Types of switching in communication channels. WAN networks with circuit and packet switching

In networks with circuit switching, subscribers are connected by a composite channel formed by network switches at the request of one of the subscribers; with this switching method, before transmitting data, it is always necessary to perform a connection establishment procedure, during which a composite channel is created.

Circuit-switched networks are good at switching data streams of constant intensity, for example, data streams created by interlocutors talking on the phone, but cannot dynamically redistribute the capacity of trunk channels between streams of subscriber channels.

To jointly share channels between network switches with multiple subscriber channels, two technologies are used: frequency division technology (FDM) and time division technology (TDM).

Frequency division is typical for analog modulation of signals, and time division is characteristic for digital coding. Frequency Division Division (FDM) technology was developed for telephone networks, but is also used for other types of networks, such as cable television networks and computer networks. During the transition to a digital form of voice representation, a new technology was developed that focuses on the discrete nature of the transmitted data - this is the time division technology (TDM).

Packet switching.

Packet switching is a subscriber switching technique that was specifically designed for the efficient transmission of computer traffic.

With the packet switching method, all messages sent by users are broken up at the source node into relatively small parts called packets. Each packet is equipped with a header, which necessarily indicates the address information necessary to deliver the packet to the destination node and other service information. WAN switches receive packets and, based on address information, transmit them to each other, and ultimately to the destination node.

Packet network switches differ from circuit switches in that they have internal buffer memory to temporarily store packets if the switch's output port is of this package busy sending another packet. In this case, the package spends some time in the packet queue in buffer memory output port, when its turn reaches it, it is transferred to the next switch.

Circuit-switched networks operate efficiently in the sense that the amount of data transmitted from all network subscribers per unit time is greater than when using a circuit-switched network. However, for each pair of subscribers, the network throughput may be lower than that of a circuit-switched network due to packet queues in the switches.

Packet sizes have a significant impact on network performance. Typically, packets on networks are 1–4 KB in size.

Packet switched networks can operate in one of two modes: datagram mode or mode virtual channels.

At datagram mode packet transmission assumes independent routing of each packet. In this case, the switch can change the route of any packet depending on the state of the network. The datagram method does not require prior connection establishment and therefore operates without delay before data transmission.

Virtual channel mode involves transmitting packets along a predetermined path - over a virtual channel. In this case, before data can be transmitted between two end nodes, a virtual circuit must be established, which is the only route connecting these nodes. The time spent on establishing a virtual channel is compensated by subsequent fast transfer the entire packet stream. The virtual channel can be dynamic And permanent.

A dynamic virtual channel is established for one communication session; for this purpose, a special service packet is sent to the network - a request to establish a connection. This packet, passing through network devices, “lays” a virtual channel through which these packets will be transmitted.

Permanent virtual circuits are created by the network administrator by manually configuring switches. Circuit switched networks have several important general properties

no matter what type of multiplexing they use. Dynamic switching networks require preliminary procedure

establishing a connection between subscribers. To do this, the address of the called subscriber is transmitted to the network, which passes through the switches and configures them for subsequent data transmission. The connection request is routed from one switch to another and eventually reaches the called party. The network may refuse to establish a connection if the capacity of the required output channel is already exhausted. For an FDM switch, the capacity of the output channel is equal to the number of frequency bands of this channel, and for a TDM switch, it is equal to the number of time slots into which the channel’s operating cycle is divided. The network also refuses the connection if the requested subscriber has already established a connection with someone else. In the first case, they say that the switch is busy, and in the second - the subscriber. The possibility of connection failure is a disadvantage of the circuit switching method.

If the connection can be established, then it is allocated a fixed frequency band in FDM networks or a fixed bandwidth in TDM networks. These values ​​remain unchanged throughout the connection period. Guaranteed network throughput once a connection is established is an important property required for applications such as voice, video or real-time facility control. However, circuit-switched networks cannot dynamically change the channel capacity at the request of a subscriber, which makes them ineffective in conditions of bursty traffic. The disadvantage of circuit-switched networks is the inability to use user equipment that works with at different speeds

Circuit-switched networks are well suited for switching constant-rate data streams, where the unit of switching is not a single byte or data packet, but a long-term synchronous data stream between two subscribers. For such flows, circuit-switched networks add a minimum of overhead to route data through the network, using the time position of each bit of the flow as its destination address in the network switches.

