throughput is equal. Internet speed - what is it and how is it measured, how to increase the speed of the Internet connection

After assessing the required bandwidth in each section of the IP network, it is necessary to decide on the choice of network and channel technologies. OSI levels. In accordance with the selected technologies, the most suitable models are determined network equipment. This question is also not easy, since throughput directly depends on the performance of the hardware, and performance, in turn, depends on the software and hardware architecture. Let us consider in more detail the criteria and methods for assessing the throughput of channels and equipment in IP networks.

Throughput Evaluation Criteria

Since the advent of teletraffic theory, many methods have been developed to calculate channel capacities. However, unlike the calculation methods applied to circuit-switched networks, the calculation of the required throughput in packet networks is quite complex and is unlikely to provide accurate results. First of all, this is due to a huge number of factors (especially inherent in modern multiservice networks), which are quite difficult to predict. In IP networks, a common infrastructure is typically shared by multiple applications, each of which may use its own distinct traffic model. Moreover, within one session, the traffic transmitted in the forward direction may differ from the traffic passing in the opposite direction. In addition, calculations are complicated by the fact that the speed of traffic between individual network nodes can change. Therefore, in most cases, when building networks, the throughput estimate is actually determined by general recommendations manufacturers, statistical studies and experience of other organizations.

In order to more or less accurately determine how much bandwidth is required for the network being designed, it is necessary first of all to know which applications will be used. Further, for each application, it is necessary to analyze how the data transfer will take place during the selected periods of time, which protocols are used for this.

For a simple example consider small applications corporate network.

Throughput Calculation Example

Suppose there are 300 working computers and the same number of IP phones on the network. It is planned to use the following services: e-mail, IP-telephony, video surveillance (Fig. 1). For video surveillance, 20 cameras are used, from which video streams are transmitted to the server. Let's try to estimate what maximum bandwidth is required for all services on the channels between the network core switches and at the junctions with each of the servers.

It should be noted right away that all calculations must be carried out for the time of the greatest network activity of users (in the theory of teletraffic - CNN, peak hours), since usually during such periods network performance is most important and the resulting delays and application failures associated with lack of bandwidth , are unacceptable. In organizations huge pressure to the network may occur, for example, at the end of the reporting period or during the seasonal influx of customers, when nai large quantity phone calls and send most of the mail messages.

Email
Returning to our example, consider an email service. It uses protocols that run on top of TCP, that is, the data transfer rate is constantly adjusted, trying to take up all the available bandwidth. Thus, we will start from the maximum value of the delay in sending a message - suppose 1 second will be enough for the user to be comfortable. Next, you need to estimate the average volume of the message sent. Let's assume that during the peaks of activity, mail messages will often contain various attachments (copies of invoices, reports, etc.), so for our example, we will take the average message size of 500 kb. And finally, the last parameter that we need to choose is the maximum number of employees who simultaneously send messages. Let's assume that during the rush job, half of the employees simultaneously press the "Submit" button in mail client. The required maximum throughput for email traffic would then be (500 kb x 150 hosts)/1 s = 75,000 kb/s or 600 Mbps. From this we can immediately conclude that for the connection mail server the network must use a Gigabit Ethernet link. In the core of the network, this value will be one of the terms that make up the total required bandwidth.

Telephony and video surveillance
Other applications - telephony and video surveillance - are similar in their streaming structure: both types of traffic are transmitted using the UDP protocol and have a more or less fixed transmission rate. The main differences are that for telephony the streams are bidirectional and limited by the call time, for video surveillance the streams are transmitted in one direction and, as a rule, are continuous.

To estimate the required bandwidth for telephony traffic, let's assume that the number of simultaneous connections passing through the gateway can reach 100 during peak periods. When using the G.711 codec in Ethernet networks the speed of one stream, taking into account headers and service packets, is approximately 100 kbps. Thus, during periods of the highest user activity, the required bandwidth in the network core will be 10 Mbps.

