Architectural principle of building networks

Today it is hardly possible to surprise anyone with the concept network connections. However, when we mention them, many of us don’t even think much about what such a connection is and how network services function. We will consider this issue briefly, as we will talk about networks and their capabilities in modern world you can write a long monograph.

Network architecture: main types

Computer networks, as follows from the basic interpretation of the term itself, represent a certain number of computer terminals connected to each other and forming a network. Today there are two main types of connections: wireless and wired. Wireless connection uses a connection via a router like Wi-Fi router. However, this is just the tip of the iceberg. The network architecture actually involves the use of several components at once, and therefore may have different classification. Today, it is customary to distinguish three types of networks: peer-to-peer networks, networks with dedicated servers, and hybrid networks that include all types of nodes. In addition to this, a separate category is represented by broadcast, local, global, private and other varieties. We will focus only on the basic concepts.

Description of networks by main types

The first place to start is with networks based on interaction " main computer on the client network." As should already be clear, the main point in in this case occupies the central terminal where the network and its components are managed. Client terminals can only send requests to provide connections and to receive information. In such a network, the main terminal cannot play the role of a client machine. Peer-to-peer networks, which are often called peer-to-peer networks, differ from the first type in that in them resources are equally distributed among all connected terminals. The simplest example is the processes of downloading files when using torrents. With such an organization final file fully or partially downloaded can be located on various computer terminals. User system, which downloads it to her computer, applies all available on at the moment network resources to download parts of the file you are looking for. The more such files there are, the higher the download speed will be. In this case, network addressing does not play a special role. The main condition is that the client machine has a special software. It will carry out client requests. The client-server network architecture is the simplest. For a simplified understanding, the connection between computer terminals can be thought of as a library in which there are shelves of books (central server), and visitors can read any material that is on the shelves. There is a relationship here: a visitor comes to the library, registers or provides already registered personal data, and after that looks for the necessary literature and reads it. This comparison is quite primitive. Modern networks work much more difficult. However, such an example is impossible would be better suited for simplified understanding.

Terminal identification issue

Let's talk a little about how computers of any type are recognized on a network. If anyone doesn’t know, when connecting, any terminal is assigned two types of IP address or unique identifier: external and internal. It is worth noting that the internal address is not unique. And the current IP address – yes. There are no two machines in the world with the same IP. This allows you to identify any device, be it a mobile device or a computer terminal. A special protocol is responsible for this. The most common and widely used protocol at the moment is IPv4. Practice shows that this protocol has already become obsolete, since it is unable to provide unique addresses due to the increased number of client devices. Just look at mobile technology; over the past ten years, the number of gadgets used has increased so much that almost every second inhabitant of the earth has a mobile phone at their disposal.

IPv6 protocol

The network architecture began to gradually change. The IPv4 protocol version has been replaced by IPv6. So far it has not yet become particularly widespread, but the future of this protocol not far off. Soon, almost all Internet providers that provide access to communication services will gradually switch to this protocol. Judge for yourself, using this protocol with a 128-bit address, you can reserve much more addresses than when using the fourth version.

Dedicated Servers

Let's look at what dedicated servers are. In this case, the designation already speaks for itself. They are designed to perform specific tasks. This is a real virtual Internet server that completely belongs to the user who rents it. This is the meaning of hosting, when the owner of the podcast of the main resource can post any information on the allocated space. In this case, it is not the tenant who is responsible for security, but the one who rents out the server space. There are many examples of such servers. Here you and personal pages, and file sharing services, and games, and mail.

Local networks

Local networks, or as they are often called “local networks,” are created to combine a limited number of terminals into one. As should already be clear, the architecture of the local network in terms of connection can be either VPN-type access or a wired connection. In both cases, you will need to be connected to the main administration server. In this case, network services can operate in dual mode: with manual entry of parameters and automatic identification, which consists of assigning an address to each machine. In principle, local networks there is one distinctive feature, which consists only in the fact that any terminal requires registration and a central server. Access to “shared” information can be either limited or complete. In this case, everything will depend on the settings. However, if you don't even look cloud services, then in essence they represent virtual network, in which users, after undergoing an authentication procedure, receive rights to access certain information, edit and download files. Sometimes this even allows for simultaneous modification of the file contents in real time.