1. Packet switching

Packet switching is a subscriber switching technique that was specifically designed for the efficient transmission of computer traffic. Experiments to create the first computer networks based on circuit switching technology showed that this type of switching does not allow achieving high overall network throughput. The crux of the problem lies in the bursty nature of traffic that typical network applications generate. For example, when accessing a remote file server, the user first views the contents of that server's directory, which results in the transfer of a small amount of data. It then opens the required file in text editor, and this operation can create quite a lot of data exchange, especially if the file contains large graphical inclusions. After displaying a few pages of a file, the user works with them locally for a while, which requires no network transfer at all, and then returns modified copies of the pages to the server - again creating intensive network transfer.

Traffic ripple factor of an individual network user, equal to ratio average intensity of data exchange to the maximum possible, can be 1:50 or 1:100. If for the described session we organize channel switching between the user’s computer and the server, then most of the time the channel will be idle. At the same time, the switching capabilities of the network will be used - part of the time slots or frequency bands of the switches will be occupied and unavailable to other network users.

When packet switching occurs, all messages transmitted by a network user are broken up at the source node into relatively small parts called packets. Let us recall that a message is a logically completed piece of data - a request to transfer a file, a response to this request containing the entire file, etc. Messages can have an arbitrary length, from several bytes to many megabytes. On the contrary, packets can usually also have a variable length, but within narrow limits, for example from 46 to 1500 bytes. Each packet is provided with a header that specifies the address information needed to deliver the packet to the destination node, as well as the packet number that will be used by the destination node to assemble the message (Figure 1.39). Packets are transported in the network as independent information blocks. Network switches receive packets from end nodes and, based on address information, transmit them to each other, and ultimately to the destination node.

Figure 1.39

Packet network switches differ from circuit switches in that they have internal buffer memory for temporary storage of packets if the output port of the switch is busy transmitting another packet at the time the packet is received (Figure 1.40). In this case, the packet remains for some time in the packet queue in the buffer memory of the output port, and when its turn reaches it, it is transferred to the next switch. This data transmission scheme allows you to smooth out traffic ripples on the backbone links between switches and thereby use them in the most effective way to increase the throughput of the network as a whole.

Figure 1.40

Indeed, for a pair of subscribers, the most effective would be to provide them with sole use of a switched communication channel, as is done in circuit-switched networks. With this method, the interaction time of this pair of subscribers would be minimal, since data would be transmitted from one subscriber to another without delay. Subscribers are not interested in channel downtime during transmission pauses; it is important for them to quickly solve their own problem. A packet-switched network slows down the process of interaction between a particular pair of subscribers, since their packets can wait in the switches while other packets that arrived at the switch earlier are transmitted along the backbone links.

However, the total amount of computer data transmitted by the network per unit time using the packet switching technique will be higher than using the circuit switching technique. This happens because the ripples of individual subscribers, in accordance with the law large numbers distributed over time. Therefore, switches are constantly and fairly evenly loaded with work if the number of subscribers they serve is really large. Figure 2.40 shows that traffic from end nodes to switches is very unevenly distributed over time. However, switches are more high level hierarchies that serve connections between lower-level switches are more evenly loaded, and the packet flow on the trunk links connecting upper-level switches is at near-maximum utilization.

The higher efficiency of packet-switched networks compared to circuit-switched networks (with equal communication channel capacity) was proven in the 60s both experimentally and using simulation modeling. An analogy with multiprogram operating systems is appropriate here. Each individual program in such a system takes longer to execute than in a single-program system, where the program is allocated all the processor time until it completes its execution. However, the total number of programs executed per unit of time is greater in a multi-program system than in a single-program system.

) is connected using a terminal device (T), which sends information to the network at the same speed. This speed is equal to the channel. If situations arise when the user transmits volumes of information that are less than the channel capacity, then the Terminal device fills the void with empty data. This is shown in Fig. 2.

Figure 2

That some of the information actually exists complemented by emptiness The recipient's Terminal device also knows, which discards the supplemented information.

Establishing a connection

To exchange information, you first need to establish a connection via . During installation, a connection may arise. Let's say two objects A and B want to exchange data (see Fig. 1). First you need to send request into the switching network, where the object specifies the address of object B. The task of sending a request is to make a connection between objects information channel, the characteristics of which are similar to a continuous connection, that is, throughout the entire time established connection data is transferred at the same speed and volume. This means that in transit switches no need to buffer information objects.

To create a connection, the request must go through a series of switches that lie on the channel from A to B, and make sure that all sections of the path are in this moment free.

