Development and design of a local computer network for an organization with two offices and a warehouse. Enterprise office LAN project Design of an organization's local network

Selecting network size and structure

The size of the network in this case refers to both the number of computers connected to the network and the distances between them. You need to clearly understand how many computers (minimum and maximum) need to be connected to the network. At the same time, it is necessary to leave room for further growth in the number of computers on the network, at least by 20–50 percent.

The required length of network communication lines also plays an important role in network design. For example, if the distances are very long, it may be necessary to use expensive equipment. In addition, as the distance increases, the importance of protecting communication lines from external electromagnetic interference increases sharply.

Network structure refers to the way the network is divided into parts (segments), as well as the way these segments are connected to each other. An enterprise network may include workgroups of computers, departmental networks, core networks, and means of communication with other networks.

Equipment selection

When choosing network equipment, many factors must be taken into account, in particular:

Level of equipment standardization and its compatibility with the most common software;

Speed of information transfer and the possibility of its further increase;

Possible network topologies and their combinations (bus, passive star, passive tree);

Network exchange control method (CSMA/CD, full duplex or token method);

Permitted types of network cable, its maximum length, immunity to interference;

Cost and technical characteristics of specific hardware (network adapters, transceivers, repeaters, hubs, switches).

Another important task is the choice of computers. If for workstations or non-dedicated servers they usually use those computers that are already available in the enterprise, then it is advisable to purchase a dedicated server specifically for the network.

Selecting Network Software

When choosing network software (software), you must first take into account the following factors:

What type of network does the networking software support: peer-to-peer, server-based, or both?

Maximum number of users (it is better to take a reserve of at least 20%);

Number of servers and their possible types;

Compatible with different operating systems and computers, as well as with other network tools;

Level of software performance in various operating modes;

Degree of operational reliability, permitted access modes and degree of data protection;

What network services are supported;

And, perhaps most importantly, the cost of software, its operation and modernization.

When choosing between Microsoft products and other network operating systems (for example, Novell), you must keep in mind that traditionally the advantages of network products (for example, NetWare network operating systems) are:

More advanced network OS architecture;

Universality and functional completeness of software;

Greater performance with this type of equipment;

Simplified network administration;

Significantly higher protection against viruses and unauthorized access;

Supports different types of users on different computer platforms.

The main advantage of Microsoft products is considered to be better compatibility with users based on the Microsoft Windows OS.

Selection, taking into account the cost of various means for building a computer network. Cable system design, network optimization and debugging.

Choice based on cost

First of all, it is necessary to determine the possible directions of financial costs (at this stage of design, the necessary prerequisites for solving this problem already exist):

Additional computers and upgrades to existing computers. Optional cost direction: if there is a sufficient number and quality of existing computers, their upgrade is not required (or required in a minimal amount - for example, to install more modern network cards); In a peer-to-peer network, a special file server is also not needed (although it is desirable).

Network hardware (cables and everything that is needed to organize a cable system, network printers, active network devices - repeaters, hubs, routers, etc.).

Network software, first of all, a network OS for the required number of workstations (with a reserve).

Payment for the work of invited specialists when organizing a cable system, installing and configuring a network OS, during periodic maintenance and urgent repairs. Optional Cost Direction: For smaller networks, many of these jobs can and should be handled by a full-time network administrator (perhaps with the help of others within the enterprise).

Cabling system design

When choosing a cable, you must first take into account the required length, as well as protection from external interference and the level of its own emissions. If the network is long and there is a need to ensure the secrecy of transmitted data or there is a high level of interference in the room, a fiber optic cable is indispensable. It should be noted that the use of fiber optics instead of electrical cables, even under fairly comfortable conditions, can significantly (by 10-50 percent) increase network performance by reducing the share of distorted information packets.

In designing cable systems for local networks, a great deal of experience has been accumulated, on the basis of which general recommendations for organizing such systems can be formulated. Moreover, there are standards under the general name “structured cabling systems (SCS)”, which are especially relevant for newly created or reconstructed relatively large local networks at the enterprise level. They are voluminous documents that describe and regulate in detail the process of creating cable connections for local networks.

Local area networksFastEthernetAndGigabit Ethernet

Fast Ethernet is the general name for a set of standards for data transmission in computer networks using Ethernet technology at speeds of up to 100 Mbit/s, as opposed to the original 10 Mbit/s.

Fast Ethernet (IEEE 802.3u)

Fast Ethernet technology is an evolutionary development of classic Ethernet technology. Its main advantages are:

increasing the capacity of network segments up to 100 Mb/s;
saving Ethernet random access method;
maintaining a star-shaped network topology and supporting traditional data transmission media - twisted pair and fiber optic cable.

These properties allow for a gradual transition from 10Base-T networks - the most popular version of Ethernet today - to high-speed networks that maintain significant continuity with familiar technology: Fast Ethernet does not require radical retraining of personnel and replacement of equipment in all network nodes. The official 100Base-T (802.3u) standard established three different physical layer specifications (in terms of the seven-layer OSI model) to support the following types of cabling systems:

100Base-TX for two-pair cable on UTP Category 5 unshielded twisted pair, or STP Type 1 shielded twisted pair;

100Base-T4 for four-pair UTP Category 3, 4, or 5 UTP cable;

100Base-FX for multimode fiber optic cable.

The Gigabit Ethernet network is a natural, evolutionary development of the concept inherent in the standard Ethernet network. Of course, it inherits all the shortcomings of its direct predecessors, for example, non-guaranteed network access time. However, the huge bandwidth makes it quite difficult to load the network to levels where this factor becomes decisive. But maintaining continuity allows you to quite simply connect Ethernet, Fast Ethernet and Gigabit Ethernet segments into a network, and, most importantly, move to new speeds gradually, introducing gigabit segments only in the busiest sections of the network. (Besides, such high bandwidth is not really needed everywhere.) If we talk about competing gigabit networks, then their use may require a complete replacement of network equipment, which will immediately lead to large costs.

The Gigabit Ethernet network retains the same CSMA/CD access method, which has proven itself in previous versions, and uses the same packet (frame) formats and the same sizes. No protocol conversion is required at the junctions with Ethernet and Fast Ethernet segments. The only thing that is needed is the coordination of exchange rates, so the main area of application of Gigabit Ethernet will be primarily the connection of Ethernet and Fast Ethernet hubs to each other.