Network architecture: historical background

Let's finally get to the large network in the world. This is, of course, the Internet. ARPANET is considered to be the prototype of the Internet. This was the name of communication, which was developed in 1969 in the United States exclusively for military purposes. However, at that time the connection was tested only between two nodes. Over time, network connections via cable were established even with terminals located in the UK. Later, identification based on TCP/IP protocols and a system for assigning domain names appeared. It was then that what is now called the Internet arose. In general, it is believed that there is no single server on the Internet on which all information could be stored. Doesn't even exist today disk drives such a capacity. Information is distributed among hundreds of thousands of individual servers of various types. In other words, the Internet can be classified equally as a peer-to-peer and a hybrid network. At the same time, on a separate machine you can create your own Internet server, which makes it possible not only to manage network parameters and save necessary information, but also provide access to it to other users. The most simple example is Wi-Fi distribution.

Basic settings and options

If we talk about parameters and settings, then everything is quite simple. Manually entering network IPs, proxies and DNS servers has not been used for a long time. Instead, providers provide automatic PC recognition services or mobile device online. In operating systems of the family Windows access These settings are accessed through the network properties with the selection of IPv4 protocol parameters. The settings indicate automatic receipt of addresses. This saves the user from manually entering data. However, in some cases, especially when setting up RDP clients or organizing access to some specific services, manual entry of this is mandatory.

Conclusion

As you can see for yourself, understanding what network architecture is is not particularly difficult. IN this review Only the main aspects of organizing the work of networks were considered. This is quite enough to explain to an untrained user how the network works on the fingers. In reality, everything is a little more complicated. In this article, we did not touch on the concepts of proxy servers, DNS, WINS, DHCP, etc. Also, software-related issues were not addressed here. Even the information presented will be quite sufficient to understand the basic principles of the functioning of networks of any type and structure.

Network layout and components. "Server" and "workstation"

A computer network (CN) is a complex complex of interconnected and coordinated functioning hardware and software components. The hardware components of a local network are computers and various communications equipment (cable systems, hubs, etc.). Software components VS are operating systems (OS) and network applications.

Network layout is the process of arranging hardware components to achieve the desired result.

Depending on how functions are distributed between computers on the network, they can act in three different roles:

1. A computer that deals exclusively with servicing requests from other computers plays the role of a dedicated network server (Fig. 1.4).

2. A computer that makes requests to the resources of another machine plays the role of a client node (Fig. 1.5).

3. A computer that combines the functions of a client and a server is a peer-to-peer node (Fig. 1.6).

Rice. 1.4. Computer - dedicated network server

Rice. 1.5. Computer as a client node

Obviously, the network cannot consist only of client or only server nodes.

The network can be built according to one of three schemes:

· peer-to-peer network – peer-to-peer network;

· network based on clients and servers – a network with dedicated servers;

· a network that includes nodes of all types – a hybrid network.

Each of these schemes has its own advantages and disadvantages that determine their areas of application.

Rice. 1.6. Computer - peer

In peer-to-peer networks, the same PC can be both a server and a client, including a client of its client. In hierarchical networks, shared resources are stored only on the server; the server itself can only be a client of another server at a higher hierarchy level.

Moreover, each of the servers can be implemented either on a separate computer or in a small LAN, or be combined on one computer with some other server.

There are also combined networks that combine best qualities peer-to-peer and server-based networks. Many administrators believe that such a network best satisfies their needs.

Network architecture defines the main elements of the network, characterizes its overall logical organization, technical support,software, describes coding methods. The architecture also defines the operating principles and user interface.