Connection refused

The only one positive thing such a connection, this means that the delay level is minimal and transmit real time/(voice, video) will be very convenient.
The negative aspects are that each physical line always transmits data at the same speed, which is inefficient. And the use of resources is also not efficient as shown in Fig. 1. The solution to circuit switching problems is multiplexing.

Packet switching

The packet switching algorithm was specially made for effective exchange computer traffic. When an entity transmits switched packets, the data is broken up at the originating node into small pieces called frames. Each package is given title, which contains the delivery address. Figure 3 shows the breakdown of the data stream into packets. Another additional field that is added to the end of the packet is limit switch. Placed there check sum , which allows you to check whether the information was changed during transmission or not.

Figure 3

Packets enter the network without pre-reservation of network channels and not with preset speed, as implemented in switched networks. And it is transmitted at the rate at which the source generates. It is assumed that a packet-switched network is always ready to receive a packet from an object, unlike a circuit-switched network.

The bandwidth reservation scheme can also be used in packet networks. But the basic idea of ​​such a reservation is fundamentally different from the idea of ​​reserving bandwidth in circuit-switched networks. The difference is that the channel capacity of a packet-switched network can dynamically change between information communication lines depending on the current tasks of each channel, which circuit switching technology cannot implement.

Circuit switching

When switching channels, such a network implements a permanent, integral physical channel between end nodes from successive connected intermediate sections using a switch. The main condition for such a channel is the same data transfer rate in each section. Equality determines that the switches of such a network should not buffer transported data. Figure 4 shows a network operating using circuit switching technology. In order for node 1 to transfer data to node 7, a special request must first be received to implement the connection to switch A, indicating the destination address 7. Switch A must designate the route of the composite channel, and then transmit the request to the next switch, in Fig. this is switch E. Then switch E sends a request to switch F, which then passes it on to node 7. Node 7 accepts the request to establish a connection, and then it responds to the original node along the assigned route.

Figure - 4

Advantages of circuit switching:

• Known and constant speed of information transfer over established channel
• Permanent and low level delays in transporting information through the network

Disadvantages of circuit switching:

• Poor implementation of bandwidth physical channels. The transmission of information may be uneven, and the dedicated channel may be idle
• Mandatory delay before transporting information due to connection establishment

Pros and cons of any network technology are relative, since in different situations pros can act as cons and vice versa.

Figure - 5

Comparison of switching methods:

Dynamic and permanent switching

Dynamic switching networks:

• it is allowed to implement a connection at the initiative of the user of this network
• switching is implemented only for the duration of the communication session, and then is terminated at the user’s initiative
• The user can implement a connection with any network user
• The time required to establish a connection between a pair of users can be from a couple of seconds to several hours and ends after completion of work - file transfer, etc.

Examples of such networks are local area networks or TCP/IP.

Constantly switched networks:

• Allows a pair of users to order a connection for a long period of time
• The connection is created by special personnel who maintain the network, and not by users
• The permanent switching mode in circuit-switched networks is called dedicated or leased circuit service.

The most popular networks in permanent switching are SDH.

Circuit switching based on time sharing

Frequency division switching was designed to transmit continuous signals, representing voice. During the transition to a digital form of voice representation, it was developed new technology multiplexing, focusing on the discrete nature of the transmitted data.

This technique is called time division multiplexing (TDM). Rice. 3.3. explains the principle of circuit switching based on TDM technology.

Rice. 3.3. Switching based on channel division in time

TDM network equipment - multiplexers, switches, demultiplexers - operates in time-sharing mode, alternately servicing all subscriber channels during its operation cycle. The operating cycle of TDM equipment is 125 μs, which corresponds to the period of voice measurements in a digital subscriber channel. This means that the multiplexer or switch manages to service any subscriber channel in a timely manner and transmit its next measurement further along the network. Each connection is allocated one time slice of the hardware operation cycle, also called a time slot. The duration of a time slot depends on the number of subscriber channels served by the TDM multiplexer or switch.

The multiplexer receives information via N input channels from end subscribers, each of which transmits data over the subscriber channel at a speed of 64 Kbps - 1 byte every 125 μs. In each cycle, the multiplexer performs the following actions:

· receiving the next byte of data from each channel;

· compiling a compressed frame from the received bytes, also called a frame;

· transmission of a compressed frame to the output channel with a bit rate equal to Nx64 Kbps.

