Infrastructure

A backup system (SRK) is a set of hardware and software tools for automatically creating backup copies of data and their rapid recovery in case of loss, damage or encryption by ransomware. The key task is to ensure data safety under any failures and cyberattacks. Read more about related aspects in the materials: SAN (storage area network) and Data warehouse.

Source: term "Backup system (SRK)"

The need for a backup system arises from the risks of data loss due to technical failures, human errors, hardware failures and cyberattacks, primarily ransomware. Backup protects against such threats by storing copies on separate media. Read more about related aspects in the materials: Data warehouse and installation works.

Source: term "Backup system (SRK)"

Building an effective backup system requires an integrated approach: determining RPO and RTO, choosing backup methods (full, incremental, differential) and storage (storage system, tape library, cloud). We also recommend familiarizing yourself with Hypervisor, Server virtualization and Postgres Pro for a complete understanding of the topic.

Source: term "Backup system (SRK)"

When building a backup system, you should avoid storing copies in the same logical environment as the main data (this does not protect against ransomware), ignoring the 3-2-1 rule and not testing recovery procedures. In the context of this topic, it is useful to study Containerization (Docker) and SMEV (interdepartmental electronic interaction system).

Source: term "Backup system (SRK)"

Backup copies can be confused with archiving: archive stores data for long-term storage, and backup is designed for rapid recovery after failures. These are different tasks with different approaches. For in-depth study, refer to the sections SMEV, COD technology and Hypervisor.

Source: term "Backup system (SRK)"

To confirm that the backup system works, regular test restores are performed — the actual recovery of data from copies in a test environment. Monitoring backup statuses, checking reports and periodic audit of recovery procedures are also used. More detailed information is available in the articles about SAN, Thin client and installation works.

Source: term "Backup system (SRK)"

Data recovery time depends on the RTO set and the recovery method: restoring from a full copy takes longer than from differential or incremental ones. For critical systems, RTO can be minutes, which requires the use of continuous data protection (CDP). Read more about related aspects in the materials: SAN (storage area network) and Hypervisor.

Source: term "Backup system (SRK)"

A virtualization server (virtualization host) is a technology or a physical computer that, with the help of special software (a hypervisor), divides its hardware resources (processors, memory, disks) into several independent virtual machines. For in-depth study, refer to the sections Containerization (Docker), Hypervisor and SAN (storage area network).

Source: term "Server virtualization"

The main types are: full hardware-assisted virtualization (the hypervisor fully emulates hardware), paravirtualization (the guest OS is modified for higher performance) and OS-level virtualization (containers that share the host kernel). For in-depth study, refer to the sections Containerization (Docker), Hypervisor and installation works.

Source: term "Server virtualization"

The advantages of server virtualization are as follows: efficient use of equipment (utilization increases from 10-15% to 70-80%), reduction of capital and operating costs, simplified management and high fault tolerance. More detailed information is available in the articles about Data warehouse, Workstation and data center (DPC).

Source: term "Server virtualization"

Server virtualization is a technology that allows dividing the resources of one physical server into several isolated virtual machines (VMs). We also recommend familiarizing yourself with SAN (storage area network) and Thin client for a complete understanding of the topic.

Source: term "Server virtualization"

Server virtualization is the process of dividing a physical server into several unique and isolated virtual servers using a software application. We also recommend familiarizing yourself with Postgres Pro, SAN (storage area network) and Thin client for a complete understanding of the topic.

Source: term "Server virtualization"

A virtual server (VPS or VDS) is an isolated virtual environment that emulates the operation of a separate physical computer. We also recommend familiarizing yourself with Backup system (SRK), Hypervisor and installation works for a complete understanding of the topic.

Source: term "Server virtualization"

Virtualization is a technology for creating software (virtual) copies of physical computers, servers or networks. For in-depth study, refer to the sections installation works, Workstation and Containerization (Docker).

Source: term "Server virtualization"

Containerization is a method of packaging a program together with all its components (code, libraries and settings) into an isolated environment — a container. For in-depth study, refer to the sections Sintezm and Postgres Pro.

Source: term "Containerization (Docker)"

The main difference is in the abstraction level: virtualization emulates a whole physical device (computer), while containerization isolates only the application itself and its dependencies within a single operating system. More detailed information is available in the articles about Thin client and Data warehouse.

Source: term "Containerization (Docker)"

Docker is a platform for creating, delivering and running containers. Docker allows packaging applications with all dependencies into images and running them in isolation on the host. More detailed information is available in the articles about SAN (storage area network), Hypervisor and Data warehouse.

Source: term "Containerization (Docker)"

Containerization in IT is a method of packaging an application together with all the dependencies necessary for its operation (libraries, settings, files) into a single image — a container. We also recommend familiarizing yourself with SAN (storage area network) and Postgres Pro for a complete understanding of the topic.

Source: term "Containerization (Docker)"

Containerization provides isolation of applications, lightweight and fast startup, portability of environments, efficient use of server resources and simplified deployment and scaling of microservices. For in-depth study, refer to the sections Server virtualization and Hypervisor.

Source: term "Containerization (Docker)"

Docker images are immutable, that is, they cannot be changed after creation. Containerization is a broader concept — a method of isolating applications, while Docker is a specific tool for its implementation. For in-depth study, refer to the sections data center (DPC), Server virtualization and Data warehouse.

Source: term "Containerization (Docker)"

Virtualization and containerization have their own strengths. Virtualization provides stronger isolation and runs any OS, while containerization is lighter and faster, ideal for microservices. They are often used together. In the context of this topic, it is useful to study Data warehouse, SAN (storage area network) and Backup system (SRK).

Source: term "Containerization (Docker)"

DPC stands for data processing center (also often called a data center). In the context of this topic, it is useful to study Data warehouse, Thin client and Backup system (SRK).

Source: term "Data center (DPC)"

A data center (DPC) is a "digital fortress", a large specialized room or building where thousands of servers and data storage systems are located. In the context of this topic, it is useful to study Postgres Pro and the virtualization technology.

Source: term "Data center (DPC)"

According to international analytical databases and industry experts, there are about 180-195 data processing centers (data centers) in Russia. Read more about related aspects in the materials: virtualization technology, SAN (storage area network) and SMEV (interdepartmental electronic interaction system).

Source: term "Data center (DPC)"

A data center (DPC) is a specialized facility or complex of buildings designed for placing, connecting and uninterrupted operation of server and network equipment. In the context of this topic, it is useful to study Workstation and Server virtualization.

Source: term "Data center (DPC)"

DPC stands for data processing center (the English term Data Center is also often used). This is a specialized high-tech complex (or a separate building) designed for placing, connecting to the internet and uninterrupted operation of server equipment, data storage systems and communication nodes. We also recommend familiarizing yourself with the scaling technology, Hypervisor and SAN (storage area network) for a complete understanding of the topic.

Source: term "Data center (DPC)"

The Uptime Institute defines four availability tiers: Tier I (basic, ~99.671% availability), Tier II (redundant components), Tier III (concurrently maintainable, ~99.982%) and Tier IV (fault-tolerant, ~99.995%). The higher the tier, the higher the reliability and the cost of the data center. We also recommend familiarizing yourself with scaling technology and Hypervisor for a complete understanding of the topic.

Source: term "Data center (DPC)"

A data center (DPC) is needed for safe placement, uninterrupted operation and connection of server equipment to the internet. We also recommend familiarizing yourself with Server virtualization, Data warehouse and SMEV (interdepartmental electronic interaction system) for a complete understanding of the topic.

Source: term "Data center (DPC)"

A hypervisor is a conductor program that allows dividing one powerful physical computer into several independent "virtual machines". In the context of this topic, it is useful to study SAN (storage area network) and Thin client.

Source: term "Hypervisor"

A hypervisor is a control program or platform that creates and manages virtual machines, while a virtual machine is an isolated virtual environment running on a hypervisor. More detailed information is available in the articles about Data warehouse, Backup system (SRK) and SAN (storage area network).

Source: term "Hypervisor"

Hypervisors are software for creating and running virtual machines. There are two types: Type 1 (bare-metal) — installed directly on hardware (VMware ESXi, Hyper-V, KVM) and Type 2 (hosted) — running on top of an operating system (VirtualBox, VMware Workstation). Read more about related aspects in the materials: Containerization (Docker), Backup system (SRK) and Thin client.

Source: term "Hypervisor"

The hypervisor ensures the security and isolation of each virtual machine, so they work autonomously, not affecting each other. In the context of this topic, it is useful to study Containerization (Docker), SMEV (interdepartmental electronic interaction system) and Data warehouse.

Source: term "Hypervisor"

Yes, Oracle VM VirtualBox is a Type 2 hypervisor that runs on top of an operating system. We also recommend familiarizing yourself with Backup system (SRK), Containerization (Docker) and Server virtualization for a complete understanding of the topic.

Source: term "Hypervisor"

A hypervisor is software that creates and manages virtual machines on a physical server. For example, VMware ESXi is a Type 1 hypervisor installed directly on the hardware, while Oracle VirtualBox is a Type 2 hypervisor running on top of an operating system. For in-depth study, refer to the sections SMEV and the virtualization technology.

Source: term "Hypervisor"

VMware ESXi is a hypervisor without an operating system, installed on top of a physical server and running at the hardware level (Type 1). More detailed information is available in the articles about SMEV, Data warehouse and Containerization (Docker).

Source: term "Hypervisor"

A thin client is a simple and low-power computer (or program) that serves only to transmit key presses and mouse movements to a remote server, where all applications and data are executed and stored. More detailed information is available in the articles about Workstation and Containerization (Docker).

Source: term "Thin client"

A thin client is a device with limited computing capabilities that connects to a centralized server infrastructure (VDI) to work with applications and data stored on servers. We also recommend familiarizing yourself with Data warehouse, Hypervisor and Containerization (Docker) for a complete understanding of the topic.

Source: term "Thin client"

The difference between a thin and a thick client is where the main computations occur and how much of the application logic is located on the user's device. In a thin client, all computing happens on the server, in a thick client — on the local device. Read more about related aspects in the materials: Containerization (Docker) and Postgres Pro.

Source: term "Thin client"

Examples of thin clients: Thinstation, LTSP, diskless stations, terminal access, Virtual Network Computing (VNC), as well as specialized hardware devices from manufacturers. In the context of this topic, it is useful to study Backup system (SRK), Server virtualization and Hypervisor.

Source: term "Thin client"

The 1C thin client is a lightweight version of the program for accessing 1C:Enterprise 8 databases, which does not perform local computing but works with the server part. More detailed information is available in the articles about Server virtualization and SMEV.

Source: term "Thin client"

Yes, for typical office tasks a thin client works like a regular PC: you connect a monitor, keyboard and mouse and use the desktop. The difference is that all applications and data are executed and stored on the server, while the device itself has minimal local resources. We also recommend familiarizing yourself with Workstation and Hypervisor for a complete understanding of the topic.

Source: term "Thin client"

A thin client is a system unit that usually has no hard disk and contains only a minimal set of hardware needed to run the thin client operating system: a low-power processor, a small amount of flash memory, a network adapter and monitor ports. More detailed information is available in the articles about SAN (storage area network) and Server virtualization.

Source: term "Thin client"

A workstation is a system of hardware and software methods for solving specific tasks. Read more about related aspects in the materials: Backup system (SRK), Containerization (Docker) and arm_spd_k.

Source: term "Workstation"

A workstation is a separate and unique product — a high-performance professional computer for resource-intensive tasks. We also recommend familiarizing yourself with Backup system (SRK) and Data warehouse for a complete understanding of the topic.

Source: term "Workstation"

The main difference is in the purpose of use and load scenarios: a server serves multiple users and processes requests, while a workstation is designed for a single specialist working with resource-intensive applications. We also recommend familiarizing yourself with Hypervisor and Postgres Pro for a complete understanding of the topic.

Source: term "Workstation"

The Workstation service (LanmanWorkstation) provides support for network connections on computers running Windows. More detailed information is available in the articles about arm_spd and Thin client.

Source: term "Workstation"

A workstation is a high-performance computer specially designed for processing complex professional tasks such as data analysis, 3D design, video editing, engineering modeling and complex calculations. We also recommend familiarizing yourself with Hypervisor and Thin client for a complete understanding of the topic.

Source: term "Workstation"

A workstation uses professional components certified by ISV (processors, ECC memory, professional video cards) and is designed for 24/7 operation under high load in resource-intensive tasks such as CAD, 3D modeling and data analysis. For in-depth study, refer to the sections Thin client and Backup system (SRK).