Terminal-host architecture;

Peer-to-peer architecture;

Client-server architecture.

Terminal-host architecture

Terminal-host computer architecture is a concept of an information network in which all data processing is carried out by one or a group of host computers.

The architecture under consideration involves two types of equipment:

The main computer where network management, data storage and processing is carried out;

Terminals designed to transmit commands to the host computer to organize sessions and perform tasks, enter data to complete tasks and obtain results.

The main computer interacts with the terminals through the MTD, as shown in Fig. 1.7.

A classic example of a host network architecture is System Network Architecture (SNA).

Rice. 1.7. Terminal-host architecture

Peer-to-peer architecture

Peer-to-peer architecture is the concept of an information network in which its resources are dispersed across all systems. This architecture is characterized by the fact that all systems in it have equal rights.

Peer-to-peer networks include small networks where any workstation can simultaneously perform the functions of a file server and workstation. On peer-to-peer LANs disk space and files on any computer can be shared. For a resource to become shared, it must be shared using services remote access network peer-to-peer operating systems. Depending on how data protection is set up, other users will be able to use the files immediately after they are created. Peer-to-peer LANs are only good enough for small workgroups.

Peer-to-peer LANs are the easiest and cheapest type of network to install. By connecting computers, users can share resources and information.

Peer-to-peer networks have the following advantages:

They are easy to install and configure;

Individual PCs are not dependent on a dedicated server;

Users are able to control their resources;

Low cost and easy operation;

Minimum equipment and software;

No need for an administrator;

Well suited for networks with no more than ten users.

The problem with peer-to-peer architecture is when computers go offline. In these cases, the types of services they provided disappear from the network. Network security can only be applied to one resource at a time, and the user must remember as many passwords as network resources. When gaining access to a shared resource, a drop in computer performance is felt. A significant disadvantage of peer-to-peer networks is the lack of centralized administration.

The use of a peer-to-peer architecture does not preclude the use of a terminal-host or client-server architecture on the same network.

Client-server architecture

Client-server architecture is a concept information network, in which the bulk of its resources are concentrated in servers serving their clients (Fig. 1.8). The architecture in question defines two types of components: servers and clients.

A server is an object that provides services to other network objects based on their requests. Service is the process of serving customers.

The server works on orders from clients and manages the execution of their jobs. After each job is completed, the server sends the results to the client that sent the job.

The service function in the client-server architecture is described by a set of application programs, in accordance with which various application processes are performed.

Rice. 1.8. Client-server architecture

The process that causes service function through certain operations is called a client. This could be a program or a user. In Fig. 1.9 shows a list of services in the client-server architecture.

Clients are workstations that use server resources and provide convenient interfaces user. User interfaces (Fig. 1.9) are procedures for user interaction with a system or network.

In networks with dedicated file server A server network operating system is installed on a dedicated stand-alone PC. This PC becomes a server. The software installed on the workstation allows it to exchange data with the server. The most common network operating systems are:

NetWare from Novel;

Windows NT from Microsoft;

AT UNIX

In addition to the network operating system networking is required application programs, realizing the benefits provided by the network.

Rice. 1.9. Client-server model

The range of tasks that servers perform in hierarchical networks is diverse and complex. To adapt to the increasing needs of users, LAN servers have become specialized. For example, in the Windows NT Server operating system there are different types of servers:

1. File servers and print servers. They control user access to files and printers. So, for example, to work with text document The first thing you do is run a word processor on your computer (PC). Next is the required document word processor, stored on the file server, is loaded into the PC's memory, and thus you can work with this document on the PC. In other words, a file server is designed to store files and data.

2. Application servers (including a database server (DB), WEB server). They run application parts of client-server applications (programs). These servers are fundamentally different from file servers in that when working with a file server required file or the entire data is copied to the requesting PC, and when working with the application server, only the request results are sent to the PC. For example, upon request, you can only get a list of employees born in September, without downloading the entire personnel database to your PC.