With the advent of ultra-fast servers and the proliferation of the most advanced high-end personal computers, the benefits of Gigabit Ethernet are becoming increasingly clear. Thus, the 64-bit PCI system bus, already a de facto standard, fully achieves the data transfer speed required for such a network.

Work on creating a Gigabit Ethernet network has been ongoing since 1995. In 1998, a standard called IEEE 802.3z (1000BASE-SX, 1000BASE-LX and 1000BASE-CX) was adopted. The development is carried out by a specially created alliance (Gigabit EthernetAlliance), which, in particular, includes such a well-known network equipment company as 3Com. In 1999, the IEEE 802.3ab (1000BASE-T) standard was adopted.

The nomenclature of Gigabit Ethernet network segments currently includes the following types:

1000BASE-SX is a segment on a multimode fiber optic cable with a light signal wavelength of 850 nm (up to 500 meters long). Laser transmitters are used.

1000BASE-LX is a segment on multimode (up to 500 meters long) and single-mode (up to 2000 meters long) fiber optic cable with a light signal wavelength of 1300 nm. Laser transmitters are used.

1000BASE-CX – segment on shielded twisted pair (up to 25 meters long).

1000BASE-T (IEEE 802.3ab standard) – a segment on quad unshielded twisted pair category 5 (up to 100 meters long). 5-level coding (PAM-5) is used, and in full duplex mode transmission is carried out on each pair in two directions.

Especially for the Gigabit Ethernet network, a method for encoding transmitted information 8V/10V has been proposed, based on the same principle as the 4V/5V code of the FDDI network (except for 1000BASE-T). Thus, the eight bits of information that need to be transmitted are associated with 10 bits transmitted over the network. This code allows you to maintain self-synchronization, easily detect the carrier (the fact of transmission), but does not require doubling the bandwidth, as is the case with the Manchester code.

To increase the 512-bit Ethernet network interval corresponding to the minimum packet length (51.2 μs in an Ethernet network and 5.12 μs in a Fast Ethernet network), special methods have been developed. In particular, the minimum packet length has been increased to 512 bytes (4096 bits). Otherwise, the 0.512 µs time interval would overly limit the length limit of a Gigabit Ethernet network. All packets less than 512 bytes in length are expanded to 512 bytes. The extension field is inserted into the packet after the checksum field. This requires additional packet processing, but the maximum allowable network size becomes 8 times larger than without such measures.

In addition, Gigabit Ethernet provides the possibility of block packet transmission (frame bursting). In this case, a subscriber who has received the right to transmit and has several packets to transmit can transmit not one, but several packets, sequentially, and addressed to different recipient subscribers. Additional transmitted packets can only be short, and the total length of all packets in a block must not exceed 8192 bytes. This solution allows you to reduce the number of network takeovers and reduce the number of collisions. When using block mode, only the first packet of the block is expanded to 512 bytes in order to check if there are any collisions on the network. Other packets up to 512 bytes may not be expanded.

Transmission on a Gigabit Ethernet network occurs in both half-duplex mode (while maintaining the CSMA/CD access method) and in a faster full-duplex mode (similar to the previous Fast Ethernet network). Full-duplex mode, which imposes no restrictions on network length (other than limitations due to signal attenuation in the cable) and ensures no collisions, is expected to become the main one for Gigabit Ethernet in the future.

The design object is the local computer network of the organization. This network must ensure the transport of information within the organization and provide the ability to interact with the global Internet. The organization for which the local network is being designed is an enterprise whose main activity is the production of high-quality furniture.

Network topology and network equipment

Network topology

When building an organization's LAN, we will use a tree structure based on the star topology. This is one of the most common topologies because it is easy to maintain.

Advantages of the topology:

· failure of one workstation does not affect the operation of the entire network as a whole;
· good network scalability;
· easy troubleshooting and network breaks;
· high network performance (subject to proper design);
· flexible administration options.

Disadvantages of topology:

· failure of the central hub will result in the inoperability of the network (or network segment) as a whole;
· network installation often requires more cable than most other topologies;
· the finite number of workstations in a network (or network segment) is limited by the number of ports in the central hub.

This topology was chosen due to the fact that it is the fastest. From a reliability point of view, it is not the best solution, since the failure of the central node leads to the shutdown of the entire network, but at the same time it is easier to find the fault.

Subscribers of each network segment will be connected to the corresponding switch (Switch). And these segments will be connected into a single network by a managed switch - the central element of the network.

The following network equipment is required:

1. Network switches or switches(Switch) - 8 pcs.-- a device designed to connect several computer network nodes within one segment.
2. Servers(server) - 1 PC.-- hardware dedicated and/or specialized for running service software on it without direct human intervention.
3. Printers (including multifunction devices)(Printer) - 5 pieces.- a device for printing digital information onto a solid medium, usually paper. Refers to computer terminal devices.
4. DVB PC card 1 pc. - This is a computer board that is designed to receive a signal from a satellite and then decrypt it.
5. Satellite antenna-1 PC.- this is the most important component of satellite Internet and satellite TV; the stability of the Internet connection, and the quality and quantity of satellite TV channels will depend on it.
6. Converter - 1 pc.- a program with which you can convert files from one format to another.

Transmission medium:

A transmission medium is a physical medium through which it is possible to distribute information signals in the form of electrical, light, etc. impulses.

To connect a PC into a single LAN, you will need a UTP5e “twisted pair” cable, which is one of the most common types of cable nowadays. It consists of several pairs of copper wires covered with a plastic sheath. The wires that make up each pair are twisted around each other, which provides protection against mutual interference. Cables of this type are divided into two classes - “Shielded twisted pair” and “Unshielded twisted pair”. The difference between these classes is that shielded twisted pair cable is more protected from external electromagnetic interference due to the presence of an additional shield of copper mesh and/or aluminum foil surrounding the cable wires. Twisted pair networks, depending on the cable category, provide transmission speeds from 10 Mbit/s - 1 Gbit/s. The length of the cable segment cannot exceed 100 m (up to 100 Mbit/s).