Source: term "Workstation"

A workstation is a powerful professional computer created for solving resource-intensive tasks: 3D modeling, video editing, engineering calculations (CAD), architectural design and data analysis. In the context of this topic, it is useful to study Thin client and arm_spd.

Source: term "Workstation"

In a data warehouse context, cleaning means data quality processes: deduplication, normalization, removing incorrect or outdated records, and setting up retention policies. This is performed during ETL/ELT stages. In the context of this topic, it is useful to study Hypervisor, the scaling technology and Backup system (SRK).

Source: term "Data warehouse (DWH)"

Data warehouse data is located in specialized storage systems and servers in a data center. Access is provided through BI tools, SQL queries and analytical interfaces. For in-depth study, refer to the sections Workstation and SMEV.

Source: term "Data warehouse (DWH)"

A data warehouse (Data Warehouse, DWH) is a centralized database for storing and analyzing information from different sources: ERP, CRM, web services, Excel files and databases. Read more about related aspects in the materials: Containerization (Docker) and Server virtualization.

Source: term "Data warehouse (DWH)"

Building a data warehouse includes the following steps: 1) identifying data sources and requirements; 2) designing the data model (star schema, Data Vault, 3NF); 3) setting up ETL/ELT processes; 4) loading historical data; 5) connecting BI tools. More detailed information is available in the articles about Containerization (Docker), Server virtualization and Hypervisor.

Source: term "Data warehouse (DWH)"

If the warehouse capacity is exhausted, options include archiving old data, increasing storage capacity using the scaling technology, optimizing data models and setting data retention policies. We also recommend familiarizing yourself with Postgres Pro, data center (DPC) and SMEV for a complete understanding of the topic.

Source: term "Data warehouse (DWH)"

A regular database (OLTP) is optimized for fast recording and updating of operational data, while a data warehouse (OLAP) is optimized for analyzing large volumes of historical data with complex queries. Here are 5 key differences: purpose, data structure, query type, history storage and load profile. We also recommend familiarizing yourself with Hypervisor, Workstation and the scaling technology for a complete understanding of the topic.

Source: term "Data warehouse (DWH)"

To provide access to a data warehouse, it is necessary to configure accounts, access rights and roles, connect BI tools and set up security policies. We also recommend familiarizing yourself with SMEV, Hypervisor and SAN (storage area network) for a complete understanding of the topic.

Source: term "Data warehouse (DWH)"

PostgreSQL is a basic free open-source DBMS with a global community, while Postgres Pro is a Russian commercial DBMS with additional optimizations, security tools and official support. Read more about related aspects in the materials: SKIF-BP, Containerization (Docker) and Backup system (SRK).

Source: term "Postgres Pro"

Postgres Pro is a Russian commercial database management system (DBMS) created by Postgres Professional. In the context of this topic, it is useful to study SAN (storage area network) and installation works.

Source: term "Postgres Pro"

The price of Postgres Pro licenses depends on the edition, the type of license (perpetual or subscription) and the number of processor cores. Read more about related aspects in the materials: Data warehouse and SAN (storage area network).

Source: term "Postgres Pro"

Postgres Pro can be downloaded from the official website of Postgres Professional (postgrespro.ru), where the installation packages and documentation are available. Read more about related aspects in the materials: Thin client, Sintezm and Hypervisor.

Source: term "Postgres Pro"

Postgres Pro is a Russian commercial database management system (DBMS) created on the basis of the open PostgreSQL platform. For in-depth study, refer to the sections SAN (storage area network) and Hypervisor.

Source: term "Postgres Pro"

A billion rows sounds intimidating, but PostgreSQL copes well with this with the right approach: partitioning, proper indexing and query optimization. We also recommend familiarizing yourself with Thin client and Workstation for a complete understanding of the topic.

Source: term "Postgres Pro"

PostgreSQL has existed for more than 30 years not because it is outdated, but because it is constantly developing. Postgres Pro continues to develop it with Russian-specific features. More detailed information is available in the articles about Server virtualization and SKIF-BP.

Source: term "Postgres Pro"

SMEV (Interdepartmental Electronic Interaction System) is a digital "bridge" or a secure messenger that allows various government agencies to instantly exchange data with each other. More detailed information is available in the articles about Data warehouse, Hypervisor and Backup system (SRK).

Source: term "SMEV (interdepartmental electronic interaction system)"

Access to the Interdepartmental Electronic Interaction System (SMEV) is available to government bodies, local self-government bodies and commercial organizations performing socially significant functions. Read more about related aspects in the materials: Backup system (SRK), data center (DPC) and SAN (storage area network).

Source: term "SMEV (interdepartmental electronic interaction system)"

SMEV is an interdepartmental electronic interaction system, whose task is to ensure the execution of state functions and the provision of state services in electronic form, as well as to ensure information interaction in the provision of these services and the execution of these functions. Read more about related aspects in the materials: Thin client, Containerization (Docker) and Data warehouse.

Source: term "SMEV (interdepartmental electronic interaction system)"

The procedure for connecting to SMEV is as follows: 1) fill out an application for registration in SMEV and submit it to the Ministry of Digital Development; 2) develop an electronic service in accordance with the requirements; 3) pass testing in the development environment; 4) get access to the production environment. For in-depth study, refer to the sections Containerization (Docker), Postgres Pro and Server virtualization.

Source: term "SMEV (interdepartmental electronic interaction system)"

Government bodies, as well as commercial and public organizations performing socially significant functions, can connect to SMEV (Interdepartmental Electronic Interaction System). For in-depth study, refer to the sections Workstation and Server virtualization.

Source: term "SMEV (interdepartmental electronic interaction system)"

Verification takes up to 5 calendar days — a notification will come to the Personal Account. In the context of this topic, it is useful to study Thin client and Server virtualization.

Source: term "SMEV (interdepartmental electronic interaction system)"

The government has approved the rules for charging fees for the use of e-government infrastructure (EGI); the fee is 4.93 rubles per registered request in the unified interdepartmental electronic interaction system (SMEV) and 10 rubles for obtaining information from the unified identification and authentication system (ESIA). In the context of this topic, it is useful to study Hypervisor and Thin client.

Source: term "SMEV (interdepartmental electronic interaction system)"

DNS is the "phone book of the internet". It converts human-readable names of sites (for example, yandex.ru) into numeric IP addresses that computers understand.

Source: term "DNS (Domain Name System)"

Use the nslookup domain.ru or dig domain.ru command in the terminal. Online services: dnschecker.org, mxtoolbox.com.

Source: term "DNS (Domain Name System)"

DoH (DNS over HTTPS) is a protocol that encrypts DNS requests via HTTPS, protecting them from interception and substitution. It is enabled in the browser or at the OS level.

Source: term "DNS (Domain Name System)"

Wi-Fi is wireless internet. The router broadcasts a signal, and your phone or laptop receives it without cables. Convenient, fast and works within the home.

Source: term "Wi-Fi (Wireless Fidelity)"

Wi-Fi 7 (802.11be) is the newest and fastest generation (up to 46 Gbit/s). Wi-Fi 6 is the current standard for most devices. Wi-Fi 5 is still relevant for home networks.

Source: term "Wi-Fi (Wireless Fidelity)"

Wi-Fi is safe when using WPA2/WPA3 encryption and a strong password. Open networks (without a password) are unsafe — attackers can intercept data. Use a VPN in public networks.

Source: term "Wi-Fi (Wireless Fidelity)"

A web server is a program that serves website pages to your browser. When you visit a site, your browser sends a request to the web server, and it returns the page.

Source: term "Web server"

NGINX is faster and more efficient under high loads, better suited for proxying. Apache is easier to configure for small projects. Most modern sites use NGINX.

Source: term "Web server"

Yes, you can install NGINX or Apache on a home computer. But for the site to work, a static IP or DDNS, open ports and stable internet are needed. For serious projects, it is better to use hosting.

Source: term "Web server"

A single server is a single point of failure: if it fails, the service stops. A server cluster combines several nodes, providing fault tolerance. The cluster also allows scaling — adding new servers without stopping the service.

Source: term "Server cluster"

The main schemes: Active-passive — one node works, the second is on standby. Active-active — all nodes work simultaneously, distributing the load. The active-active scheme provides better performance.

Source: term "Server cluster"

A server cluster usually uses a common data storage system (DSS) to which all nodes have access. This ensures data consistency. Building a fault-tolerant cluster without a common DSS is impossible.

Source: term "Server cluster"

Kubernetes, Proxmox VE, VMware vSphere, OpenStack are used for cluster management. Load balancers (Nginx, HAProxy) are also used. Virtualization technology underlies many solutions.

Source: term "Server cluster"

A minimum cluster for fault tolerance usually consists of 3 nodes (to ensure quorum). For high-load systems, the number of nodes can reach hundreds or even thousands.

Source: term "Server cluster"

A high availability cluster is designed to ensure uninterrupted operation of critical applications. When one node fails, its resources are automatically switched to another. This type is often used in banking and at CII facilities.

Source: term "Server cluster"

The main risks: the high cost of equipment and software, the complexity of configuration and administration, the need for highly qualified personnel. Professional design helps minimize these risks.

Source: term "Server cluster"

For business, the choice depends on security, integration and volume requirements. Yandex 360, VK Cloud, SberDisk are popular in Russia. For personal data, FSTEC-certified solutions integrated with CIPF are recommended.

Source: term "Cloud storage"

Storing documents in the cloud is safe when using encryption (TLS for transmission, AES-256 for storage) and two-factor authentication. It is also important to choose a provider with data centers in Russia.

Source: term "Cloud storage"

Free tariffs: Yandex Disk — 5 GB, Mail.ru Cloud — 8 GB, Google Drive — 15 GB, Microsoft OneDrive — 5 GB, Dropbox — 2 GB.

Source: term "Cloud storage"

Cloud storages integrate with document flow systems via API, providing automatic document loading and synchronization. For example, SINTEZ-M supports integration with cloud storages.

Source: term "Cloud storage"

Yes, most cloud storages support automatic backup. Specialized backup solutions are available, for example Veeam. The 3-2-1 principle is recommended for reliability.

Source: term "Cloud storage"

For government bodies and CII facilities, cloud storages must be included in the Register of Russian Software, certified by FSTEC, support CIPF and store data in Russia.

Source: term "Cloud storage"

Most cloud storages provide file versioning and a recycle bin that allows restoring previous versions. Immutable storage is recommended to protect against encryption.

Source: term "Cloud storage"

1U (1 unit) is a standard height measurement equal to 44.45 mm. Servers can be 1U, 2U, 4U high. A 42U rack can hold 42 1U servers.

Source: term "Server rack"

Width of 19 inches, height 42U (about 2 m), depth from 600 to 1200 mm. In data centers racks of 1000-1200 mm are most often used.

Source: term "Server rack"

A rack is an open metal frame. A cabinet is a closed structure with doors and locks for additional protection.

Source: term "Server rack"

For example, a 42U rack can hold 42 1U servers or 21 2U servers. On average 15-20 servers fit considering cooling gaps.

Source: term "Server rack"

Cold corridor / hot corridor principle. Temperature 18-27°C and humidity 40-60% are maintained in data centers.

Source: term "Server rack"

Physical protection: locks, access control systems, video surveillance and door sensors. Grounding and fire safety compliance are also required.

Source: term "Server rack"

Consider equipment depth, load, cooling requirements, cable management, DCIM compatibility and physical security.

Source: term "Server rack"

Archiving frees up space for long-term storage. Backup creates copies for recovery after a failure. The processes are often combined in backup systems (RBS).

Source: term "Data archiving"

A combined strategy is optimal: full backup weekly, incremental daily. For rapid recovery, the differential approach is preferable.

Source: term "Data archiving"

ZIP or 7Z — open formats guaranteeing future readability. Use lossless compression and check integrity with checksums.

Source: term "Data archiving"

Use CIPF encryption, the 3-2-1 rule and WORM protection in backup systems.

Source: term "Data archiving"

Yes, high compression (LZMA2 in 7Z) requires more time to unpack. For critical data, ZIP is often chosen to reduce RTO.

Source: term "Data archiving"

Data transfer speed and consistency. Deduplication, client-side compression and specialized protocols are used.

Source: term "Data archiving"

From several hours to a day depending on file type, disk performance and compression. Automation via backup systems runs processes at night.

Source: term "Data archiving"

Cloud security is the protection of your data stored in the cloud (for example, in Yandex Disk, Google Drive or corporate cloud systems). It is like security for your digital files: encryption, access control, protection from hacking. Read more about encryption in the article Encryption.