3. Mail servers control transmission emails between network users.

4. Fax servers manage the flow of incoming and outgoing fax messages through one or more fax modems.

5. Communication servers manage the flow of data and mail messages between a given LAN and other networks or remote users via modem and telephone line. They also provide access to the Internet.

6. Directory services server is designed to search, store and protect information on the network. Windows NT Server unites PCs into logical domain groups, the security system of which gives users various access rights to any network resource.

The client is the initiator and uses email or other server services. In this process, the client requests a service, establishes a session, gets the results it wants, and reports completion.

Server-based networks have best characteristics and increased reliability. The server owns the main network resources that are accessed by other workstations.

In modern client-server architecture, there are four groups of objects: clients, servers, data and network services. Clients are located in systems at user workstations. Data is mainly stored on servers. Network Services are shared servers and data. In addition, services manage data processing procedures.

Client-server architecture networks have the following advantages:

Allows you to organize networks with a large number workstations;

Provide centralized management of user accounts, security and access, which simplifies network administration;

Efficient access to network resources;

The user needs one password to log into the network and to gain access to all resources to which user rights apply.

Along with the benefits of the network client- server architecture They also have a number of disadvantages:

A server malfunction can render the network inoperable;

Require qualified personnel for administration;

They have a higher cost of networks and network equipment.

Selecting a Network Architecture

The choice of network architecture depends on the purpose of the network, the number of workstations and the activities performed on it.

You should choose a peer-to-peer network if:

The number of users does not exceed ten;

All cars are close to each other;

There are small financial possibilities;

There is no need for a specialized server such as a database server, fax server or any other;

There is no possibility or need for centralized administration.

You should choose a client-server network if:

The number of users exceeds ten;

Requires centralized management, security, resource management, or backup;

A specialized server is required;

Requires access to the global network;

It is required to share resources at the user level.

The most common architectures:

· Ethernet ether– broadcast) – broadcast network. This means that all stations on the network can receive all messages. Topology – linear or star-shaped. Data transfer speed 10 or 100 Mbit/s.

· Arcnet ( Attached Resource Computer Network– computer network of connected resources) – broadcast network. Physical topology is a tree. Data transfer rate 2.5 Mbit/s.

· Token Ring (relay ring network, token passing network) – a ring network in which the principle of data transmission is based on the fact that each ring node waits for the arrival of some short unique sequence of bits – marker– from an adjacent previous node. The arrival of the token indicates that it is possible to transmit a message from this node further along the flow. Data transfer speed 4 or 16 Mbit/s.

FDDI ( Fiber Distributed Data Interface) – network architecture high speed transmission data on fiber optic lines. Transfer speed – 100 Mbit/sec. Topology – double ring or mixed (including star or tree subnets). The maximum number of stations in the network is 1000. Very high cost of equipment.

ATM ( Asynchronous Transfer Mode) is a promising, expensive architecture that provides the transmission of digital data, video information and voice over the same lines. Transfer speed up to 2.5 Gbps. Optical communication lines.

The main task solved when creating computer networks is to ensure compatibility of equipment in terms of electrical and mechanical characteristics and ensure compatibility information support(programs and data) by coding system and data format. The solution to this problem belongs to the field of standardization and is based on the so-called model OSI (interoperability model) open systems– Model of Open System Interconnections). OSI model was created on the basis of technical proposals from the International Standards Institute ISO (International Standards Organization).

According to the model OSI architecture computer networks should be considered at different levels (the total number of levels is up to seven). The highest level is applied. At this level the user interacts with computing system. The lowest level is physical. It ensures the exchange of signals between devices. Data exchange in communication systems occurs by moving it from the upper level to the lower one, then transporting it and, finally, playing it back on the client's computer as a result of moving from the lower level to the upper one.