Table 1. Number of equipment in the network

Equipment	Quantity
Switches
Switch D-Link Switch 10port (8UTP 10/100/1000Mbps + 2Combo 1000Base-T/SFP)
Switch D-Link Switch 16 port (16UTP 10/100Mbps) (swich1,2,3,4,5,6,7)
Server sS7000B/pro2U (SX20H2Mi): Xeon E5-2650/ 16 GB/ 2 x 1 TB SATA RAID
PC (2 configurations)
Epson AcuLaser M2400DN (A4, 35 ppm, 1200dpi, USB2.0/LPT, network, duplex printing)
Satellite DVB card TeVii S 470 PCI-E (DVB-S2)
Satellite antenna LANS-7.5 Direct focus parabolic mesh antenna with azimuth fixed suspension AZ/EL 2.30m F/D=0.375
Converter MultiCo< EC-202C20-BB>10/100Base-TX to 100Base-FX converter (1UTP, 1SC)

Cable system calculation:

To calculate the cost of cables, let's assume that the average distance between computers in a department and the corresponding switch is 10 meters, then approximately 850 m of UTP 5e cable will be needed.

To cover the distance from the switches to the central managed switch (+ connect the manager) you will need 350 m of UTP 5e cable. network local cable internet

Let's increase the cost of UTP 5e cable by 10% (for waste and installation defects) and get approximately 1350 m.

In total, you will need 100 pieces of twisted pair cable, which will require 200 RJ-45 connectors. Taking into account defects - 220.

Moscow State Mining University

Department of Automated Control Systems

Course project

in the discipline "Computer networks and telecommunications"

on the topic: “Design of a local area network”

Completed:

Art. gr. AS-1-06

Yuryeva Ya.G.

Checked:

Prof., Doctor of Technical Sciences Shek V.M.

Moscow 2009

Introduction

1 Design task

2 Description of the local area network

3 Network topology

4 Local network diagram

5 OSI reference model

6 Justification for choosing a local network deployment technology

7 Network protocols

8 Hardware and software

9 Calculation of network characteristics

Bibliography

A local area network (LAN) is a communications system that connects computers and peripheral equipment in a limited area, usually no more than several buildings or one enterprise. Currently, a LAN has become an integral attribute in any computing systems with more than 1 computer.

The main advantages provided by a local network are the ability to collaborate and quickly exchange data, centralized data storage, shared access to shared resources such as printers, the Internet and others.

Another important function of a local network is the creation of fault-tolerant systems that continue to function (albeit not fully) if some of their elements fail. In a LAN, fault tolerance is ensured through redundancy and duplication; as well as flexibility in the operation of individual parts (computers) included in the network.

The ultimate goal of creating a local network in an enterprise or organization is to increase the efficiency of the computing system as a whole.

Building a reliable LAN that meets your performance requirements and has the lowest cost requires starting with a plan. In the plan, the network is divided into segments, and a suitable topology and hardware are selected.

The bus topology is often called a linear bus. This topology is one of the simplest and most widespread topologies. It uses a single cable, called a backbone or segment, along which all computers on the network are connected.

In a network with a “bus” topology (Fig. 1.), computers address data to a specific computer, transmitting it over a cable in the form of electrical signals.

Fig.1. Bus topology

Data in the form of electrical signals is transmitted to all computers on the network; however, only the one whose address matches the recipient address encrypted in these signals receives information. Moreover, at any given time, only one computer can transmit.

Since data is transmitted to the network by only one computer, its performance depends on the number of computers connected to the bus. The more there are, i.e. The more computers waiting to transfer data, the slower the network.

However, it is impossible to derive a direct relationship between network bandwidth and the number of computers in it. Since, in addition to the number of computers, network performance is influenced by many factors, including:

· hardware characteristics of computers on the network;

· the frequency with which computers transmit data;

· type of running network applications;

· type of network cable;

· distance between computers on the network.

The bus is a passive topology. This means that computers only “listen” to data transmitted over the network, but do not move it from sender to recipient. Therefore, if one of the computers fails, it will not affect the operation of the others. In active topologies, computers regenerate signals and transmit them across the network.

Signal reflection

Data, or electrical signals, travel throughout the network - from one end of the cable to the other. If no special action is taken, the signal reaching the end of the cable will be reflected and will not allow other computers to transmit. Therefore, after the data reaches the destination, the electrical signals must be extinguished.

Terminator

To prevent electrical signals from being reflected, terminators are installed at each end of the cable to absorb these signals. All ends of the network cable must be connected to something, such as a computer or a barrel connector - to increase the cable length. A terminator must be connected to any free - unconnected - end of the cable to prevent electrical signals from being reflected.

Information technology infrastructure is mainly based on a local area network, therefore, depending on how well it is designed and created local area network (LAN), depend on the quality indicators of the functioning of the infrastructure as a whole.

Specialists of Mosproekt-Engineering LLC are always ready to design a LAN for you your office, enterprise, or other objects, which in turn will allow you to combine workplaces, office equipment, various installations and elements, including computers and microprocessors, into one whole system.

The process of creating a LAN includes three stages:

LAN design taking into account the necessary regulatory documents, coordination of design documentation with the customer and with various authorities (if necessary);
- assembly, installation and integration of LAN network elements into a single whole;
- commissioning and transfer of LAN for use to the customer.
When preparing design documentation, design engineers of Mosproekt-Engineering LLC take into account the possibility of using components from various world-famous manufacturers, such as Hyperline, Krone and other manufacturers, in the designed LAN network.
Specialists Mosproekt-Engineering LLC will promptly carry out all the necessary preparatory (pre-design) work, namely, carry out a survey of the premises, and, if necessary, a survey of the adjacent territory, engineering surveys, draw up plans for the location of workplaces, office equipment, servers, various network elements and other devices.
If it is necessary to combine the LAN of individual divisions, branches, buildings into a single territorial distributed network, the specialists of Mosproekt-Engineering LLC are ready to offer you this type of project, namely, the design of territorial distribution networks. When designing a LAN, at the initiative of the customer, our specialists will provide the ability to connect to the LAN on the principle of remote access to the equipment of specialists - freelancers working remotely; also, remote connection may be required for full-time employees who are away from the office, for example, on business trips, at various sites or in the field conditions. Remote access to the LAN is provided taking into account the cybersecurity standards established by the customer organization.

What is a LAN in the daily life of a business/office?