Source: term "Cloud security"

The main threats: data leaks (due to configuration errors), incorrect configuration of cloud resources (open buckets, public databases), account compromise (weak passwords, lack of two-factor authentication) and insider threats (errors or malicious actions of employees). About how to protect accounts, read the article Verification.

Source: term "Cloud security"

Responsibility is divided between the provider and the client. The provider protects the infrastructure (data centers, equipment, networks). The client is responsible for access configuration, operating systems, data encryption and the security of end devices. This is called the shared responsibility model. Read more about the division of responsibility in the article Infrastructure.

Source: term "Cloud security"

The main standards: ISO 27001 (international information security management standard), PCI DSS (for processing payment data), FSTEC and FSB (Russian certificates for government organizations). Also in force is 242-FZ on data localization — the data of Russian citizens must be stored on servers in the Russian Federation. About what PCI DSS is, read the article PCI DSS.

Source: term "Cloud security"

When choosing, check: the presence of security certificates (ISO 27001, PCI DSS, FSTEC), support for data encryption, the presence of data centers in Russia (to comply with 242-FZ), round-the-clock technical support and SLA. Also study user reviews and the provider's security reports. About choosing cloud solutions, read the article Fintech.

Source: term "Cloud security"

This is a principle according to which cloud security is divided between the provider and the client. The provider is responsible for the security of the infrastructure (physical servers, networks, hypervisors). The client is responsible for the security of what he places in the cloud (operating systems, applications, data, access settings). It is important to understand this in order not to allow gaps in protection. About data protection, read the article Information security.

Source: term "Cloud security"

The consequences can be serious: leakage of confidential data, financial losses, regulatory fines (up to 6 million rubles under 152-FZ), loss of client trust, stoppage of business processes. In some cases — criminal liability for managers. About how to protect data and avoid fines, read the article Information security.

Source: term "Cloud security"

A virtual machine is a “computer inside a computer”. It is a program that creates a virtual computer with its own processor, memory and disk but uses the resources of your real PC. You can install another operating system on it and work in it as on a separate device. Read about how this is used in the cloud in the article Cloud security.

Source: term "Virtual machine"

VirtualBox is a free open-source program, ideal for home use and beginners. VMware is a powerful commercial solution that provides higher performance and works better with 3D graphics. VMware Workstation Pro is now free for personal use. Choose VirtualBox for simple tasks, VMware for professional work. Read about choosing tools in the article Infrastructure.

Source: term "Virtual machine"

Yes, but this requires certain conditions. A virus can infect the main system through shared folders, the clipboard or vulnerabilities in the hypervisor itself (VM Escape). To protect yourself: disable shared folders, do not use the clipboard, set up an isolated network. Read about protection from viruses in the article Antivirus software.

Source: term "Virtual machine"

Virtual machines are divided by hypervisor type: type 1 (Bare-Metal) — installed directly on the hardware (VMware ESXi, Hyper-V), type 2 (hosted) — run inside an operating system (VirtualBox, VMware Workstation). They also differ by purpose: server VMs, desktop VMs, containers and emulators.

Source: term "Virtual machine"

For Windows, the best choice is VMware Workstation Pro (high performance) or VirtualBox (free and simple). If you have Windows Pro, you can use the built-in Hyper-V. For home use I recommend VirtualBox, for professional work — VMware. Read about choosing business solutions in the article Low-code.

Source: term "Virtual machine"

Download and install VirtualBox or VMware. Click “Create”, choose a name and OS type, allocate RAM (2-4 GB), create a virtual disk (20-50 GB). Then specify the ISO image of the operating system and run the installation as on an ordinary computer. Read more about configuration in the article Infrastructure.

Source: term "Virtual machine"

At a minimum — 2 GB for lightweight Linux systems, 4 GB for Windows 10/11. For comfortable work, 8-16 GB is recommended on the host system to allocate 4-8 GB to the virtual machine. The more tasks you plan to perform in the VM, the more RAM you will need. Read about system requirements in the article Operating system.

Source: term "Virtual machine"

An operating system is the main program on a computer or smartphone that manages all the “hardware” and allows you to use the device. Without an OS, a computer simply will not turn on — it is like a brain for the technology. The most popular operating systems: Windows, macOS, Linux for computers, Android and iOS for phones. Read more about choosing an OS in the article Infrastructure.

Source: term "Operating system"

For computers and laptops — Windows (the most popular), macOS (for Apple computers) and Linux (free, for programmers). For smartphones and tablets — Android (the most popular) and iOS (for iPhone). There are also Russian operating systems: Astra Linux, RED OS, Alt OS. Read about Russian developments in the article Import substitution.

Source: term "Operating system"

Among computers — Windows (about 72% of the market). Among all devices (including smartphones) — Android (more than 70% of the global market). Windows is popular for work and games, Android — for mobile devices. Read about choosing an OS for business in the article Fintech.

Source: term "Operating system"

Linux is free, more secure (fewer viruses), lighter (works on old hardware) and gives full control over the system. Windows is more convenient for beginners and has more programs and games. The choice depends on your tasks: if you are a programmer or administrator — Linux, for an ordinary user — Windows. Read about choosing an OS in the article Infrastructure.

Source: term "Operating system"

This is an OS developed in Russia to replace the foreign Windows and macOS. The main Russian operating systems are Astra Linux (used in government bodies), RED OS and Alt OS. They are based on Linux, have FSTEC and FSB certificates and meet the security requirements for work with state secrets. Read about Russian developments in the article Import substitution.

Source: term "Operating system"

For most users, the best choice is Windows 11, since it has the most programs, games and hardware compatibility. If you work in a creative field (design, video, music), macOS is an excellent option. For programmers and system administrators — Linux. Read about choosing an OS for work in the article Fintech.

Source: term "Operating system"

Linux is harder for beginners to learn, has fewer games and professional programs (for example, Adobe Photoshop). There may also be problems with drivers for non-standard hardware. However, for programmers, servers and advanced users these downsides are not critical. Read about choosing an OS in the article Infrastructure.

Source: term "Operating system"

A domain is the address of a website on the internet that you enter in the browser. For example, google.com or yandex.ru are domains. They are needed so that people can easily find websites without remembering complex digital IP addresses. Read more about how the internet works in the article Infrastructure.

Source: term "Domain"

A domain is only the address of the site (for example, yoursite.ru). A website is the project itself with texts, images and code, which is stored on hosting. A domain can be compared to the address of a house, and a website — to the house itself where you live. Read about choosing hosting in the article Infrastructure.

Source: term "Domain"

Domain zones are divided into national (.ru — Russia, .kz — Kazakhstan, .de — Germany), international (.com — commercial, .org — non-profit, .net — network) and new thematic zones (.shop — stores, .tech — technologies, .blog — blogs). The choice of zone depends on the goals of your project. Read about choosing a domain for business in the article Fintech.

Source: term "Domain"

Choose a registrar (for example, Reg.ru, Beget, Nic.ru), check the availability of the name, choose a zone, fill in the form and pay for registration. The domain is rented for a period from 1 to 10 years. After registration, configure DNS to link the domain to hosting. Read more about configuration in the article DNS.

Source: term "Domain"

The cost of a domain depends on the zone and the registrar. On average, a domain in the .ru zone costs 500-800 rubles per year, .com — 800-1200 rubles, .rf — 300-500 rubles. Premium domains (short and catchy) can cost from tens of thousands to millions of rubles. Read about choosing a domain for business in the article Low-code.

Source: term "Domain"

Technically, yes, but it will be inconvenient. Without a domain, a website is accessible only by IP address (for example, 192.168.0.1), which is hard to remember and impossible to use for branding. For a serious project, a domain is mandatory. Read about creating websites in the article Low-code.

Source: term "Domain"

A subdomain is a part of the main domain that is created for individual sections of a website. For example, in the address mail.yandex.ru, mail is a subdomain. Subdomains are used to separate services: mail, blog, store, forum. They are created free of charge in the hosting control panel. To study related concepts, we also recommend reading about CDN and DHCP.

Source: term "Domain"

A CDN is a network of servers around the world that store copies of your website (images, video, styles). When a user opens the site, files are loaded from the server closest to him, not from your main hosting. This makes the site faster, especially for users from other countries or regions. Read more about choosing IT solutions in the article Infrastructure.

Source: term "CDN"

Hosting is the place where your website is stored (the main server). A CDN is an additional network of servers that speeds up the delivery of content to users. Hosting stores all files, while a CDN stores only their copies for fast distribution. They work together: hosting is like the main house, while a CDN is like warehouses in different cities with copies of goods. Read about choosing hosting in the article Low-code.

Source: term "CDN"

In Russia, local providers operate: Ngenix (the largest, part of Rostelecom Data Center), Selectel, CDNvideo, EdgeTsentr, cdnnow!, as well as cloud Yandex Cloud CDN and VK Cloud CDN. Among foreign ones, Cloudflare has retained several nodes in the Russian Federation, but its use may be unstable due to restrictions. For Russian projects, it is better to choose local CDNs. Read more about choosing providers in the article Fintech.

Source: term "CDN"

Prices depend on the volume of traffic. Some providers have fixed tariffs (for example, 600 rubles per month for 1000 GB at cdnnow!). Others offer pay-as-you-go — from 0.5 to 1.5 rubles per 1 GB of traffic. Some cloud providers (Yandex Cloud) give free limits of up to 150 GB. For small websites, CDN costs amount to 500-3000 rubles per month. Read about choosing tariffs in the article Infrastructure.

Source: term "CDN"

Some CDNs have free tariffs: Cloudflare (the most popular free CDN in the world), as well as Russian services like TurboFlare. However, free tariffs usually have restrictions on traffic, speed or functionality. For serious projects with large traffic, it is better to use paid solutions. Read about choosing a CDN for business in the article Low-code.

Source: term "CDN"

Even if your website is fast in your region, it can be slow for users from other cities or countries. A CDN solves this problem by storing copies of files around the world. A CDN also protects from DDoS attacks, reduces the load on the server and improves SEO. If you have a global audience or plan to expand it, a CDN is necessary. Read about protection from DDoS in the article WAF.

Source: term "CDN"

Connecting a CDN usually takes 10-15 minutes: choose a provider, register, create a CDN resource and specify the address of your main server. Then in the DNS settings of your domain, create a CNAME record pointing to the address of the CDN provider. Configure caching rules and an SSL certificate. Most CMS (WordPress, 1C-Bitrix) have plugins for automatic integration. Read more about configuration in the article Infrastructure. We also recommend reading about the concept of a Domain.

Source: term "CDN"

DHCP is a system that automatically distributes IP addresses to devices in a network. When you connect a phone to Wi-Fi, the DHCP server (usually a router) assigns it an address, subnet mask and other settings by itself. You do not need to enter anything manually — everything happens automatically. Read more about how networks work in the article Infrastructure.

Source: term "DHCP"

DHCP distributes IP addresses to devices in a network, while DNS converts domain names of websites (for example, yandex.ru) into IP addresses. DHCP is like an administrator who issues numbers (addresses) to devices. DNS is like a phone book that uses a name (domain) to find a number (IP address). They work together: DHCP issues an address, and DNS helps find a website by its name. Read more about DNS in the article DNS.

Source: term "DHCP"

If you disable DHCP, devices will stop automatically receiving IP addresses. Every device (phone, laptop, TV) will have to be configured manually — IP address, subnet mask, gateway and DNS. This is very inconvenient, especially if you have many devices. Without DHCP, the network stops being “plug-and-play” — a new device will not be able to connect without manual configuration. Read about network configuration in the article Infrastructure.

Source: term "DHCP"

The lease time is the period for which an IP address is allocated to a device. When the time expires, the device can receive a new address or renew the lease. For home, 24 hours is usually set, for public networks (cafes, hotels) — 1-2 hours. A short lease time helps free up addresses faster when a device disconnects from the network.

Source: term "DHCP"

DHCP is convenient for most devices — it is automatic and requires no configuration. A static IP is needed for servers, printers and other devices that require a permanent address (for example, for external access). In home networks, almost all devices use DHCP, and for a printer or NAS you can reserve a permanent IP through the DHCP settings. Read about choosing settings in the article Infrastructure.

Source: term "DHCP"

Go to the router settings through the browser (usually 192.168.0.1 or 192.168.1.1). Find the LAN, Network or DHCP section. Set the switch to the “Enable” position, specify the address range (for example, 192.168.1.100-200) and the lease time. Save the settings and restart the router. After that, devices will automatically receive IP addresses. Read more about configuration in the article Infrastructure.

Source: term "DHCP"

Manual configuration of IP addresses for every device takes a lot of time and is fraught with errors — you can accidentally assign two devices the same address (an IP conflict). DHCP automates this process, prevents conflicts and simplifies network management, especially if you have many devices. In modern networks, DHCP is used almost everywhere.