Rice. 8. Control levels and protocols of the OSI model

To ensure the necessary compatibility at each of the seven possible levels of architecture computer network There are special standards called protocols. They determine the nature of the hardware interaction of network components (hardware protocols) and the nature of the interaction of programs and data (software protocols). Physically, protocol support functions are performed by hardware devices (interfaces) and software (protocol support programs). Programs that support protocols are also called protocols.

Each level of the architecture is divided into two parts:

· specification of services;

· protocol specification.

A service specification defines what a layer does, and a protocol specification defines how it does it, and each particular layer may have more than one protocol.

Let's look at the functions performed by each layer of software:

1. The physical layer makes connections to the physical channel, so disconnects from the channel, and manages the channel. The data transfer rate and network topology are determined.

2. Data Link Layer adds auxiliary characters to the transmitted arrays of information and controls the correctness of the transmitted data. Here the transmitted information is divided into several packets or frames. Each packet contains source and destination addresses, as well as error detection.

3. The network layer determines the route for transmitting information between networks, provides error handling, and also manages data flows. The main task of the network layer is data routing (data transfer between networks).

4. The transport layer connects lower layers (physical, data link, network) with upper levels, which are being implemented software. This level separates the means of generating data on the network from the means of transmitting it. Here the information is divided according to a certain length and the destination address is specified.

5. The session layer manages communication sessions between two interacting users, determines the beginning and end of a communication session, the time, duration and mode of a communication session, synchronization points for intermediate control and recovery during data transfer; restores the connection after errors during a communication session without data loss.

6. Representative – controls the presentation of data in the form required by the user program, performs data compression and decompression. The task of this level is to convert data when transmitting information into a format that is used in information system. When receiving data this level data representation performs the reverse transformation.

7. Application layer interacts with application network programs, serving files, and also performs computational, information retrieval work, logical transformations of information, transmission of mail messages, etc. The main task of this level is to provide a convenient interface for the user.

At different levels, different units of information are exchanged: bits, frames, packets, session messages, user messages.

This course will cover three types of architectures:

• 
  architecture terminal - main computer;
• 
  peer-to-peer architecture;
• 
  client-server architecture.

Architecture terminal - main computer

Rice. 1.1 Architecture terminal - host computer

The architecture under consideration involves two types of equipment:

• 
  The main computer where network management, data storage and processing is carried out.
• 
  Terminals designed to transmit commands to the host computer to organize sessions and perform tasks, enter data to complete tasks and obtain results.

A classic example of a network architecture with host computers is the systems network architecture ( System Network Architecture - SNA).

Peer-to-peer architecture

TO peer-to-peer networks include small networks where any workstation can simultaneously perform the functions of a file server and a workstation. IN peer-to-peer LAN Disk space and files on any computer can be shared. For a resource to become shared, it must be shared using remote access services of networked peer-to-peer operating systems. Depending on how data protection is set up, other users will be able to use the files immediately after they are created. Peer-to-peer LAN Only good enough for small workgroups.

Rice. 1.2 Peer-to-peer architecture

Peer-to-peer LAN are the easiest and cheapest type of network to install. They require on the computer, except network card and network media, operating system only Windows 95 or Windows for Workgroups. By connecting computers, users can share resources and information.

Peer-to-peer networks have the following advantages:

• 
  they are easy to install and configure;
• 
  individual PCs do not depend on a dedicated server;
• 
  users are able to control their resources;
• 
  low cost and easy operation;
• 
  minimum hardware and software;
• 
  no need for an administrator;
• 
  well suited for networks with no more than ten users.

The use of a peer-to-peer architecture does not exclude the use of a terminal-host architecture or a client-server architecture on the same network.

Client-server architecture

Server - is an object that provides service to other network objects upon their requests. Service is a customer service process.

Rice. 1.3 Client-server architecture

The server works on orders from clients and manages the execution of their jobs. After each job is completed, the server sends the results to the client that sent the job.

The service function in the client-server architecture is described by a set of application programs, in accordance with which various application processes are performed.

A process that calls a service function using certain operations is called client. This could be a program or a user. In Fig. 1.6 shows a list of services in the client-server architecture.