A LAN is a high-tech, “smart” communication system that combines into a single whole system personal computers, office equipment, servers, telephony, monitoring systems, security, accounting and access control, management systems, other systems and elements, including various processors, microprocessors, chips, devices, controllers, control panels, software. The purpose of a LAN in an enterprise, office, or other structures is to securely, quickly and synchronously transfer data of various types (text, graphics, sound, video, and others) between personal computers and servers and other elements interacting with the system. The LAN allows you to receive, process and display various types of information on PC screens from units, devices, controllers, control panels, sensors, sensors, and other equipment connected to the system, as well as control them by setting the necessary parameters. A LAN makes it possible to quickly and securely access and manage databases. A LAN is also an opportunity to create mail hosting on its basis, that is, corporate mail, relatively safe and controlled access for personnel to external network resources (the Internet).
There are a great many possibilities and advantages of a LAN, they can be continued to be listed for a long time, but in our opinion, we have told you the main points. However, it is necessary to understand that for the correct and uninterrupted operation of the LAN system, administration is necessary, and the larger the system, the more difficult it is to maintain. For this purpose, special software products are provided, for example, operating systems installed on servers. Such software products are produced by many world-famous companies, such as Microsoft, Apple and others. It is worth noting that in order to fully protect information, it is necessary to choose the right programs for protecting and monitoring the state of the LAN - specialists from Mosproekt-Engineering LLC will competently advise you on such issues.

LAN consists of many independent, separate systems, as well as subsystems, segments, modules and elements, let's call them for convenience - LAN units. So, LAN design represents the development of separate projects for each LAN unit, subsequently combined into a common project, according to the principle “from a private project to a general one.” We design many LAN units ourselves, for example, individual systems, subsystems, and in our projects we also provide for the need or possibility of using standard LAN units, that is, developments from various manufacturers with well-known names, we are talking about ready-made modules, servers, processors, microprocessors, controllers, control panels, various devices, components, and so on, including software. The specialists of Mosproekt-Engineering LLC will help you select ready-made LAN units from global manufacturers, or they will develop them independently, then carry out the general design based on what is chosen.

Upon completion of the design work for the LAN, the customer receives the following design documents, namely:

A diagram reflecting the interaction between LAN electronic computers and a software product
- a diagram reflecting a structured cabling system (SCS), in other words, a document that includes graphic information about the building’s telephone network and LAN laying along with equipment. Graphic information about the telephone installation of the building and the LAN itself is reflected in this document in the form of an application on the plan of the building (office, factory, store, etc.). Note that preparing the SCS diagram requires more labor compared to other work, and therefore is considered separately from other work.
Design work over the interaction between electronic computers LAN.
As a result, a diagram is drawn up that reflects the deployment of a LAN, in other words, a diagram on which symbols of computers and other equipment are applied, indicating the installed software product, as well as the information flows formed in this case.
Design work on LAN cable systems.
A documentation package is generated, which includes the documents necessary for designing a LAN in a particular building.
The name of the documents included in the package and their content must strictly comply with the regulations of GOST R 21.1703-2000.
The LAN project is drawn up strictly according to the recommendations emanating from GOST 21.101-97.
The lack of a project will simply not allow you to install a LAN if the network covers large areas, a large building, especially if we are talking about a group of buildings.

The most important sections of the LAN project are:

1. Diagram reflecting the structure of the LAN;
2. Working documentation (graphic) - diagrams, drawings, explications, etc.;
3. Equipment classification.
The diagram reflecting the structure of the LAN is intended mainly for general visualization of the communications system. Working documentation in the form of graphic documents is intended for the correct assembly of the LAN. Equipment classification is important for the formation of cost estimates, contracts (agreements), acts, technical specifications for installation work, other documents, as well as for the execution of contracts for the manufacture and supply of equipment, for the general assembly of a LAN.

DEVELOPMENT OF LAN PROJECTS IS THE INITIAL AND NECESSARY STAGE TO CREATE A RELIABLE SUPPORT FOR THE UNINTERRUPTED OPERATION OF THE INFRASTRUCTURE OF AN ENTERPRISE, OFFICE AND MANY OTHER OBJECTS.

The specialists of Mosproekt-Engineering LLC are always ready to offer you many versions of design solutions, taking into account your wishes, based on your financial capabilities, as well as the technical characteristics of your premises.

Moscow State Mining University

Department of Automated Control Systems

Course project

in the discipline "Computer networks and telecommunications"

on the topic: “Design of a local area network”

Completed:

Art. gr. AS-1-06

Yuryeva Ya.G.

Checked:

Prof., Doctor of Technical Sciences Shek V.M.

Moscow 2009

Introduction

1 Design task

2 Description of the local area network

3 Network topology

4 Local network diagram

5 OSI reference model

6 Justification for choosing a local network deployment technology

7 Network protocols

8 Hardware and software

9 Calculation of network characteristics

Bibliography

The ultimate goal of creating a local network in an enterprise or organization is to increase the efficiency of the computing system as a whole.

In a network with a “bus” topology (Fig. 1.), computers address data to a specific computer, transmitting it over a cable in the form of electrical signals.

Fig.1. Bus topology

· hardware characteristics of computers on the network;

· the frequency with which computers transmit data;

· type of running network applications;

· type of network cable;

· distance between computers on the network.

Signal reflection

Terminator

Network integrity violation

A network cable breaks when it is physically broken or one of its ends is disconnected. It is also possible that there are no terminators at one or more ends of the cable, which leads to reflection of electrical signals in the cable and termination of the network. The network is falling.

The computers themselves on the network remain fully operational, but as long as the segment is broken, they cannot communicate with each other.

The concept of a star network topology (Fig. 2.) comes from the field of mainframe computers, in which the head machine receives and processes all data from peripheral devices as an active data processing node. This principle is applied in data transmission systems. All information between two peripheral workstations passes through the central node of the computer network.

Fig.2. Star topology

Network throughput is determined by the computing power of the node and is guaranteed for each workstation. There are no data collisions. Cabling is quite simple as each workstation is connected to a node. Cabling costs are high, especially when the central node is not geographically located in the center of the topology.

When expanding computer networks, previously made cable connections cannot be used: a separate cable must be laid from the center of the network to the new workplace.

The star topology is the fastest of all computer network topologies because data transfer between workstations passes through a central node (if its performance is good) over separate lines used only by these workstations. The frequency of requests to transfer information from one station to another is low compared to that achieved in other topologies.

The performance of a computer network primarily depends on the power of the central file server. It can be a bottleneck in the computer network. If the central node fails, the entire network is disrupted. The central control node - the file server - implements the optimal protection mechanism against unauthorized access to information. The entire computer network can be controlled from its center.