Source: term "DHCP"

Wireless networks are the connection of devices without wires, over radio waves. This is Wi-Fi at home, Bluetooth headphones, mobile internet 4G/5G. They allow you to connect to the internet and exchange data anywhere without being tied to cables. Read more about technologies in the article Bluetooth.

Source: term "Wireless networks"

Wireless networks are divided by range: WPAN (up to 10 m — Bluetooth, NFC), WLAN (up to 100 m — Wi-Fi), WMAN (up to tens of km — WiMAX) and WWAN (global — 4G, 5G, satellite internet). For home and office, Wi-Fi is used, for mobile devices — cellular communication. Read about home networks in the article Home networks.

Source: term "Wireless networks"

Wi-Fi is a way to connect devices to the internet inside a room (a local network). Wireless internet is a way to deliver the internet to your home or device (through 4G/5G or a satellite). Wi-Fi distributes the internet inside the home, while wireless internet delivers it to the home. They work together: the provider gives wireless internet, and the router distributes it over Wi-Fi.

Source: term "Wireless networks"

Almost all modern devices: smartphones, laptops, tablets, smart watches, fitness bracelets, wireless headphones, smart speakers, TVs, game consoles, robot vacuum cleaners, smart light bulbs and other internet of things devices. Even cars use wireless networks for navigation and updates.

Source: term "Wireless networks"

The main risks: data interception (if the network is not password-protected), attacks through router vulnerabilities, connection of third-party devices to your network, theft of passwords through fake Wi-Fi points (Evil Twin). To protect yourself, use WPA2/WPA3 encryption, strong passwords and regularly update the router firmware. Read about data protection in the article Cloud security.

Source: term "Wireless networks"

A wired network (Ethernet) provides higher speed, stability and security. A wireless one (Wi-Fi) is more convenient — you can connect from any point of the house without being tied to a cable. For desktop computers and game consoles, it is better to use a cable, for mobile devices — Wi-Fi. Often both options are used together.

Source: term "Wireless networks"

On a phone or laptop, open the Wi-Fi settings, turn it on, select the network from the list and enter the password. If the network is open (without a password), the connection will happen automatically, but this is unsafe. Connecting to a corporate network may require additional configuration (login, certificate). Read about setting up home networks in the article Home networks. To study related concepts, we also recommend reading about the Internet of things and NFC.

Source: term "Wireless networks"

A home network is the union of all your devices (phone, laptop, TV, printer) into a single network through a router. They can talk to each other and access the internet through one channel. It is like a common house where all residents can talk and go outside through a common door (the router). Read about how wireless networks work in the article Wireless networks.

Source: term "Home networks"

The main components: a router (distributes the internet), devices (laptops, phones, TVs), cables for wired connection and access points for expanding Wi-Fi. The router is the “brain” of the network, which manages all traffic and distributes the internet between devices. Read about choosing a router in the article Infrastructure.

Source: term "Home networks"

Connect the provider's cable to the router, turn it on, go to the settings (192.168.0.1 or 192.168.1.1), set a network name and password, choose WPA2 or WPA3 encryption. Then connect devices — find the network in the Wi-Fi list and enter the password. The whole process takes 10-15 minutes. Read more about configuration in the article Infrastructure.

Source: term "Home networks"

A home network is a general concept that includes all connection methods (wired and wireless). Wi-Fi is only a wireless way to connect to a home network. A home network can be wired (cables), wireless (Wi-Fi) or hybrid. Wi-Fi is a part of a home network.

Source: term "Home networks"

Yes, a home network can work without internet access. You can exchange files between computers, print on a shared printer, watch movies from a media server or play local network games. The internet is needed only to reach the global network, but the internal functions of a home network work even without it.

Source: term "Home networks"

The choice of provider depends on your address. In Moscow and large cities, MTS, Beeline, Rostelecom and Dom.ru are popular. In the regions — local providers. Compare tariffs by speed and price, read neighbors' reviews. The optimal speed for a family of 3-4 people is 100-300 Mbps. Read about choosing a provider in the article Fintech.

Source: term "Home networks"

Use a strong password (at least 12 characters, with digits and letters), enable WPA2 or WPA3 encryption, hide the network name (SSID) from outsiders, disable WPS, regularly update the router firmware. You can also set up a guest network for visitors so that they do not have access to your devices. Read about data protection in the article Cloud security.

Source: term "Home networks"

Bluetooth is a wireless communication technology over short distances. It allows you to connect headphones to a phone, a keyboard to a computer, a watch to a smartphone — all without wires. It is like an invisible cable that connects your devices. Read about other wireless technologies in the article Wireless networks.

Source: term "Bluetooth"

Bluetooth connects devices to each other (headphones to a phone) and consumes little power. Wi-Fi connects devices to the internet and consumes more power. Bluetooth is for personal accessories, Wi-Fi is for the internet and networks. They complement each other: Bluetooth for short-range communication, Wi-Fi for internet access. Read more about Wi-Fi in the article Wireless networks.

Source: term "Bluetooth"

The main versions: Bluetooth 4.0 (introduced the energy-saving BLE mode), Bluetooth 5.0 (increased the range to 240 m and doubled the speed), Bluetooth 5.2 (LE Audio), Bluetooth 5.3 (improved stability) and the newest Bluetooth 6.0 (precise positioning down to centimeters). For most devices, Bluetooth 5.0 and higher is enough.

Source: term "Bluetooth"

Yes, Bluetooth works without the internet. It connects devices directly over a radio channel. You can listen to music through headphones, transfer files between phones, control a smart home — all of this works without Wi-Fi and mobile internet. The internet is not needed for Bluetooth to work.

Source: term "Bluetooth"

Yes, Bluetooth is absolutely safe for health. The radiation power of Bluetooth devices is 1-10 milliwatts — this is 1000 times less than that of a smartphone. The signal belongs to non-ionizing radiation, which cannot damage DNA or heat tissues. Large studies do not confirm a link between Bluetooth and diseases.

Source: term "Bluetooth"

It is recommended to turn off Bluetooth when it is not needed. This saves battery (albeit a little) and improves security — fraudsters will not be able to connect to your device through Bluetooth vulnerabilities. In public places, keep Bluetooth off; at home you can leave it on for accessories. Read about data protection in the article Digital footprint.

Source: term "Bluetooth"

Turn on the headphones and switch them to pairing mode (usually hold the power button for 3-5 seconds). On the phone, turn on Bluetooth, find the headphones in the list of available devices and press “Connect”. If a password is asked, it is usually 0000 or 1234. After the first connection, the headphones will connect automatically. Read about device setup in the article Infrastructure.

Source: term "Bluetooth"

NFC is a technology that allows devices to exchange data over a very short distance (up to 10 cm). The most famous example is paying in a store with a phone: you simply bring your smartphone to a terminal. NFC is also used for metro travel, transferring files between phones and managing a smart home. Read about how NFC works in payments in the article Fintech.

Source: term "NFC"

On Android: go to “Settings” → “Connections” (or “Network & Internet”) → find the NFC item and move the slider to “On”. You can also enable it through the notification shade (swipe down) — the NFC icon should be there. On the iPhone, NFC is enabled by default and does not require manual activation — it is automatically activated when using payment applications (Apple Pay). Read more about configuration in the article Infrastructure.

Source: term "NFC"

A quick way: type the command *#06# on the phone keypad — if there is NFC, you will see a line with the EID (Embedded Identity Document). Or go to “Settings” and enter “NFC” in the search — if the item exists, the module is supported. You can also check in “Settings” → “Connections” — look for the NFC or Contactless Payments item. On iPhone, NFC is present in all models from iPhone 6 and above, but it is used only for Apple Pay.

Source: term "NFC"

To set up payment via NFC: enable NFC on your phone (on Android), install a payment application (Google Wallet, Mir Pay, SberPay, T-Pay), add a bank card (scan it with the camera or enter manually). Confirm the card via SMS or the bank application. Then simply unlock the phone (place a finger or enter a password) and bring it to a payment terminal. You do not need to open the application for payment — it is enough that the screen is active. Read about payment systems in the article Fintech.

Source: term "NFC"

Yes, NFC is safe. The range is only 2-10 cm, so an attacker cannot “read” the data at a distance. In addition, payment requires unlocking the screen (password, fingerprint or Face ID). The payment data itself is not transmitted — instead of the card number, an encrypted digital token is used (tokenization), which is useless for attackers. Banks also use additional transaction verification. Read about data protection in the article Digital footprint.

Source: term "NFC"

You do not have to turn off NFC — it almost does not consume battery (less than 1% per day) and is safe. However, for maximum security, experts recommend disabling it in public places (transport, streets) to rule out the theoretical possibility of an attack (although such attacks are extremely rare due to the short range). At home, you can leave it on for comfort. Read about mobile device security in the article Digital footprint.

Source: term "NFC"

Yes, NFC works without the internet. Data transfer between devices happens directly over a radio channel at 13.56 MHz. Contactless payment does not require the internet — the terminal contacts the bank through its own network (for example, through GSM or Ethernet), not through your phone. However, some functions (for example, checking a card balance or activating some payment applications) may require the internet, but the main NFC function works offline.

Source: term "NFC"

eSIM is a virtual SIM card. It is built into the phone and does not require a plastic card. You connect a number through the internet — scan a QR code or install it through the operator's application. It is like an electronic version of a regular SIM that cannot be lost or broken. Read about how mobile networks work in the article Wireless networks.

Source: term "eSIM"

A regular SIM is a plastic card that can be taken out and moved to another phone. eSIM is a chip soldered into the phone. It cannot be taken out. eSIM is more convenient: you can connect numbers online, store several profiles, travel without changing cards. The downside is that it is harder to transfer a number to another phone. Read about network configuration in the article Infrastructure.

Source: term "eSIM"

On Android: “Settings” → “Connections” → “SIM card manager” → “Add eSIM” → scan the QR code. On iPhone: “Settings” → “Cellular” → “Add eSIM” → scan the QR code. The QR code is issued by the operator when you order eSIM. An internet connection is needed for activation. Read more about configuration in the article Infrastructure.

Source: term "eSIM"

The main downside is the binding to a specific device. You cannot simply move the card to another phone. If the phone breaks or is lost, you need to contact the operator for a new QR code. Also, activating eSIM requires the internet. In some countries, eSIM support is still limited. Read about choosing devices in the article IMEI.

Source: term "eSIM"

eSIM is supported by all iPhones from XS/XR and newer, all Google Pixel from 3 and newer, all Samsung Galaxy S20/S21/S22/S23/S24 and newer, many Xiaomi models (13, 14 and newer), Honor, Huawei, Oppo. Enter the command *#06# — if the list contains EID, the phone supports eSIM. Read about checking devices in the article IMEI.

Source: term "eSIM"

Yes, eSIM works legally in Russia. All major operators (MTS, MegaFon, Beeline, T2, Yota, SberMobile) support eSIM. You can connect a new number or transfer an existing one from a physical SIM to eSIM. Activation requires a passport or identity confirmation through State Services. Read about network configuration in the article Infrastructure.

Source: term "eSIM"

When switching to eSIM, the physical SIM card stops working — the number is transferred to the built-in chip. You can throw away the physical card or keep it in case you need to switch back. Some operators allow you to have a physical SIM and eSIM with different numbers on one phone at the same time. Read about configuring home networks in the article Home networks.

Source: term "eSIM"

IMEI is a unique number assigned to every phone at the factory. It is like a passport or fingerprint of your device. Even if two phones are of the same model, their IMEIs will differ. Phones with two SIM cards have two IMEIs. Read about how mobile networks work in the article Wireless networks.

Source: term "IMEI"

The fastest way is to type the command *#06# on the phone keypad. The number appears on the screen instantly. You can also see the IMEI in the settings (iPhone: “Settings” → “General” → “About”; Android: “Settings” → “About phone”), on the phone box or on the case (often engraved on the SIM tray). Read about setting up devices in the article Infrastructure.

Source: term "IMEI"

IMEI is needed for identifying a device in cellular networks. It helps operators recognize a phone, and the police block stolen devices. IMEI is also used to check the authenticity of a device at purchase and for warranty service in service centers. Read about data protection in the article Digital footprint.

Source: term "IMEI"

Technically, yes, but only law enforcement bodies can do this through mobile operators. Ordinary users, websites or applications cannot do it. If a phone is stolen, you should report the IMEI to the police — they can track the device when it appears in the network. Do not trust websites that promise to find a phone by IMEI for money — this is fraud. Read about security in the article Digital footprint.