Video surveillance traffic is calculated quite simply and accurately. Suppose, in our case, video cameras transmit streams of 4 Mbps each. The required bandwidth will be equal to the sum of the speeds of all video streams: 4 Mbps x 20 cameras = 80 Mbps.

Finally, it remains to add the resulting peak values ​​for each of network services: 600 + 10 + 80 = 690 Mbps. This will be the required bandwidth in the network core. The design should also consider the possibility of scaling so that the communication links can serve the traffic of a growing network for as long as possible. In our example, the use of Gigabit Ethernet will be enough to satisfy the requirements of the services and at the same time be able to seamlessly grow the network by connecting more nodes.

Of course, the given example is far from being a reference one - each case must be considered separately. In reality, the network topology can be much more complex (Fig. 2), and throughput needs to be estimated for each of the network sections.

It should be taken into account that VoIP traffic (IP telephony) is distributed not only from phones to the server, but also directly between phones. In addition, different departments of the organization may have different network activity: the technical support service makes more phone calls, the project department is more active than others. email, the engineering department consumes Internet traffic more than others, etc. As a result, some parts of the network may require more bandwidth than others.

Useful and total bandwidth

In our example, when calculating the IP telephony stream rate, we took into account the codec used and the size of the packet header. This is an important detail to keep in mind. Depending on the encoding method (codecs used), the amount of data transmitted in each packet, and the link layer protocols used, the total bandwidth of the stream is formed. It is the total bandwidth that should be taken into account when estimating the required network bandwidth. This is most relevant for IP telephony and other applications that use low-rate real-time streaming, in which the size of the packet headers is a significant fraction of the packet size. For clarity, let's compare two VoIP streams (see table). These streams use the same compression, but different payload sizes (the actual digital audio stream) and different link layer protocols.

The data transfer rate in its purest form, without taking into account the headers of network protocols (in our case, the digital audio stream), is a useful bandwidth. As can be seen from the table, with the same useful throughput of streams, their total throughput can vary greatly. Thus, when calculating the required network bandwidth for telephone calls during peak loads, especially for telecom operators, the choice of channel protocols and flow parameters plays a significant role.

Equipment selection

The choice of link-layer protocols is usually not a problem (today the question is more often about how much bandwidth an Ethernet channel should have), but even an experienced engineer can find it difficult to choose the right equipment.

Development network technologies Along with the growing demands of applications for network bandwidth, it forces manufacturers of network equipment to develop new software and hardware architectures. Often, at a single manufacturer, there are at first glance similar models of equipment, but designed to solve different network problems. Take, for example, Ethernet switches: along with the usual switches used in enterprises, most manufacturers have switches for building storage networks, for organizing operator services, etc. Models of one price category differ in their architecture, "sharpened" for specific tasks.

In addition to overall performance, the choice of hardware should also be driven by the supported technologies. Depending on the type of equipment, a certain set of functions and types of traffic can be processed at the hardware level, without using CPU and memory resources. In this case, the traffic of other applications will be processed at the software level, which greatly reduces the overall performance and, as a result, the maximum throughput. For example, multilayer switches, due to their sophisticated hardware architecture, are capable of transmitting IP packets without performance degradation when all ports are fully loaded. Moreover, if we want to use more complex encapsulation (GRE, MPLS), then such switches (at least inexpensive models) are unlikely to suit us, since their architecture does not support the corresponding protocols, and in best case such encapsulation will occur at the expense of a low performance CPU. Therefore, to solve such problems, one can consider, for example, routers whose architecture is based on a high-performance central processor and depends more on software than hardware implementation. In this case, at the expense of maximum throughput, we get a huge set of supported protocols and technologies that are not supported by switches of the same price category.

Overall equipment performance

In the documentation for their equipment, manufacturers often indicate two values ​​\u200b\u200bof the maximum throughput: one is expressed in packets per second, the other in bits per second. This is due to the fact that most of the performance of network equipment is usually spent on processing packet headers. Roughly speaking, the equipment must accept the packet, find a suitable switching path for it, form a new header (if necessary) and pass it on. Obviously, in this case, it is not the amount of data transmitted per unit of time that plays a role, but the number of packets.