Clients are workstations that use server resources and provide convenient user interfaces. User Interfaces These are the procedures for how a user interacts with a system or network.

The client is the initiator and uses email or other server services. In this process, the client requests a service, establishes a session, gets the results it wants, and reports completion.

Rice. 1.4 Client-server model

IN networks with a dedicated file server on a dedicated standalone PC a server network operating system is installed. This PC becomes server. Software ( BY), installed on a workstation, allows it to exchange data with the server. The most common network operating systems are:

• 
  NetWare from Novel;
• 
  Windows NT from Microsoft;
• 
  AT UNIX
• 
  Linux.

Server-based networks have better characteristics and increased reliability. Server owns the main resources of the network, which other workstations access.

In modern client-server architecture, four groups of objects are distinguished: clients, servers, data and network services. Clients are located in systems at user workstations. Data is mainly stored on servers. Network services are shared servers and data. In addition, services manage data processing procedures.

Client-server architecture networks have the following advantages:

• 
  allow you to organize networks with a large number of workstations;
• 
  provide centralized management of user accounts, security and access, which simplifies network administration;
• 
  efficient access to network resources;
• 
  the user needs one password to log on to the network and to gain access to all resources to which the user's rights apply.

• 
  a server malfunction can make the network inoperable, at least loss of network resources;
• 
  require qualified personnel for administration;
• 
  have a higher cost of networks and network equipment.

Selecting a Network Architecture

You should choose a peer-to-peer network if:

• 
  the number of users does not exceed ten;
• 
  all cars are close to each other;
• 
  there are small financial opportunities;
• 
  there is no need for a specialized server such as a database server, fax server or any other;
• 
  there is no possibility or need for centralized administration.

• 
  the number of users exceeds ten;
• 
  requires centralized management, security, resource management, or backup;
• 
  a specialized server is required;
• 
  need access to the global network;
• 
  it is required to share resources at the user level.

Hello, dear site visitors! Networks must support a wide range of applications and services, and must operate through large quantities various types physical infrastructures. The term network architecture in this context refers to both technologies that support the infrastructure and software services and the protocols that carry messages over this architecture. As in general, evolve, we find four main characteristics underlying architectures that must be implemented to meet user expectations: , , and .

A scalable network can quickly expand to support new users and applications without impacting the performance of the service provided to existing users. Thousands of new users and service providers connect every week. The ability of the network to support these new relationships depends on the layered hierarchical design of the underlying physical infrastructure and logical architecture of the network. The operation of each layer allows users or service providers to connect to the Internet without disrupting functionality the whole network. Technological developments continually improve the messaging capabilities and performance of physical infrastructure components at every layer. These developments, together with new ways of defining and localizing individual users V unified network allow you to develop in accordance with the needs and requirements of users.

Quality of Service (from the English Quality of Service or QoS)

On present moment provides an acceptable level of fault tolerance and scalability for users. But new applications made available to users through interconnected networks create higher expectations for the quality of services provided. Voice communication and video transmissions require a level of consistent quality and uninterrupted transmission that was not required for traditional computer applications. The quality of these services is measured in comparison to listening/viewing the same audio or video presentations directly (not through ). Traditional voice and video networks are designed to support one type of transmission, and therefore they are able to provide an acceptable level of quality. New requirements to support this quality of service in a converged network are changing the way network architectures are designed and implemented.

Evolved from a tightly controlled, unified network of educational and government organizations into a globally available network, which has become a means of personal and business communications. As a result, network security requirements have changed. Expectations of security and privacy associated with the use of internetworks for confidential and business communications classified information, exceed the capabilities that the current network architecture. Rapid expansion into areas of communications that have not been served by traditional communications increases the need to build security into the network architecture. Because of this, a lot of research and development efforts are being put into this area, and many tools and procedures are being implemented to plug the inherent security holes in the network architecture.

Thank you for your attention!.