Source: term "IMEI"

Yes, a mobile operator can block a phone by IMEI if you report the theft. After blocking, the device will not be able to connect to the operator's network, even if an attacker changes the SIM card. To block, contact the police and your operator with an application. Read about device protection in the article Information security.

Source: term "IMEI"

Fraudsters can use your IMEI to clone a device, add it to blacklists or blackmail you. If an attacker gets your IMEI, he can reflash it onto another phone, and if a crime is committed with it, the police may come after you. Therefore, do not publish the IMEI on forums, in ads and do not tell it to strangers. Read about protection from fraudsters in the article Fraud.

Source: term "IMEI"

Yes, phones with two SIM cards have two IMEIs — one for each slot. Usually they are displayed simultaneously when entering the command *#06#. The first IMEI refers to the first slot, the second — to the second one. Both numbers are unique for this particular device. Read about configuring two SIM cards in the article Home networks.

Source: term "IMEI"

Backup is creating copies of your important files in case they are lost. It is like insurance for your data. If you accidentally delete a file, your computer breaks down or a virus encrypts your data, you can restore everything from a backup copy. Read about data protection in the article Cloud security.

Source: term "Data backup"

Three main types: full (all data is copied), incremental (only changes since the last copy) and differential (changes since the last full copy). Full is the most reliable, but slow. Incremental is the fastest and most economical. Differential is a compromise between them. Read about setting up backups in the article Infrastructure.

Source: term "Data backup"

This is the golden rule of backup: 3 copies of data (original + 2 backups), 2 different types of media (for example, an external disk and the cloud), 1 copy offsite (outside the home or office). This guarantees that you will not lose data even in case of serious problems. Read about applying the rule in the article Infrastructure.

Source: term "Data backup"

The best option is to combine different places: cloud storage (Google Drive, Yandex Disk, iCloud) for automatic synchronization, external hard drives for local storage and remote storage (a server in another city or a second cloud). The 3-2-1 rule recommends storing copies in different places on different media.

Source: term "Data backup"

The frequency depends on how often your data changes. For everyday documents and work files — daily. For personal photos and videos — once a week. For system backups — once a month. Set up automatic backup so you do not forget to do it. Read about setting up automation in the article Low-code.

Source: term "Data backup"

If you do not do backup, you risk losing all your data in case of a system failure, disk failure, virus attack or device theft. According to statistics, more than 60% of companies that lost their data close within six months. For individuals, the consequences are the loss of family photos, documents and work files. Read about protection from viruses in the article Antivirus software.

Source: term "Data backup"

Restoring depends on where the copy is stored. In cloud services, you need to log into your account and download the necessary files. On an external disk — connect it to the computer and copy the files back. For system backups, use built-in tools (Time Machine on macOS, System Restore in Windows). Read more about restoring in the article Infrastructure.

Source: term "Data backup"

A DDoS attack is an artificial overload of a website or server with a huge number of requests from different computers. The goal is to make the resource stop opening for ordinary users. Attackers use a network of infected devices (a botnet) around the world. Read about ways to protect against such attacks in the article Information security.

Source: term "DDoS attack"

Main signs: a sharp jump in traffic without visible reasons, server errors 502/503/504, slow loading or complete unavailability of the website, strange geography of requests (from countries where you have no audience). If you notice these signs, check the server logs and contact your hosting provider. Read about problem diagnostics in the article Information security incident.

Source: term "DDoS attack"

DoS (Denial of Service) is an attack from one device. DDoS (Distributed Denial of Service) is an attack from many sources (distributed), using a botnet. A DDoS attack is harder to block, since the traffic comes from legitimate addresses of real users whose devices were infected. Read about network threats in the article Firewall.

Source: term "DDoS attack"

Organizing or participating in a DDoS attack entails criminal liability. In Russia — under Articles 272-274 of the Criminal Code of the Russian Federation (up to 7 years of imprisonment and fines up to 2 million rubles). In the USA and EU countries — from 10 to 20 years of imprisonment. Civil lawsuits for damages are also possible. Read about legal consequences in the article Cyber fraud.

Source: term "DDoS attack"

The most effective way is to connect cloud protection services (Anti-DDoS) that filter traffic before it reaches your server. We recommend using Cloudflare, DDoS-Guard or built-in solutions from Yandex Cloud. Setting up request rate limiting and using a WAF also helps. Read about comprehensive protection in the article Information security.

Source: term "DDoS attack"

A sandbox is an isolated environment for safely running suspicious files or testing code. All changes remain inside the sandbox and do not affect the main system. It is like a children's sandbox — you can play, build and break, but everything outside it stays clean. Read about protection from viruses in the article Antivirus software.

Source: term "Sandbox"

For safely analyzing suspicious files and programs. A file is run in an isolated environment where specialists observe its behavior: what it tries to change, what files it creates, where it sends data, what system calls it executes. If it is a virus, it will not be able to infect the main system, and analysts will get information about its behavior to create signatures and protection. Read about protection from threats in the article EDR / XDR.

Source: term "Sandbox"

A virtual machine is a full-fledged computer emulator where you can install an OS and work as on a regular PC. It is isolated, but requires significant resources. A sandbox is usually a more lightweight isolated environment for running individual programs, often at the OS or application level. Virtual machines are often used as sandboxes, but not the other way around. Read about virtualization in the article Virtualization.

Source: term "Sandbox"

Popular solutions: Sandboxie (Windows), Windows Sandbox (built into Windows 10/11 Pro), Cuckoo Sandbox (for malware analysis in corporate SOCs), VirtualBox/VMware (full-fledged VM sandboxes), Firejail (for Linux). The choice depends on the tasks: Sandboxie or Windows Sandbox are suitable for simple testing, Cuckoo or virtual machines for professional analysis. Read about virtual machines in the article Virtual machine.

Source: term "Sandbox"

Yes, some complex viruses can detect that they are running in a sandbox (sandbox evasion) and do not show malicious activity. Also, with incorrect isolation configuration, a virus can “escape” and infect the system (sandbox escape). Therefore, it is important to use proven solutions, regularly update software, use several layers of isolation and combine sandboxes with other protection means such as EDR and SIEM.

Source: term "Sandbox"

Yes, many modern antivirus products use sandboxes for behavioral analysis of suspicious files. When an unknown file is detected, the antivirus can run it in a cloud or local sandbox, analyze its behavior and decide whether to block it. This allows identifying new threats for which signatures do not yet exist. Read about antiviruses in the article Antivirus software.

Source: term "Sandbox"

Windows 10/11 Pro and Enterprise have a built-in Windows Sandbox feature. To enable it, go to Control Panel -> Programs -> Turn Windows features on or off and check “Windows Sandbox”. After a restart you can launch the isolated environment through the Start menu. You can also use Sandboxie or VirtualBox to create virtual machines. For corporate use, it is recommended to design protected environments using professional services.

Source: term "Sandbox"

An API gateway is a single entry point for all requests to microservices. It accepts a request, checks the user's rights, limits the frequency of requests and redirects it to the needed microservice. Like a reception desk in a large office. Read about microservices in the article Client-server architecture.

Source: term "API gateway"

An API gateway centrally handles authentication, authorization, request limitation, routing, data transformation and response aggregation. This offloads microservices and simplifies API management. Read about security in the article WAF.

Source: term "API gateway"

Cloud: Yandex API Gateway, AWS API Gateway, Azure API Management. Open Source: Kong, Traefik, Tyk, Envoy, NGINX. On-premise: Apigee, IBM API Connect. The choice depends on the infrastructure and budget. Read about solutions in the article Proxy server.

Source: term "API gateway"

An ordinary proxy simply redirects requests “as is”. An API gateway is a “smart” proxy: it checks authentication, limits requests, transforms data, aggregates responses from several services and collects analytics. Read about proxies in the article Proxy server.

Source: term "API gateway"

Not exactly. A load balancer distributes load between instances of one service. An API gateway does more: it routes between different services, checks rights and transforms data. It can include load balancing, but that is only one of its functions. Read about load balancing in the article Server cluster.

Source: term "API gateway"

This is a model where a client (your browser or application) requests data, and a server processes it and returns it. Like an order in a restaurant: you (the client) place an order, the kitchen (the server) prepares and serves the dish. Read about how servers work in the article Web server.

Source: term "Client-server architecture"

In a file-server model, files are stored on a server and clients open and process them themselves (for example, shared access to documents). In a client-server model, the client only sends requests, and all processing is done by the server. This is safer and more efficient for complex systems. Read about databases in the article DBMS.

Source: term "Client-server architecture"

Two-tier — the client directly accesses the DB server. Three-tier — an application server with business logic stands between the client and the DB. Multi-tier — caching, queues and integration services are added. Read about corporate systems in the article 1C:Enterprise.

Source: term "Client-server architecture"

Centralized data storage, high security, ease of scaling, backup and updates. Clients do not store data, so if a device breaks, the information will not be lost. Read about data protection in the article Data backup.

Source: term "Client-server architecture"

A single point of failure — if the server goes down, the system is unavailable. Dependence on network quality. High load on the server with a large number of users. To solve these problems, clustering and load balancing are used. Read about clusters in the article Server cluster.

Source: term "Client-server architecture"

Open Source is programs whose code is open to everyone. You can not only use them for free, but also see how they work, modify them for your needs and distribute your versions. It is like a recipe for a dish — you can cook it, change the ingredients and share the recipe. Read about choosing software in the article Alternative software.

Source: term "Open source software (Open Source)"

Linux (OS), LibreOffice (office), Firefox (browser), PostgreSQL and MySQL (DB), NGINX and Apache (web servers), GIMP (graphics), Blender (3D), Python and JavaScript (languages). This is only a small part — there are millions of Open Source projects. Read about solutions in the article Software import substitution.

Source: term "Open source software (Open Source)"

Free Software puts the emphasis on the ethical freedoms of the user. Open Source — on practical advantages (quality, security). In practice it is almost the same thing, but the philosophy is different. Read about licenses in the article Licenses.

Source: term "Open source software (Open Source)"

Free availability, security (the code is checked by thousands of developers), transparency (no hidden functions), vendor independence, flexibility (you can customize it for yourself) and a huge community. Read about selecting in the article Enterprise architecture.

Source: term "Open source software (Open Source)"

On paid services (installation, configuration, support), commercial support (Red Hat), cloud services (SaaS), the Open Core model (the base is free, extensions are paid) and donations/sponsorship. Read about business models in the article Fintech.

Source: term "Open source software (Open Source)"

A server is a dedicated computer that constantly runs and processes requests from other devices: computers, laptops, smartphones. It stores data, launches applications and distributes them to users over the network. For example, when you open a website, the browser sends a request to the web server, and it returns the page. A server differs from an ordinary PC in higher reliability, performance and the ability to work around the clock.

Source: term "Server"

A server is designed for continuous operation under high load and servicing multiple users at the same time. It uses server processors, memory with error correction (ECC), redundant power supplies and disk arrays that protect data from loss. A server usually has no monitor or keyboard — it is managed remotely, and its components are often hot-swappable without stopping operation.

Source: term "Server"

By design: tower (Tower), rack-mount (Rack, installed in a server rack) and blade servers (dense boards in a common chassis). By purpose: file, web, mail, database and application servers, proxy and print servers. In large systems, servers are combined into clusters for load distribution and fault tolerance.

Source: term "Server"

A server allows corporate data to be stored centrally, organizes shared access to files and databases, deploys internal systems (1C, mail, document flow) and provides secure remote access for employees. This is more reliable and convenient than storing data on separate computers, and a server also simplifies backup and information security management.

Source: term "Server"

Assess the number of users and programs, the load and growth plans. For 10–50 users, one server with a single processor, 32–64 GB of RAM, two network ports and a RAID array of several disks is enough. Be sure to provide an uninterruptible power supply and backup. A reliable choice is a 1U–2U rack server with redundant power supplies in a server rack.

Source: term "Server"

Server infrastructure is a set of servers, storage systems, network equipment and software that ensure the operation of a company's information systems. It includes virtualization, clustering, backup and monitoring. For critical systems, servers are placed in data centers with guaranteed power supply and cooling.

Source: term "Server"

The cost depends on the class: a server for a small office costs from 100 to 300 thousand rubles, for a medium business — from 300 thousand to 1 million, high-performance rack and blade servers for data centers — from 1 million. OS and DBMS licenses, electricity, maintenance and technical support are added to the expenses. For the public sector, it is important to take into account import substitution and certification requirements.

Source: term "Server"

One or two small-business servers are usually placed in a server room or a telecommunications cabinet with air conditioning and uninterruptible power. For medium and large businesses, placement in a data center is more profitable: power supply, cooling, secure access, communication channels and maintenance are provided there. Placement in a data center reduces the risks of downtime and relieves the company's IT department.