The order of the bytes in the holder corresponds to the number of the input channel from which this byte was received. The number of subscriber channels served by the multiplexer depends on its speed. For example, the T1 multiplexer, the first industrial multiplexer to use TDM technology, supports 24 input subscriber channels, producing T1 standard output clips transmitted at a bit rate of 1.544 Mbps.

The demultiplexer performs the opposite task - it parses the bytes of the compressed frame and distributes them across its several output channels, while it considers that the sequence number of the byte in the frame corresponds to the number of the output channel.

The switch receives a compressed frame over a high-speed channel from the multiplexer and writes each byte from it into a separate cell of its buffer memory, and in the order in which these bytes were packed into the compressed frame. To perform a switching operation, bytes are retrieved from the buffer memory not in the order they were received, but in an order that corresponds to the subscriber connections supported on the network. So, for example, if the first subscriber on the left side of the network in Fig. 3.3 must connect to the second subscriber on the right side of the network, then the byte written to the first buffer memory cell will be retrieved from it second. "Stirring" in the right way bytes in a clip, the switch provides the connection of end subscribers in the network.

Once allocated, a time slot number remains at the disposal of the input channel-output slot connection for the entire lifetime of that connection, even if the transmitted traffic is bursty and does not always require the allocated number of time slots. This means that a connection in a TDM network always has a known and fixed throughput that is a multiple of 64 Kbps.

The operation of TDM equipment is similar to the operation of packet-switched networks, since each byte of data can be considered an elementary packet. However, unlike a computer network packet, a TDM network "packet" does not have a unique address. Its address is the serial number in the clip or the number of the allocated time slot in the multiplexer or switch. Networks using TDM technology require synchronous operation all equipment. A violation of synchronization destroys the required switching of subscribers, since address information is lost. Therefore, redistribution of time slots between different channels in TDM equipment is impossible, even if in some cycle of the multiplexer the time slot of one of the channels turns out to be redundant, since there is no data for transmission at the input of this channel at that moment (for example, a telephone network subscriber is silent).

TDM networks can support either dynamic switching mode or persistent switching mode, and sometimes both modes. For example, the main mode of digital telephone networks operating on the basis of TDM technology is dynamic switching, but they also support permanent switching, providing their subscribers with a dedicated circuit service.

There is equipment that supports only constant switching mode. This includes T1/E1 type equipment, as well as high-speed SDH equipment. Such equipment is used to build primary networks, the main function of which is to create dedicated channels between switches that support dynamic switching.

Today, almost all data - voice, image, computer data - is transmitted in digital form. Therefore, dedicated TDM technology channels that provide Lower level for transmitting digital data are universal channels for building networks of any type: telephone, television and computer.

Circuit-switched networks have several important common properties, regardless of the type of multiplexing they use.

1. Networks with dynamic switching require a preliminary procedure for establishing a connection between subscribers. To do this, the address of the called subscriber is transmitted to the network, which passes through the switches and configures them for subsequent data transmission. The connection request is routed from one switch to another and eventually reaches the called party. The network may refuse to establish a connection if the capacity of the required output channel is already exhausted. For an FDM switch, the capacity of the output channel is equal to the number of frequency bands of this channel, and for a TDM switch - the number of time slots into which the channel's operating cycle is divided. The network also refuses the connection if the requested subscriber has already established a connection with someone else. In the first case they say that the switch is busy, and in the second - the subscriber. The possibility of connection failure is a disadvantage of the circuit switching method.

2. If the connection can be established, then it is allocated a fixed frequency band in FDM networks or a fixed bandwidth in TDM networks. These values ​​remain unchanged throughout the connection period. The guaranteed network throughput after the connection is established is important property, necessary for applications such as voice, image, or real-time object control. However, dynamically change throughput Circuit-switched networks cannot provide channels at the request of the subscriber, which makes them ineffective in conditions of bursty traffic.

3. The disadvantage of circuit-switched networks is the inability to use user equipment operating at different speeds. The individual parts of a composite circuit operate at the same speed because circuit-switched networks do not buffer user data.

Computer networks Lecture No. 1 6th semester.

Evaluation of finished products?

Accounting for finished products and their sale?

Production cost accounting?