Advantages

· Failure of one workstation does not affect the operation of the entire network as a whole;

· Good network scalability;

· Easy search for faults and breaks in the network;

· High network performance;

· Flexible administration options.

Flaws

· Failure of the central hub will result in the inoperability of the network as a whole;

· Laying a network often requires more cable than most other topologies;

· A finite number of workstations, i.e. the number of workstations is limited by the number of ports in the central hub.

With a ring topology (Fig. 3.) of the network, workstations are connected to one another in a circle, i.e. workstation 1 with workstation 2, workstation 3 with workstation 4, etc. The last workstation is connected to the first. The communication link is closed in a ring.

Fig.3. Ring topology

Laying cables from one workstation to another can be quite complex and expensive, especially if the geographical location of the workstations is far from the ring shape (for example, in a line). Messages circulate regularly in circles. The workstation sends information to a specific destination address, having previously received a request from the ring. Message forwarding is very efficient since most messages can be sent "on the road" over the cable system one after another. It is very easy to make a ring request to all stations.

The duration of information transfer increases in proportion to the number of workstations included in the computer network.

The main problem with a ring topology is that each workstation must actively participate in the transfer of information, and if at least one of them fails, the entire network is paralyzed. Faults in cable connections are easily localized.

Connecting a new workstation requires a short-term shutdown of the network, since the ring must be open during installation. There is no limit on the length of a computer network, since it is ultimately determined solely by the distance between two workstations. A special form of ring topology is a logical ring network. Physically, it is mounted as a connection of star topologies.

Individual stars are switched on using special switches (English Hub – concentrator), which in Russian are also sometimes called “hub”.

When creating global (WAN) and regional (MAN) networks, the MESH mesh topology is most often used (Fig. 4.). Initially, this topology was created for telephone networks. Each node in such a network performs the functions of receiving, routing and transmitting data. This topology is very reliable (if any segment fails, there is a route along which data can be transferred to a given node) and is highly resistant to network congestion (a route that is least congested with data transmission can always be found).

Fig.4. Mesh topology.

When developing the network, the “star” topology was chosen due to its simple implementation and high reliability (a separate cable goes to each computer).

1) FastEthernet using 2 switches (Fig. 5)

2 segment

1 segment

Rice. 6. FastEthernet topology using 1 router and 2 switches.

4Local network diagram

Below is a diagram of the location of computers and cable routing on floors (Fig. 7, 8).

Rice. 7. Layout of computers and cable routing on the 1st floor.

Rice. 8. Layout of computers and cable routing on the 2nd floor.

This scheme was developed taking into account the characteristic features of the building. The cables will be located under the artificial flooring, in channels specially designated for them. The cable will be pulled to the second floor through a telecommunications cabinet, which is located in the utility room, which is used as a server room where the server and router are located. Switches are located in the main rooms in cabinets.

Layers interact top-down and bottom-up through interfaces and can also interact with the same layer of another system using protocols.

The protocols used at each layer of the OSI model are presented in Table 1.

Table 1.

Protocols of the OSI model layers

OSI layer	Protocols
Applied	HTTP, gopher, Telnet, DNS, SMTP, SNMP, CMIP, FTP, TFTP, SSH, IRC, AIM, NFS, NNTP, NTP, SNTP, XMPP, FTAM, APPC, X.400, X.500, AFP, LDAP, SIP, ITMS, ModbusTCP, BACnetIP, IMAP, POP3, SMB, MFTP, BitTorrent, eD2k, PROFIBUS
Representation	HTTP, ASN.1, XML-RPC, TDI, XDR, SNMP, FTP, Telnet, SMTP, NCP, AFP
Session	ASP, ADSP, DLC, Named Pipes, NBT, NetBIOS, NWLink, Printer Access Protocol, Zone Information Protocol, SSL, TLS, SOCKS
Transport	TCP, UDP, NetBEUI, AEP, ATP, IL, NBP, RTMP, SMB, SPX, SCTP, DCCP, RTP, TFTP
Network	IP, IPv6, ICMP, IGMP, IPX, NWLink, NetBEUI, DDP, IPSec, ARP, RARP, DHCP, BootP, SKIP, RIP
Duct	STP, ARCnet, ATM, DTM, SLIP, SMDS, Ethernet, FDDI, Frame Relay, LocalTalk, Token ring, StarLan, L2F, L2TP, PPTP, PPP, PPPoE, PROFIBUS
Physical	RS-232, RS-422, RS-423, RS-449, RS-485, ITU-T, xDSL, ISDN, T-carrier (T1, E1), Ethernet standard modifications: 10BASE-T, 10BASE2, 10BASE5, 100BASE- T (includes 100BASE-TX, 100BASE-T4, 100BASE-FX), 1000BASE-T, 1000BASE-TX, 1000BASE-SX

It should be understood that the vast majority of modern networks, for historical reasons, only roughly correspond in general terms to the ISO/OSI reference model.

The actual OSI protocol stack developed as part of the project was perceived by many as too complex and virtually unimplementable. It involved the abolition of all existing protocols and their replacement with new ones at all levels of the stack. This made the stack very difficult to implement and was the reason for its abandonment by many vendors and users who had made significant investments in other networking technologies. In addition, the OSI protocols were developed by committees that proposed different and sometimes conflicting characteristics, leading to many parameters and features being declared optional. Because too much was optional or left to the developer's choice, different vendors' implementations simply could not interoperate, thus defeating the very idea of the OSI design.

As a result, OSI's attempt to agree on common standards for networking was supplanted by the TCP/IP protocol stack used on the Internet and its simpler, more pragmatic approach to computer networking. The Internet's approach was to create simple protocols with two independent implementations required for a protocol to be considered a standard. This confirmed the practical feasibility of the standard. For example, the definitions of the X.400 email standards consist of several large volumes, and the definition of Internet mail (SMTP) is only a few dozen pages in RFC 821. However, it is worth noting that there are numerous RFCs that define extensions to SMTP. Therefore, at the moment, complete documentation on SMTP and extensions also takes up several large books.

Most protocols and specifications in the OSI stack are no longer in use, such as X.400 email. Only a few survived, often in greatly simplified form. The X.500 directory structure is still in use today, largely due to the simplification of the original cumbersome DAP protocol, which became known as LDAP and became an Internet standard.