Source: term "Server"

A personal computer (PC) is a universal computing device for individual work: processing documents, calculations, accessing the internet, programming and multimedia. It consists of a system unit (or an all-in-one), a monitor and input devices and runs under an operating system — Windows, Linux or macOS.

Source: term "Personal computer (PC)"

A PC is designed for one user and performs his local tasks. A server serves many users at once: stores data, runs applications and responds to network requests. Servers use more reliable components, redundant power and work around the clock, while a PC is simpler, cheaper and focused on interactive human work.

Source: term "Personal computer (PC)"

Desktop PCs are easy to upgrade and optimal in performance; all-in-ones are compact, with all components in the monitor case; laptops are portable; nettops are minimalist for the office. Workstations for complex calculations and thin clients for working in virtual desktop infrastructure are distinguished separately.

Source: term "Personal computer (PC)"

The main components: motherboard, processor, RAM, storage drive (SSD or HDD), video card, power supply and case. The presence and power of a discrete video card determine the suitability of the PC for games and graphics, and the amount of memory and the speed of the drive determine performance in work tasks.

Source: term "Personal computer (PC)"

For most office tasks, a PC with a mid-range processor, 8–16 GB of RAM, a 512 GB SSD and integrated graphics is enough. Organizations should take into account the requirements of tender documentation, including the possibility of delivery under import substitution programs, certified OS and warranty service.

Source: term "Personal computer (PC)"

A desktop PC gives more performance for the same money, is easier to upgrade and simpler to maintain. A laptop is compact, allows working anywhere and consumes less energy, but is more expensive, limited in expansion and less reliable under intensive load. The choice depends on the mobility of the employee and the nature of the tasks.

Source: term "Personal computer (PC)"

A workstation is a high-performance PC for CAD, 3D modeling, video editing and scientific calculations: a multi-core processor, 32–128 GB of memory and a professional video card. The term “immersive PC” is used for virtual and augmented reality systems that require maximum graphics performance.

Source: term "Personal computer (PC)"

A computer is upgraded when it stops coping with tasks: system boot takes minutes, programs “slow down”, there is not enough memory and disk space. Most often it is enough to increase RAM and replace the HDD with an SSD — this gives a noticeable speed gain without replacing the whole PC. Full replacement is planned after 5–8 years of operation or when key components fail.

Source: term "Personal computer (PC)"

A network switch is a device that connects computers, servers and other network nodes into one local network and transmits data strictly between the required devices. It is used to connect computers in an office, power cameras and access points, organize VLANs and build a network core in a data center.

Source: term "Network switch"

A switch operates within one local network and transmits frames between its nodes based on MAC addresses. A router connects different networks and directs packets between them by IP addresses, and also provides internet access, NAT and inter-network protection.

Source: term "Network switch"

Unmanaged — simple, without configuration; managed — with VLAN, QoS and monitoring; access, aggregation and core switches — by the level of the network hierarchy; PoE switches — power devices over a network cable; industrial — for harsh operating conditions.

Source: term "Network switch"

Count the number of devices to be connected and add 20–30% of headroom for growth. For a small office, 8–24 ports are usually enough, for a medium company — 48-port models with several 10 Gbps uplink ports to the server room and higher-level equipment.

Source: term "Network switch"

VLAN (virtual local area network) is a function of a managed switch that allows one physical network to be logically divided into several isolated segments: for example, separate VLANs for employees, guests, servers and IP telephony. This increases security and simplifies traffic management.

Source: term "Network switch"

PoE (Power over Ethernet) is a technology for supplying power to a device over the same network cable that carries data. PoE switches power surveillance cameras, WiFi access points and IP phones, which simplifies installation and does not require separate outlets at each device.

Source: term "Network switch"

A managed switch allows configuring VLANs, QoS priorities, port aggregation, mirroring and traffic monitoring. This is necessary for corporate networks that require segmentation, performance control and information security.

Source: term "Network switch"

For the data center core, high-performance 10/25/40/100 Gbps switches with redundant power supplies and fans, support for expansion modules and management via CLI/NETCONF are chosen. For server racks, models with port aggregation for connection to the core and storage systems are used.

Source: term "Network switch"

A router is a device that connects networks to each other and directs internet traffic along the right path. A home router provides internet to devices, distributes Wi-Fi and assigns each device a local address via DHCP.

Source: term "Router"

A switch operates within one network and transmits frames by MAC addresses. A router connects different networks and directs IP packets along routes, providing internet access, NAT and inter-network protection. A home router combines a router, a switch and an access point.

Source: term "Router"

Home and office — all-in-one with Wi-Fi; corporate — with VPN for branches and support for multiple providers; edge — at the edge of the network; core — in the provider backbone network; industrial — for harsh conditions.

Source: term "Router"

Routing is the selection of the optimal path for transmitting packets between networks. The router looks at the recipient's IP address, consults the routing table (static or built by OSPF/BGP protocols) and forwards the packet to the required network.

Source: term "Router"

NAT (network address translation) allows many devices on an internal network to access the internet through one external IP address. The router changes the internal source address to its external one and remembers the connection to return the response to the right device. This also hides the internal network from the outside world.

Source: term "Router"

The number of employees and devices, the required channel speed, the number of wired ports (usually 8–48) and the need for management are assessed. For a branch network, corporate routers with VPN, QoS and support for internet channel redundancy are chosen.

Source: term "Router"

DHCP is a service for automatic assignment of network settings to devices: IP address, subnet mask, gateway and DNS servers. Thanks to DHCP, devices connect to the network without manual configuration — the router itself issues the parameters to them for the lease period.

Source: term "Router"

Static routing — routes are written manually by the administrator: simple and predictable, but does not adapt to failures. Dynamic routing uses protocols (OSPF, BGP) that automatically build route tables and bypass failed links, which is convenient in large networks.

Source: term "Router"

A UPS is a device that provides uninterrupted power to equipment during power grid failures. It protects servers and computers from data loss during power outages and from damage during voltage surges and interference.

Source: term "UPS (uninterruptible power supply)"

Standby (Offline) — switch to batteries when the mains fails; line-interactive (Line-Interactive) — additionally stabilize voltage; double conversion (Online) — the load is constantly powered through the inverter, ideal for servers and data centers.

Source: term "UPS (uninterruptible power supply)"

Sum up the power consumption of the equipment in watts and add 20–30% of reserve. For example, for a server with an 800 W power supply, a UPS of 1000–1200 VA is suitable. Power is specified in volt-amperes (VA) or watts (W) — compare the model with the required load.

Source: term "UPS (uninterruptible power supply)"

5–15 minutes of autonomy is enough for a correct system shutdown. For longer operation support, external batteries are added or a diesel generator is connected, which takes over the load after the UPS stops.

Source: term "UPS (uninterruptible power supply)"

An Online UPS constantly powers the load from the inverter through batteries, so the voltage is always stable and there is no switching. A Line-Interactive UPS corrects deviations with a stabilizer, but briefly switches to batteries during a complete outage. For critical servers and storage systems, Online is chosen.

Source: term "UPS (uninterruptible power supply)"

Yes — a UPS will protect the PC from loss of unsaved data and damage during an outage or voltage surge. For home and office, a 600–1000 VA line-interactive UPS is enough for a correct shutdown.

Source: term "UPS (uninterruptible power supply)"

This is a function that allows replacing batteries without disconnecting the UPS and the load. It allows maintaining an uninterruptible power supply in the server room without interrupting the operation of critical equipment and is available in models with removable battery modules.

Source: term "UPS (uninterruptible power supply)"

A UPS provides instant battery backup operation (seconds to minutes) but has a limited capacity. A generator provides long-term power but takes several minutes to start and spin up. In server rooms they are used together: the UPS supports the load while the diesel generator starts.

Source: term "UPS (uninterruptible power supply)"

SCS is a standardized cable infrastructure of a building that uniformly connects workstations with switches and servers. It is used for computer networks, telephony and security systems, simplifies maintenance and allows the network to be expanded without reworking the cable wiring.

Source: term "Structured cabling system (SCS)"

SCS includes a work area (outlets), a horizontal subsystem (cables to the floor cabinet), a patch field (patch panels and patch cords), a backbone between floors and an equipment room with the network core. All lines are documented and tested.

Source: term "Structured cabling system (SCS)"

Ordinary laying is done chaotically and for a specific task. SCS is built according to standards with a strict hierarchy and universal outlets: a computer, phone or camera can be connected to the same outlet. This simplifies network administration and scaling.

Source: term "Structured cabling system (SCS)"

For new offices, Cat 6A is recommended — up to 10 Gbps and PoE+ support. For typical tasks, Cat 5e/6 (up to 1–10 Gbps) is enough. Fiber optics are used between buildings and floors. The choice is made with a headroom of 10–15 years of network development.

Source: term "Structured cabling system (SCS)"

SCS installation includes design of the scheme, laying of cable routes and cables, installation of cabinets and patch panels, mounting of outlets at workstations, labeling and testing of each line for category compliance. At the end, the customer receives documentation and a system passport.

Source: term "Structured cabling system (SCS)"

The backbone subsystem connects telecommunications rooms to each other and between buildings. It is built on fiber optic cables (SM/MM) and/or copper pairs and provides the network core channels connecting floor switches to the server room.

Source: term "Structured cabling system (SCS)"

The cost depends on the area, the number of workstations and the cable category. The estimate includes cables, cabinets, patch panels, outlets and installation and testing work. An accurate assessment is given by the project after a site survey — contact a system integrator.

Source: term "Structured cabling system (SCS)"

Yes. Standard SCS outlets are used to connect IP cameras and Wi-Fi access points when PoE is supported by the switches. The universality of SCS allows LAN, telephony, video surveillance and ACS to be combined in one infrastructure.

Source: term "Structured cabling system (SCS)"

A laptop is a portable computer with a built-in screen, keyboard and battery. You can take it to work, meetings and travel; it runs offline for several hours and, when connected to the grid, works like a regular PC.

Source: term "Laptop"

A laptop packs all components into a compact case with a built-in battery and screen; it is mobile and does not take up space. A desktop PC is cheaper, more powerful and easier to repair or upgrade, but requires a fixed position and outlets.

Source: term "Laptop"

Office (home) — for documents and internet; gaming — with a powerful video card; professional — for video, 3D and design; business — with long battery life and security; ultrabooks — thin, light and mobile; graphics workstations — for computing loads.

Source: term "Laptop"

Focus on tasks. For documents and mail — 8 GB RAM, SSD and an Intel Core i3/Ryzen 3 class processor. For video and design — 16–32 GB RAM and a discrete video card. Important: screen quality (Full HD IPS), ports and 3–5 hours of battery life.

Source: term "Laptop"

Modern working models run on battery for 4–10 hours: office tasks — about 8–10 hours, video calls and multi-window work reduce the time to 4–6 hours. Battery life depends on brightness, load and energy-saving settings.

Source: term "Laptop"

Yes. Connect an external monitor, keyboard and docking station to a laptop and it works like a full-fledged workplace. This scheme — a laptop with a dock — combines the mobility of a portable computer and the ergonomics of a desktop at the office.

Source: term "Laptop"

The disk (SSD) affects loading speed and system responsiveness — 256–512 GB is a comfortable minimum. RAM affects multitasking and the ability to run heavy programs. The ideal is a fast SSD plus 16 GB RAM; both are usually expanded only by some laptops.

Source: term "Laptop"

A programmer needs a fast processor (Intel Core i5/i7, Ryzen 5/7), 16–32 GB RAM, an SSD of 512 GB or more and a good keyboard. The operating system is chosen for the stack (Windows, macOS or Linux). A comfortable 14–15″ screen and a long battery life are desirable.

Source: term "Laptop"

A monitor is a screen that displays information from a computer. It converts the signal from the video card into an image, allows seeing documents, sites and programs. Diagonal, resolution and matrix type determine the image quality.

Source: term "Monitor"

IPS gives the most accurate color reproduction and wide viewing angles; VA provides deep black and high contrast; TN has a fast response but poor angles and colors. For office and design work, IPS, for multimedia, VA, for fast games, TN or fast IPS are recommended.

Source: term "Monitor"

Full HD (1920×1080) is enough for office text work. For graphics and CAD, 2K (2560×1440) or 4K (3840×2160) is better — more workspace and sharpness. The higher the resolution, the more powerful the video card is required.

Source: term "Monitor"

The refresh rate (60/144/240 Hz) shows how many times per second the image updates. High frequency makes the picture smoother: fast movement in games and scrolling does not blur. For documents and regular work, 60 Hz is enough.