Concept, classification and evaluation of Finnish investments? Financial investments – investing in their organizations Money and other free resources in assets are not connected. with core activities and the creation of durable facilities. Finnish investments include: 1.state and muniz. securities 2.securities of other organizations 3.contributions to the authorized capital of other organizations, incl. under a simple partnership agreement 4. provision of loans to other organizations, deposits in loans to organizations 5. accounts receivable acquired on the basis of assignment of legal claims. Finnish investments do not include: 1. own shares purchased from shareholders for subsequent resale or cancellation. 2. bills issued to the seller when paying for goods, works and services sold. 3. an organization’s investment in a property that has a material form and is provided for a fee for temporary use 4. precious metals, jewelry, works of art and other similar valuables acquired not for sale common species activities. To accept assets for accounting purposes as Finnish investments, the following conditions must be simultaneously met: 1. availability of properly executed documents 2. transfer of risks associated with Finnish investments 3. ability to bring similar benefits in the future . A one-time fulfillment of 3 conditions is required. Finnish classification. attachments: 1. In connection with the authorized capital: Finnish. investments for the purpose of forming the authorized capital (purchase of shares), Finnish. investments not related to the formation of authorized capital. 2. By type of ownership: state and corporate 3. By term: long-term (over 1 year) and short-term. Finnish assessment. attachments: Finn. investments are accepted for accounting according to the initial st-ti (valid general rules grade). When contributing to the authorized capital, the initial cost is determined by a monetary valuation agreed upon with the founders. If the cost exceeds 200mrokt required. assessment by recent appraisers.

Switching is the process of connecting subscribers of a communication network through transit nodes.

Communication networks must ensure communication between their subscribers. Subscribers can be computers, segments local networks, fax machines or telephone interlocutors. Typically in networks public access it is impossible to provide each pair of subscribers with their own physical line communications that they could monopolize and use at any time. Therefore, the network always uses some method of switching subscribers, which ensures the division of existing physical channels between several communication sessions and between network subscribers.

Each subscriber is connected to switches individual line connection assigned to this subscriber. Communication lines stretched between switches are shared by several subscribers, that is, they are used together.

There are four fundamentally various schemes switching subscribers in networks:

• Circuit switching (CC, circuit switching) - organization of a composite channel through several transit nodes from several sequentially “connected” channels for the duration of message transmission (operational switching) or for more long term(permanent/long-term switching - switching time is determined administratively, that is, a technician came and physically switched channels for an hour, a day, a year, forever, etc., then came and unswitched).
• Message switching (KS, message switching) is the division of information into messages that are transmitted sequentially to the nearest transit node, which, having received the message, remembers it and transmits it further in the same way. It turns out something like a conveyor belt.
• Packet switching (CP, packet switching) - splitting a message into “packets” that are transmitted separately. The difference between a message and a packet: the size of a packet is limited technically, messages are limited logically. In this case, if the route of packets between nodes is determined in advance, they speak of a virtual channel (with connection establishment). Example: IP packet switching. If for each packet the problem of finding a path is solved anew, they speak of a datagram (connectionless) method of packet switching.
• Cell switching (cell switching) - combines the properties of circuit-switched networks and packet-switched networks; when switching cells, packets always have a fixed and relatively small size.

Externally, these diagrams correspond to those shown in Fig. 1 network structure, but their capabilities and properties are different.

Rice. 1. General structure switched networks

Circuit-switched networks have a richer history, having evolved from the first telephone networks. Packet switching networks are relatively new, having emerged in the late 60s as a result of experiments with the first global computer networks. Each of these schemes has its own advantages and disadvantages, but according to the long-term forecasts of many experts, the future belongs to packet switching technology, as it is more flexible and universal.

Circuit switching
When switching channels switching network forms a continuous composite physical channel between end nodes from intermediate channel sections connected in series by switches. The condition that several physical channels with serial connection form a single physical channel, the data transmission rates in each of the constituent physical channels are equal. Equality of speeds means that the switches of such a network do not have to buffer the transmitted data.

In a circuit-switched network, before transmitting data, it is always necessary to perform a connection establishment procedure, during which a composite channel is created. And only after that you can start transferring data.

For example, if the network shown in Fig. 1 operates using circuit switching technology, then node 1, in order to transmit data to node 7, must first transmit special request to establish a connection to switch A, indicating the destination address 7. Switch A must select a route for creating a composite channel, and then transmit the request to the next switch, in in this case E. Switch E then transmits the request to switch F, which in turn transmits the request to node 7. If node 7 accepts the request to establish a connection, it sends a response to the original node through the already established channel, after which the composite channel is considered switched, and the nodes 1 and 7 can exchange data over it.

Rice. 2. Establishing a composite channel

The circuit switching technique has its advantages and disadvantages.