The collapse of the OSI project in 1996 dealt a serious blow to the reputation and legitimacy of the organizations involved, especially the ISO. The biggest omission of the OSI creators was their failure to see and acknowledge the superiority of the TCP/IP protocol stack.

To select a technology, consider a table comparing FDDI, Ethernet and TokenRing technologies (Table 2).

Table 2. Characteristics of FDDI, Ethernet, TokenRing technologies

Characteristic	FDDI	Ethernet	Token Ring
Bit speed, Mbit/s	100	10	16
Topology	Double ring of trees	Tire/star	Star/ring
Data transmission medium	Fiber Optic, Category 5 UTP	Thick coax, thin coax,	Shielded or unshielded twisted pair, fiber optic
Maximum network length (without bridges)	(100 km per ring)	2500 m	40000 m
Maximum distance between nodes	2 km (no more than 11 dB of loss between nodes)	2500 m	100 m
Maximum number of nodes	(1000 connections)	1024	260 for shielded twisted pair, 72 for unshielded twisted pair

After analyzing the table of characteristics of FDDI, Ethernet, TokenRing technologies, the choice of Ethernet technology (or rather its modification FastEthernet), which takes into account all the requirements of our local network, is obvious. Since TokenRing technology provides data transfer speeds of up to 16 Mbit/s, we exclude it from further consideration, and due to the complexity of implementing FDDI technology, it would be most reasonable to use Ethernet.

7Network protocols

The seven-layer OSI model is theoretical and contains a number of shortcomings. Real network protocols have to deviate from it, providing unintended capabilities, so the binding of some of them to OSI layers is somewhat arbitrary.

The main flaw of OSI is the ill-conceived transport layer. On it, OSI allows data exchange between applications (introducing the concept of port - application identifier), however, the ability to exchange simple datagrams in OSI is not provided - the transport layer must form connections, ensure delivery, control the flow, etc. Real protocols implement this possibility .

Network transport protocols provide the basic functionality that computers need to communicate with a network. Such protocols implement complete, efficient communication channels between computers.

The transport protocol can be thought of as a registered mail service. The transport protocol ensures that the transmitted data reaches the specified destination by checking the receipt received from it. It performs monitoring and error correction without higher level intervention.

The main network protocols are:

NWLink IPX/SPX/NetBIOS Compatible Transport Protocol (NWLink) is Novell's NDIS-compatible 32-bit implementation of the IPX/SPX protocol. The NWLink protocol supports two application programming interfaces (APIs): NetBIOS and Windows Sockets. These interfaces allow computers running Windows to communicate with each other, as well as with NetWare servers.

The NWLink transport driver is an implementation of NetWare low-level protocols such as IPX, SPX, RIPX (Routing Information Protocol over IPX) and NBIPX (NetBIOS over IPX). The IPX protocol controls the addressing and routing of data packets within and between networks. The SPX protocol ensures reliable delivery of data by maintaining the correct transmission sequence and acknowledgment mechanism. The NWLink protocol provides NetBIOS compatibility by building a NetBIOS layer on top of the IPX protocol.

IPX/SPX (from the English Internetwork Packet eXchange/Sequenced Packet eXchange) is a protocol stack used in Novell NetWare networks. The IPX protocol provides the network layer (packet delivery, an analogue of IP), SPX - the transport and session layer (an analogue of TCP).

The IPX protocol is designed to transport datagrams on connectionless systems (much like IP or NETBIOS, developed by IBM and emulated by Novell), and provides communications between NetWare servers and end stations.

SPX (Sequence Packet eXchange) and its improved modification SPX II are transport protocols of the ISO 7-layer model. This protocol guarantees packet delivery and uses a sliding window technique (a distant analogue of the TCP protocol). In case of loss or error, the packet is resent, the number of repetitions is set programmatically.

NetBEUI is a protocol that complements the NetBIOS interface specification used by the network operating system. NetBEUI formalizes a transport layer frame that is not standardized in NetBIOS. It does not correspond to any specific layer of the OSI model, but covers the transport layer, network layer and LLC sublayer of the data link layer. NetBEUI interacts directly with NDIS at the MAC level. Thus, it is not a routable protocol.

The transport part of NetBEUI is NBF (NetBIOS Frame protocol). Nowadays, NBT (NetBIOS over TCP/IP) is usually used instead of NetBEUI.

As a rule, NetBEUI is used in networks where it is not possible to use NetBIOS, for example, on computers with MS-DOS installed.

Repeater(English repeater) - designed to increase the distance of a network connection by repeating the electrical signal “one to one”. There are single-port repeaters and multi-port repeaters. In twisted pair networks, a repeater is the cheapest means of combining end nodes and other communications devices into a single shared segment. Ethernet repeaters can have a speed of 10 or 100 Mbit/s (FastEthernet), the same for all ports. Repeaters are not used for GigabitEthernet.

Bridge(from the English bridge - bridge) is a means of transmitting frames between two (or more) logically heterogeneous segments. According to the logic of operation, it is a special case of a switch. The speed is usually 10 Mbit/s (switches are more often used for FastEthernet).

Hub or hub(from the English hub - activity center) - a network device for combining several Ethernet devices into a common segment. Devices are connected using twisted pair, coaxial cable or optical fiber. A hub is a special case of a concentrator

The hub operates at the physical layer of the OSI network model and repeats the signal arriving at one port to all active ports. If a signal arrives on two or more ports at the same time, a collision occurs and the transmitted data frames are lost. This way, all devices connected to the hub are in the same collision domain. Hubs always operate in half-duplex mode; all connected Ethernet devices share the available access bandwidth.

Many hub models have simple protection against an excessive number of collisions arising due to one of the connected devices. In this case, they can isolate the port from the general transmission medium. For this reason, network segments based on twisted pair are much more stable than segments on a coaxial cable, since in the first case each device can be isolated from the general environment by a hub, and in the second case, several devices are connected using one cable segment, and, in In case of a large number of collisions, the hub can only isolate the entire segment.

Recently, hubs have been used quite rarely; instead, switches have become widespread - devices that operate at the data link level of the OSI model and increase network performance by logically separating each connected device into a separate segment, a collision domain.

Switch or switch(from English - switch) Switch (switching hub) According to the principle of frame processing, it is no different from the bridge. Its main difference from a bridge is that it is a kind of communication multiprocessor, since each of its ports is equipped with a specialized processor that processes frames using the bridge algorithm regardless of the processors of other ports. Due to this, the overall performance of the switch is usually much higher than that of a traditional bridge with a single processing unit. We can say that switches are new generation bridges that process frames in parallel.