Source: term "Monitor"

This is a function that synchronizes the screen refresh rate with the video card output, synchronizing the frame output. As a result, graphic crosses and tearing disappear during games and dynamic scenes. Useful for gamers, designers and video editors.

Source: term "Monitor"

For design and finance, a 27″ IPS with 2K/4K resolution, accurate color rendering (covering sRGB/Adobe RGB) and speaker calibration is recommended. A 100% gamut guarantee and width for comfortable work with documents are important.

Source: term "Monitor"

Two monitors increase workspace: on one you open mail and chat, on the other — working documents or a trading terminal. This reduces window switching and speeds up work for analysts, operators and developers.

Source: term "Monitor"

For an office, a 23–27″ IPS with Full HD resolution and eye protection (flicker-free, low blue light) is recommended. Adjustable stand height and comfortable eye training reduce fatigue during a shift. Ports for connecting to a laptop docking are also useful.

Source: term "Monitor"

A network card (NIC) is a device for connecting a computer to a network. It forms network frames, labels them with its MAC address, transmits them over cable or Wi-Fi and receives data addressed to this computer.

Source: term "Network interface card (NIC)"

In data transfer speed: 1 Gbps is a classic for workstations and small servers; 10 Gbps and higher is for data center servers, storage systems and the network core, where high-load bandwidth is required.

Source: term "Network interface card (NIC)"

A MAC address is a unique six-byte address defined by the manufacturer (the first three bytes) and the device serial number. In Windows it is viewed with the ipconfig /all command, in Linux — with ip link, on servers — on the label or in the OS interface.

Source: term "Network interface card (NIC)"

Several ports allow aggregating channels (increasing bandwidth and reliability), separating traffic (management/data/backup) and connecting the server to several networks at once. The cards support VLAN and protocol offload.

Source: term "Network interface card (NIC)"

Offload is the transfer of part of the processing of network protocols (checksums, packet segmentation) to the card hardware instead of the processor. This reduces the CPU load and speeds up processing of large data flows on servers.

Source: term "Network interface card (NIC)"

For a laptop, a built-in Wi-Fi 6/6E is usually enough; for stable operation over cable, a USB Gigabit Ethernet adapter is used. For professional tasks, a USB-C adapter with 2.5–10 Gbps support is suitable.

Source: term "Network interface card (NIC)"

SFP/SFP+ is a removable optical module installed in a slot on a network card or switch for connecting fiber optics. It determines the speed (1–25 Gbps) and the type of line (single-mode or multimode fiber).

Source: term "Network interface card (NIC)"

A network card operates at the second (data link) layer of the OSI model: it forms and transmits frames with MAC addresses in the transmission medium. Routing between networks is performed higher, at the network layer, by routing devices.

Source: term "Network interface card (NIC)"

An access point (AP) is a device that creates a wireless Wi-Fi network and connects clients to it. It connects by cable to a switch and broadcasts the network over the radio, allowing laptops and smartphones to work without wires.

Source: term "Access point"

A router hands out IP addresses, performs NAT and provides internet access. An access point only adds wireless access to an existing network, without doing routing. A home router combines both a router and an access point.

Source: term "Access point"

Modern Wi-Fi 6 access points support from 50 to 200 clients depending on the model and load. With a large number of devices and high speed requirements, a network of several APs with a controller is built.

Source: term "Access point"

Roaming is a technology where a client automatically switches between access points without dropping the connection when moving around a building. The controller coordinates the APs, and a video call or work session is not interrupted.

Source: term "Access point"

PoE is powering a device over a network cable. An access point receives power and data over one twisted pair, which simplifies mounting on a ceiling or wall and does not require a separate outlet at the installation point.

Source: term "Access point"

A radio survey is conducted to calculate the number of access points, they are mounted on the ceiling with PoE power, connected to switches and configured with a single SSID with roaming through a controller. Guest traffic is isolated via VLAN and limited.

Source: term "Access point"

Wi-Fi 5 (802.11ac) — up to 3.5 Gbps at 5 GHz; Wi-Fi 6 (802.11ax) — higher speed, better performance in dense environments and lower latency; Wi-Fi 6E — adds the 6 GHz band. The choice of standard determines the speed and capacity of the network.

Source: term "Access point"

If the apartment is large or the walls weaken the signal of the built-in router, a separate access point or a mesh system provides stable coverage. The access point is connected by cable to the router and creates an additional Wi-Fi zone.

Source: term "Access point"

A patch panel is a panel with a row of ports (usually 24 or 48) mounted in a rack. Workstation cables are connected to the ports on the back, and on the front they are connected with patch cords to a switch, creating a neat switching point.

Source: term "Patch panel"

A patch panel organizes the cabling system: trunk lines are not connected directly to the switch, but come to the panel. Moving and replacing equipment is done with a short patch cord without redoing the cable runs.

Source: term "Patch panel"

A patch panel is a passive device: it only connects connectors and physically passes the signal without processing data. A switch is an active device that receives frames and directs them by MAC addresses to the desired port.

Source: term "Patch panel"

For standard office networks, category 5e (up to 1 Gbps) is enough, for modern 2.5–10 Gbps — category 6 (up to 10 Gbps over 55 m) and 6A (up to 10 Gbps over 100 m). The panel category must match the cable and outlets.

Source: term "Patch panel"

An optical patch panel contains LC or SC adapters for fiber optics and is used for backbone connections between cabinets and rack rows. It provides high speed and protection from electromagnetic interference.

Source: term "Patch panel"

Panels with 24 and 48 ports (1U high) are common. The 1U format with 48 ports saves space in the rack, but is denser to route. For small offices, one or two 24-port panels are enough.

Source: term "Patch panel"

A cable organizer is a bar with holes or a channel installed under the patch panel for neat routing of patch cords. It maintains the bend radius, reduces clutter and makes rack maintenance easier.

Source: term "Patch panel"

Shielded panels (FTP/STP) are used where there is strong interference and high requirements for protection of transmitted data: in industrial areas and near power lines. In a regular office, an unshielded UTP panel is enough.

Source: term "Patch panel"

An all-in-one PC is a computer in which the system unit and monitor are combined in one case. It takes up less space than a classic PC and connects to power, keyboard and mouse with one set of wires.

Source: term "All-in-one PC"

There is no separate system unit in an all-in-one PC: the processor, memory and storage are located behind the display. An all-in-one PC is harder to upgrade — usually only RAM and storage replacement are available, while in a PC almost all components can be changed.

Source: term "All-in-one PC"

In most all-in-one PCs, RAM and an SSD can be replaced — service compartments are provided for this. The processor and video card in compact cases are soldered and cannot be replaced.

Source: term "All-in-one PC"

Office all-in-one PCs with a 21–24-inch diagonal, universal 27-inch models for design and multimedia, as well as compact models for cash registers and information kiosks. By screen, IPS models and touchscreen options are distinguished.

Source: term "All-in-one PC"

For office tasks, a processor of the Core i3 or Ryzen 3 level, 8 GB of RAM and an SSD from 256 GB are enough. Stable domain operation, a quiet cooler and vendor warranty support are important.

Source: term "All-in-one PC"

Office models cost comparably to classic PCs of the same configuration, while models with accurate color rendering and touchscreens are more expensive. The final price depends on the diagonal, processor and amount of memory.

Source: term "All-in-one PC"

Usually all-in-one PCs are equipped with USB ports (including USB-C), HDMI or DisplayPort for an external monitor, an RJ-45 jack for a wired network and an audio jack. The set of ports depends on the model and manufacturer.

Source: term "All-in-one PC"

An all-in-one PC has a larger display, often more powerful components at the same price and does not require a docking station. A laptop wins in mobility, so the choice depends on whether work outside the workplace is needed.

Source: term "All-in-one PC"

A thick client is a computer that processes applications and data locally, without constant dependence on a server. Programs run on the device itself, and the server is used for storage and data exchange.

Source: term "Thick client"

A thick client processes data itself and works autonomously, while a thin client only displays the image created by a server. A thin client is cheaper and simpler to maintain, but requires a stable network.

Source: term "Thick client"

For office tasks, a modern processor, 8 GB of RAM and an SSD from 256 GB are enough. Requirements grow if the application works with heavy graphics or local databases.

Source: term "Thick client"

Yes, a thick client continues to work when the network fails, using local data and applications. After the connection is restored, the changes are synchronized with the server.

Source: term "Thick client"

In banks and companies they are used for teller workstations, programs with cryptography and applications working with peripheral equipment: printers, scanners and document readers.

Source: term "Thick client"

The equipment is more expensive compared to thin clients, centralized updating is more difficult and the risk of data leakage is higher if sensitive information is stored locally.

Source: term "Thick client"

Yes, an all-in-one PC can work as a thick client if applications and data are processed on the device itself. The term describes an architecture, not a specific type of computer.

Source: term "Thick client"

Centralized management, disk encryption, application updates and security policies that restrict the installation and launch of unauthorized programs are needed.

Source: term "Thick client"

A printer is a device for outputting digital documents and images to paper or other media. It receives print jobs from a computer or network and outputs them on sheets of the required format.

Source: term "Printer"

Laser — for mass printing of documents, inkjet — for high-quality color printing, dot matrix — for checks and multi-layer forms, thermal — for checks and labels on thermal paper.

Source: term "Printer"

A printer only prints. An MFD additionally scans, copies and sends documents by email or fax, which is convenient if document digitization is needed.

Source: term "Printer"

For large volumes, a laser printer with duplex printing and a network port is suitable. For bank operating halls — compact models with low noise and check printing.

Source: term "Printer"

A print server is a computer or device that accepts print jobs from users and sends them to network printers. It centralizes print management and its accounting.

Source: term "Printer"

Laser printers usually have a speed of 20–50 pages per minute, inkjet ones — 10–30. The speed depends on the quality mode and the type of document.

Source: term "Printer"

These are cartridges with toner or ink, image-forming units and thermal heads. The cost of ownership depends on the price per print and the life of the consumables.

Source: term "Printer"

Authorization at the device, holding jobs on the print server until the user confirms, and print accounting are used. This allows controlling who prints what.

Source: term "Printer"

An MFD is a multifunction device that combines a printer, scanner and copier in one case. Many models also send documents by email or fax.

Source: term "MFD (multifunction device)"

A printer only prints, while an MFD also scans and copies. This allows digitizing documents and making copies without separate devices, saving space and money.

Source: term "MFD (multifunction device)"

Laser — for fast printing of documents, inkjet — for high-quality color printing. By function, models with a copier, automatic document feeder and fax are distinguished.

Source: term "MFD (multifunction device)"

An automatic document feeder (ADF) is a mechanism that automatically passes a stack of sheets through the scanner. It is convenient for quickly digitizing multi-page documents: you do not need to place each sheet separately.

Source: term "MFD (multifunction device)"

For document flow, a laser device with a speed of from 25 pages per minute, an automatic feeder and a network connection is suitable. The cost per print and cartridge life are important.

Source: term "MFD (multifunction device)"

Scanning turns a paper document into an electronic copy — an image or PDF. The copy can be saved to a computer, a network folder or sent by email.

Source: term "MFD (multifunction device)"

The price per print depends on the type of device and consumables. Laser black-and-white models give the cheapest print among office equipment, inkjet color ones are more expensive.

Source: term "MFD (multifunction device)"

Yes, most modern MFDs support network connection via RJ-45 or Wi-Fi. Several employees can send print and scan jobs from a shared device.

Source: term "MFD (multifunction device)"

A scanner is a device for transferring a paper document or image into an electronic copy: an image or PDF file. It illuminates the page and captures the reflected light with a sensor.

Source: term "Scanner"

Flatbed — the document is placed on glass, sheet-fed — with an automatic feeder for a stack of sheets, handheld and book. There are also separate scanners for passports and barcodes.

Source: term "Scanner"

A sheet-fed scanner automatically draws sheets in and scans them on both sides in one pass. It processes tens of pages per minute and is suitable for mass digitization.

Source: term "Scanner"

A scanner only digitizes documents, while an MFD additionally prints and copies. For large digitization volumes, a separate sheet-fed scanner is faster and more reliable.

Source: term "Scanner"

300 dpi is enough for text, 600 dpi — for barcodes and banking documents, 1200 dpi and higher — for photos and small details. High resolution increases the file size.

Source: term "Scanner"

OCR (text recognition) is a software function that turns a scanned image into editable text. It allows searching and processing the contents of documents.

Source: term "Scanner"

The scanner saves files to a shared folder or document flow system. The program recognizes text, rotates pages and transfers the document to the repository with accounting and a signature.

Source: term "Scanner"

They quickly and reliably digitize applications, passports and payment orders. Scanners with IR and UV illumination recognize document security features, speeding up customer service.

Source: term "Scanner"

Server components are the hardware elements of a server: processors, RAM, storage drives, power supplies and network controllers. They are designed for round-the-clock load and high reliability.

Source: term "Server components"

They are designed for constant load, support more cores, memory and communication channels. Server memory uses ECC error correction, and storage drives have an increased life and protection against power loss.

Source: term "Server components"

ECC is memory error correction. ECC modules detect and correct single-bit errors, which is critical for databases and financial systems, where data correctness matters most.

Source: term "Server components"

SSDs for operating systems and databases (fast, with a large write endurance) and HDDs for archives and backup copies (capacious). NVMe drives over PCIe give maximum speed.

Source: term "Server components"

RAID is combining storage drives into an array for reliability and speed. Mirroring (RAID 1) duplicates data, striping with parity (RAID 5/6) stores recovery information, allowing a disk failure to be survived.

Source: term "Server components"

1+1 redundancy: if one power supply fails, the second takes over the load. This ensures continuous operation and the ability to replace without stopping the server.

Source: term "Server components"

A BMC (iLO, iDRAC) is a server service board that allows remotely powering on, powering off and monitoring the server even when the operating system is off. It is the basis of remote administration.

Source: term "Server components"

Start from the tasks: a file server requires capacious drives, databases need fast CPUs and a lot of ECC memory, virtualization needs a balance of cores and RAM. Balance is important: a bottleneck reduces all performance.

Source: term "Server components"

A media converter is a device that converts a signal between copper cable (twisted pair) and optical fiber. It allows connecting copper devices through an optical backbone over long distances.

Source: term "Media converter"

Twisted pair works at distances up to 100 meters, and fiber optics — for kilometers. A media converter links these media when you need to connect buildings or a remote object without replacing the switch.

Source: term "Media converter"

By speed — 10/100 Mbps and gigabit, by power — with PoE for cameras and access points, by design — desktop and chassis for racks. The optical port is often made as an SFP slot.

Source: term "Media converter"

Single-mode converters work with single-mode fiber at distances up to tens of kilometers, multimode ones — with multimode fiber over short distances (up to 500 m and less). The choice depends on the line length.

Source: term "Media converter"

PoE (power over Ethernet) allows the converter to supply power to a connected device — a camera or access point — over twisted pair. This eliminates the need for a separate power cable.

Source: term "Media converter"

Connect the twisted pair to the RJ-45 port, insert a suitable SFP module into the optical port and connect it to the fiber. The paired converter at the other end performs the reverse conversion.

Source: term "Media converter"

A chassis is a multi-module case for installation in a rack, combining dozens of converters with common power and redundancy. It is convenient for centralized placement in a server room.

Source: term "Media converter"

Yes, it operates at the physical layer and does not analyze packets. The converter is transparent to protocols: it only changes the signal form between the copper and optical media.

Source: term "Media converter"

An SFP module is a removable transceiver for connecting optics to a switch, server or media converter. It sets the speed, fiber type and range of the communication line.

Source: term "SFP module (transceiver)"

SFP supports speeds up to 1 Gbps, SFP+ — up to 10 Gbps. The form factor is compatible, but the speed is determined by both the module and the device port.

Source: term "SFP module (transceiver)"

Take into account the port speed, fiber type and range: multimode 850 nm up to 550 m, single-mode 1310/1550 nm over tens of kilometers. Modules in a pair must have the same wavelength.

Source: term "SFP module (transceiver)"

SX — multimode up to 550 m, LX — up to 10 km, EX — up to 40 km, ZX — up to 80 km. There are BiDi over one fiber, DWDM for several channels and SFP-T for copper.

Source: term "SFP module (transceiver)"

Yes, SFP is designed for hot swapping: the module can be inserted or removed while the device is running. The port may temporarily lose the connection at the same time.

Source: term "SFP module (transceiver)"

Single-mode modules work with single-mode fiber over long distances, multimode ones — with multimode over short distances. The module type must match the fiber type.

Source: term "SFP module (transceiver)"

BiDi (bidirectional) transmits and receives a signal over one fiber at different wavelengths. This saves fiber when cable capacity is insufficient.

Source: term "SFP module (transceiver)"

Modules are a sensitive component that can fail. A reserve of identical models and documenting transmitter-receiver pairs speed up the restoration of a communication line.

Source: term "SFP module (transceiver)"

An optical communication line (FOL) is a data transmission channel over fiber optics using light pulses. It provides high speed, long range and immunity to electromagnetic interference.

Source: term "Optical communication lines"

Optics transmits data over kilometers instead of 100 meters for copper, supports speeds up to hundreds of Gbps and is not subject to interference. But it is more expensive to install and maintain.

Source: term "Optical communication lines"

Single-mode fiber has a small core and transmits one beam over long distances. Multimode has a large core with several beams and is used over short distances.

Source: term "Optical communication lines"

A laser source turns an electrical signal into light pulses that travel along the fiber. At the receiving end, a photodetector returns the light to electrical form.

Source: term "Optical communication lines"

The optical signal is not emitted outward and does not create electromagnetic interference, which reduces the risk of unauthorized data extraction. A fiber break is easily detected by instruments.

Source: term "Optical communication lines"

An optical distribution frame is a device for connecting and distributing optical cables. It secures fibers, provides neat switching and communication line control.

Source: term "Optical communication lines"

Fibers are joined by fusion welding — this gives minimal signal loss. The welding points are protected with cassettes in closures or distribution frames and checked for losses with a reflectometer.

Source: term "Optical communication lines"

For connecting buildings and remote branches, backbones between cabinets, video surveillance systems and banking networks. Optics transmits data, voice and video streams simultaneously.

Source: term "Optical communication lines"

A data center cooling system is a set of equipment for removing heat from servers and maintaining an acceptable temperature and humidity in the data center. It prevents overheating and failures of equipment.

Source: term "Data center cooling system"

Air cooling with precision air conditioners, liquid (water) cooling with cold liquid supplied to servers, and their combinations. Free cooling with outside air is used to save money.

Source: term "Data center cooling system"

Servers are installed in rows: the front sides form a “cold” aisle, where cold air is supplied, and the rear sides form a “hot” aisle, where heated air is discharged. Their separation increases cooling efficiency.

Source: term "Data center cooling system"

The acceptable air temperature at the server intake is usually 18–27 °C, humidity 40–60%. The exact values are set by equipment vendors and the ASHRAE recommendations.

Source: term "Data center cooling system"

A precision air conditioner (CRAC/CRAH) cools the air in a machine room around the clock with high accuracy. CRAC uses a refrigeration circuit, CRAH — cold water from a chiller.

Source: term "Data center cooling system"

Liquid cooling supplies cold liquid directly to servers through rear door heat exchangers or submerges the equipment in a dielectric liquid. It is more efficient than air at high rack density.

Source: term "Data center cooling system"

N+1 redundancy means that one reserve unit is added to the required number of air conditioners: if any unit fails, cooling continues without interruption.

Source: term "Data center cooling system"

PUE is a data center energy efficiency indicator, the ratio of total energy consumption to IT equipment consumption. A value close to 1 means efficient infrastructure and cheap cooling.

Source: term "Data center cooling system"

A KVM switch is a device for managing several servers and computers from one keyboard, monitor and mouse. It saves space in the rack and simplifies access to equipment.

Source: term "KVM switch"

KVM stands for Keyboard, Video, Mouse. The switch combines these three ports of several computers into one control console.

Source: term "KVM switch"

An IP-KVM is a switch with a network interface for remote management of servers. The administrator works with the BIOS and the system over the network from any computer, without being at the rack.

Source: term "KVM switch"

Desktop — for several computers, console — built into a rack with a retractable screen, local and remote (IP). Models differ in interfaces: VGA, DVI, HDMI, DisplayPort.

Source: term "KVM switch"

It allows managing many servers with one set of keyboard and monitor, saving space in the rack and the administrator's time. An IP-KVM gives access to servers without physical presence.

Source: term "KVM switch"

Switching is done with hot keys, a button on the case or through an on-screen menu. Modern models support automatic discovery of connected computers.

Source: term "KVM switch"

No, a KVM is transparent to the operating system: servers see an ordinary keyboard, monitor and mouse. Special software is required only for the extended functions of an IP-KVM.

Source: term "KVM switch"

Access is protected by encryption, authentication and differentiation of user rights. Built-in audit records sessions, and power redundancy increases fault tolerance.

Source: term "KVM switch"

A passport reader is a device that automatically recognizes identity document data, reads the machine-readable zone (MRZ), the photo and checks security features.

Source: term "Passport and visa document reader"

The MRZ is one or two lines of characters in the lower part of the passport page encoding the full name, date of birth, document number and validity period. The reader recognizes it with a camera and fills in the data automatically.

Source: term "Passport and visa document reader"

The device illuminates the document with visible, infrared and ultraviolet light and checks security features: engraving patterns, marks and elements invisible in ordinary light.

Source: term "Passport and visa document reader"

Desktop ones with a passport tray, standalone wall-mounted terminals for checkpoints and mobile models. There are versions with an RFID module for contactless passports.

Source: term "Passport and visa document reader"

It automatically fills client data into applications and contracts, recognizing the MRZ and photo. This speeds up service and reduces manual entry errors.

Source: term "Passport and visa document reader"

Yes, models with a built-in NFC/RFID module read the data of a contactless passport chip. Special access keys provided by the standards are used to read the chip.

Source: term "Passport and visa document reader"

Usually 1–2 seconds. The device captures the page, recognizes the MRZ and photo, and the software transfers the data to the required system.

Source: term "Passport and visa document reader"

At checkpoints and entrances to register visitors, in access control systems, security services and hotels. The data is checked against the photo and verified for authenticity.

Source: term "Passport and visa document reader"

An ACS terminal is an access control device that checks a user identifier (card, PIN, biometrics) and controls a lock or turnstile. It is the point of entry to the access control system.

Source: term "ACS terminal"

Card (RFID), with a PIN code, biometric (fingerprint, face) and combined. By installation site — wall-mounted for doors, passage ones for turnstiles and outdoor ones with IP65 protection.

Source: term "ACS terminal"

An employee presents an identifier, the terminal checks it against the rights database and, if allowed, sends a signal to the actuator. All events are written to a log.

Source: term "ACS terminal"

It is verification by two factors, for example, a card and a fingerprint or a PIN code. It is used for vaults and rooms with special security requirements.

Source: term "ACS terminal"

Usually 3–10 cm for standard cards and up to several meters for long-range readers. The choice depends on the type of identifier and the task.

Source: term "ACS terminal"

Anti-passback is a function that does not allow passing through a turnstile again until the previous door has closed. It prevents strangers from following an employee.

Source: term "ACS terminal"

These are terminal connection interfaces: Wiegand transmits a signal from the reader to the controller, RS-485 is an industrial bus for connecting several devices to the ACS server.

Source: term "ACS terminal"

In server rooms, cash zones, vaults and service premises. They work with passport readers for visitors and fingerprint scanners for employees.

Source: term "ACS terminal"

It is a biometric device that reads the papillary pattern of a finger and compares it with a saved template. It is used for identification in ACS and authentication systems.

Source: term "Fingerprint scanner"

Optical ones photograph the finger, capacitive ones measure the charge difference, ultrasonic ones penetrate under the skin, thermal ones record temperature. The choice depends on the conditions and protection requirements.

Source: term "Fingerprint scanner"

During registration, the fingerprint is read and turned into a protected template. During verification, the finger is scanned again, the template is compared with the saved one and, on a match, access is allowed.

Source: term "Fingerprint scanner"

Modern scanners are protected from forgery: they check for a live finger by blood flow and use capacitive or ultrasonic technologies. The level of protection depends on the model and the tasks.

Source: term "Fingerprint scanner"

FAR is the false acceptance rate of an unauthorized person, FRR is the false rejection rate of a registered user. For banking systems, devices with a low FAR are important.

Source: term "Fingerprint scanner"

Templates are stored in a protected repository — the terminal database or the ACS server, often in the form of a mathematical model, not an image. Their processing is carried out with consent and in accordance with personal data law.

Source: term "Fingerprint scanner"

Moisture and skin damage distort the papillary pattern, so some scanners refuse to work with wet or dirty fingers. Ultrasonic models work better under such conditions.

Source: term "Fingerprint scanner"

At checkpoints and vaults for strict access control, at workstations to log into the system and confirm operations. They are combined with passport readers and ACS.

Source: term "Fingerprint scanner"