This is a device designed to connect several computer network nodes within one segment. Unlike a hub, which distributes traffic from one connected device to all others, a switch transmits data only directly to the recipient. This improves network performance and security by freeing other network segments from having to (and being able to) process data that was not intended for them.

The switch operates at the data link layer of the OSI model, and therefore, in general, can only unite hosts of the same network by their MAC addresses. Routers are used to connect multiple networks based on the network layer.

The switch stores a special table in memory (ARP table), which indicates the correspondence of the host MAC address to the switch port. When the switch is turned on, this table is empty and the switch is in learning mode. In this mode, data arriving on any port is transmitted to all other ports of the switch. In this case, the switch analyzes data packets, determining the MAC address of the sending computer, and enters it into a table. Subsequently, if a packet destined for that computer arrives on one of the switch ports, that packet will be sent only to the corresponding port. Over time, the switch builds a complete table for all its ports, and as a result, the traffic is localized.

Switches are divided into managed and unmanaged (the simplest). More complex switches allow you to manage switching at the data link and network levels of the OSI model. They are usually called accordingly, for example Level 2 Switch or simply abbreviated L2. The switch can be managed via Web interface protocol, SNMP, RMON (a protocol developed by Cisco), etc. Many managed switches allow you to perform additional functions: VLAN, QoS, aggregation, mirroring. Complex switches can be combined into one logical device - a stack, in order to increase the number of ports (for example, you can combine 4 switches with 24 ports and get a logical switch with 96 ports).

Interface converter or converter(English mediaconverter) allows you to make transitions from one transmission medium to another (for example, from twisted pair to optical fiber) without logical signal conversion. By amplifying the signals, these devices can overcome limitations on the length of communication lines (if the restrictions are not related to propagation delay). Used to connect equipment with different types of ports.

Three types of converters are available:

× RS-232 converter<–>RS-485;

× USB Converter<–>RS-485;

× Ethernet Converter<–>RS-485.

RS-232 converter<–>RS-485 converts the physical parameters of the RS-232 interface into RS-485 interface signals. Can operate in three reception and transmission modes. (Depending on the software installed in the converter and the state of the switches on the converter board).

USB converter<–>RS-485 - this converter is designed to organize an RS-485 interface on any computer that has a USB interface. The converter is made in the form of a separate board connected to the USB connector. The converter is powered directly from the USB port. The converter driver allows you to create a virtual COM port for the USB interface and work with it as with a regular RS-485 port (similar to RS-232). The device is detected immediately when connected to the USB port.

Ethernet Converter<–>RS-485 - this converter is designed to provide the ability to transmit RS-485 interface signals over a local network. The converter has its own IP address (set by the user) and allows access to the RS-485 interface from any computer connected to the local network and with the appropriate software installed. To work with the converter, 2 programs are supplied: Port Redirector – support for the RS-485 interface (COM port) at the network card level and the Lantronix configurator, which allows you to bind the converter to the user’s local network, as well as set the parameters of the RS-485 interface (baud rate, number of data bits, etc.) The converter provides completely transparent data reception and transmission in any direction.

Router or router(from the English router) is a network device used in computer data networks, which, based on information about the network topology (routing table) and certain rules, makes decisions about forwarding network layer packets of the OSI model to their recipient. Typically used to connect multiple network segments.

Traditionally, a router uses the routing table and the destination address found in the data packets to forward the data. By extracting this information, it determines from the routing table the path along which the data should be transmitted and routes the packet along this route. If there is no described route in the routing table for an address, the packet is discarded.

There are other ways to determine the forwarding route of packets using, for example, the source address, the upper layer protocols used, and other information contained in the network layer packet headers. Often, routers can translate source and recipient addresses (NAT, Network Address Translation), filter the transit data stream based on certain rules to limit access, encrypt/decrypt transmitted data, etc.

Routers help reduce network congestion by dividing it into collision and broadcast domains, as well as packet filtering. They are mainly used to combine networks of different types, often incompatible in architecture and protocols, for example, to combine Ethernet local networks and WAN connections using DSL, PPP, ATM, Frame relay, etc. protocols. A router is often used to provide access from local network to the global Internet, performing the functions of address translation and firewall.

A router can be either a specialized device or a PC computer that performs the functions of a simple router.

Modem(an abbreviation made up of words mo duulator- dem odulator) is a device used in communication systems and performing the function of modulation and demodulation. A special case of a modem is a widely used peripheral device for a computer that allows it to communicate with another computer equipped with a modem through a telephone network (telephone modem) or a cable network (cable modem).

The end network equipment is the source and recipient of information transmitted over the network.

Computer (workstation), connected to the network, is the most versatile node. The applied use of a computer on a network is determined by the software and installed additional equipment. For long-distance communications, a modem is used, internal or external. From a networking point of view, the “face” of a computer is its network adapter. The type of network adapter must match the purpose of the computer and its network activity.

Server is also a computer, but with more resources. This implies its higher network activity and importance. It is advisable to connect servers to a dedicated switch port. When installing two or more network interfaces (including a modem connection) and the corresponding software, the server can play the role of a router or bridge. Servers generally need to have a high-performance operating system.

Table 5 shows the parameters of a typical workstation and its cost for the local network being developed.

Table 5.

Work station

		System unit.GH301EA HP dc5750 uMT A64 X2-4200+(2.2GHz),1GB,160GB,ATI Radeon X300,DVD+/-RW,Vista Business
		Hewlett-Packard GH301EA dc 5750 series computer. This system unit is equipped with an AMD Athlon™ 64 X2 4200+ processor with a frequency of 2.2 GHz, 1024 MB of DDR2 RAM, a 160 GB hard drive, a DVD-RW drive and Windows Vista Business installed.
		Price: RUB 16,450.00
			Monitor. TFT 19 “Asus V W1935

			Price: 6,000.00 rub.
Input Devices
Mouse	Genius GM-03003				172 rub.
Keyboard				208 rub.
total cost					RUB 22,830

Table 6 shows the server parameters.

Table 6.

Server

			DESTEN System unit DESTEN eStudio 1024QM
			Processor INTEL Core 2 Quad Q6600 2.4GHz 1066MHz 8Mb LGA775 OEM Motherboard Gigabyte GA-P35-DS3R ATX Memory module DDR-RAM2 1Gb 667Mhz Kingston KVR667D2N5/1G - 2 Hard drive 250 Gb Hitachi Deskstar T7K500 HDP725 025GLA380 7200RPM 8Mb SATA-2 - 2 Video adapter 512MB Zotac PCI -E 8600GT DDR2 128 bit DVI (ZT-86TEG2P-FSR) DVD drive RW NEC AD-7200S-0B SATA Black ZALMAN HD160XT BLACK housing.
			Price: RUB 50,882.00
			Monitor. TFT 19 “Asus V W1935
			Type: LCD LCD technology: TN Diagonal: 19" Screen format: 5:4 Max resolution: 1280 x 1024 Inputs: VGA Vertical scan: 75 Hz Horizontal scan: 81 KHz
			Price: 6,000.00 rub.
Input Devices
Mouse		Genius GM-03003			172 rub.
Keyboard	Logitech Value Sea Gray (refresh) PS/2			208 rub.
total cost					RUB 57,262

The server software includes:

× Operating system WindowsServer 2003 SP2+R2

× ABBY FineReader Corporate Edition v8.0 (server license)

× Network administration program SymantecpcAnywhere 12 (server)

Workstation software includes:

× Operating system WindowsXPSP2

× Antivirus program NOD 32 AntiVirusSystem.

× Microsoft Office 2003 (pro)

× ABBY FineReader Corporate Edition v8.0 software package (client license)

× Network administration program Symantec pcAnywhere 12 (client)

× User programs

For real networks, an important performance indicator is network utilization, which is a percentage of the total bandwidth (not divided between individual subscribers). It takes into account collisions and other factors. Neither the server nor the workstations contain tools for determining network usage; special hardware and software tools such as protocol analyzers are designed for this, not always available due to the high cost.

For busy Ethernet and FastEthernet systems, 30% network utilization is considered a good value. This value corresponds to the absence of long-term downtime in the network and provides sufficient reserve in case of peak load increases. However, if the network utilization rate is 80...90% or more for a significant time, then this indicates that the resources are almost completely used (at a given time), but does not leave a reserve for the future.

To carry out calculations and conclusions, you should calculate the performance in each network segment.

Let's calculate the payload Pп:

where n is the number of segments of the designed network.

P0 = 2*16 = 32Mbps

The total actual load Pf is calculated taking into account collisions and the magnitude of access delays to the data transmission medium:

, Mbit/s,

(3)

where k is the access delay to the data transmission medium: for the Ethernet family of technologies – 0.4, for TokenRing – 0.6, for FDDI – 0.7.

RF = 32*(1+0.4) = 44.8 Mbit/s

Since the actual load Pf > 10 Mbit/s, then, as previously assumed, this network cannot be implemented using the Ethernet standard; it is necessary to use FastEthernet technology (100 Mbit/s).

Because Given that we do not use hubs in the network, there is no need to calculate the double signal turnaround time. (There is no collision signal)

Table 7 shows the final calculation of the cost of a network built on 2 switches. ( Option 1).

Table 6.

Table 8 shows the final calculation of the cost of a network built on 2 switches and 1 router. ( Option 2).

Table 8.

№	Name		Price for 1 unit. (rub.)	Total (RUB)
1	RJ-45 plugs	86	2	172
2	RJ-45 UTP cable, lev.5e	980m.	20	19 600
3	TrendNet N-Way Switch TEG S224 (10/100Mbps, 24 port, +2 1000Mbps Rack Mount)	2	3714	7 428
4	Router, Router D-Link DIR-100	1	1 250	1 250
5	Work station	40	22 830	913 200
6	Sunrise XD Server (Tower/RackMount)	1	57 262	57 262
Total:				998912

As a result, we get two network options that do not differ significantly in cost and meet the standards for network construction. The first network option is inferior to the second option in terms of reliability, even though network design using the second option is slightly more expensive. Therefore, the best option for building a local network would be option two - a local network built on 2 switches and a router.

To ensure reliable operation and improve network performance, changes to the network structure should be made only taking into account the requirements of the standard.

To protect data from viruses, you must install anti-virus programs (for example, NOD32 AntiVirusSystem), and to restore damaged or erroneously deleted data, you should use special utilities (for example, the utilities included in the NortonSystemWorks package).

Although the network is built with a performance reserve, you should still take care of network traffic, so use the administration program to monitor the intended use of intranet and Internet traffic. The use of NortonSystemWorks utility applications (such as defragmentation, cleaning the registry, fixing current errors using WinDoctor), as well as regular anti-virus scanning at night, will have a beneficial effect on network performance. You should also divide the loading of information from another segment in time, i.e. try to ensure that each segment addresses the other in the time allotted to it. The installation of programs that are not related to the immediate area of the company's activities should be prevented by the administrator. When installing a network, it is necessary to mark the cable so as not to encounter difficulties when servicing the network.

Network installation should be carried out through existing channels and ducts.

For reliable operation of the network, it is necessary to have an employee responsible for the entire local network and involved in optimizing it and increasing productivity.

Peripheral equipment (printers, scanners, projectors) should be installed after the specific assignment of work station responsibilities.

For preventive purposes, the integrity of the cables in the secret floor should be periodically checked. When dismantling the equipment, you should handle the equipment carefully so that it can be used again.

In addition, it is necessary to limit access to the server room and to cabinets with switches.

1. V.G. Olifer, N.A. Olifer - St. Petersburg. Peter 2004

2. http://ru.wikipedia.org/wiki/

3. V.M. Shek, T.A. Kuvashkina “Guidelines for course design in the discipline Computer Networks and Telecommunications” - Moscow, 2006

4. http://catalog.sunrise.ru/

5. V.M. Shek. Lectures on the discipline “Computer networks and telecommunications”, 2008.

« This is the official instruction for Alcatel PIXI 3(4.5) 4027D in Russian, which is suitable for Android 4.4. In case you have updated your... »

Installing TTF fonts on your computer How to remove an installed font

« What is a font? A font is a graphical representation of a set of characters (numbers, letters and symbols). Fonts are installed with the operating system... »

Setting up a wi-fi router from Netbynet

« In this article, we will look at setting up the D-Link DIR-300 router using the example of the NetByNet provider...Manual for setting up the D-Link DIR-300 router. 1.... »