If we compare two streams transmitted at the same rate but with different packet sizes, then the stream with a smaller packet size will require more performance. This fact should be taken into account if the network is supposed to use, for example, a large number of IP telephony streams - the maximum throughput in bits per second here will be much less than the declared one.

It is clear that with mixed traffic, and even taking into account additional services(NAT, VPN), as it happens in the vast majority of cases, it is very difficult to calculate the load on equipment resources. Often, equipment manufacturers or their partners conduct Stress Testing different models under different conditions and the results are published on the Internet in the form comparison tables. Familiarization with these results greatly simplifies the task of choosing an appropriate model.

Pitfalls of modular equipment

If the selected network equipment is modular, then, in addition to the flexible configuration and scalability promised by the manufacturer, you can get a lot of "pitfalls".

When choosing modules, you should carefully read their description or consult the manufacturer. It is not enough to be guided only by the type of interfaces and their number - you also need to familiarize yourself with the architecture of the module itself. It is not uncommon for similar modules when, when transmitting traffic, some are able to process packets autonomously, while others simply forward packets to the central processing unit for further processing (accordingly, for the same external modules, the price for them can vary several times). In the first case overall performance equipment and, as a result, its maximum throughput turn out to be higher than in the second, since part of its work CPU shifts to the module processors.

In addition, modular equipment often has a blocking architecture (when the maximum throughput is lower than the total speed of all ports). This is due to the limited bandwidth of the internal bus through which the modules exchange traffic between themselves. For example, if a modular switch has a 20 Gbps internal bus, then only 20 ports can be used for its line card with 48 Gigabit Ethernet ports when fully loaded. Such details should also be kept in mind and when choosing equipment, carefully read the documentation.

When designing IP networks, bandwidth is a key parameter that will affect the architecture of the network as a whole. For a more accurate estimate of throughput, you can follow these guidelines:

1. Study the applications that you plan to use on the network, the technologies they use, and the volumes of transmitted traffic. Use the advice of developers and the experience of colleagues to take into account all the nuances of these applications when building networks.
2. Study in detail network protocols and technologies that are used by these applications.
3. Read the documentation carefully when choosing equipment. To have some stock of ready-made solutions, check out the product lines of different manufacturers.

As a result, at right choice technologies and equipment, you can be sure that the network will fully meet the requirements of all applications and, being flexible and scalable enough, will last for a long time.

1. What is the process of information transfer?

Transfer of information- the physical process by which the movement of information in space is carried out. They recorded the information on a disk and transferred it to another room. This process is characterized by the presence of the following components:

The source of information. Information receiver. Information carrier. transmission medium.

Information transfer scheme:

Information source - information channel - information receiver.

Information is presented and transmitted in the form of a sequence of signals, symbols. From the source to the receiver, the message is transmitted through some material medium. If technical means of communication are used in the transmission process, then they are called information transmission channels (information channels). These include telephone, radio, TV. Human sense organs play the role of biological information channels.

The process of transferring information technical channels communication takes place according to the following scheme (according to Shannon):

The term "noise" refers to various kinds of interference that distort the transmitted signal and lead to loss of information. Such interferences, first of all, arise for technical reasons: poor quality of communication lines, insecurity from each other of various flows of information transmitted over the same channels. Used for noise protection different ways, for example, the use of various kinds of filters that separate the useful signal from the noise.

Claude Shannon developed a special coding theory that provides methods for dealing with noise. One of the important ideas of this theory is that the code transmitted over the communication line must be redundant. Due to this, the loss of some part of the information during transmission can be compensated. However, you can not make the redundancy too large. This will lead to delays and higher communication costs.

2. General information transfer scheme

3. List the communication channels you know

Communication channel (eng. channel, data line) - a system of technical means and a signal propagation environment for transmitting messages (not just data) from a source to a recipient (and vice versa). A communication channel, understood in a narrow sense ( communication path), represents only the physical medium of signal propagation, for example, a physical communication line.

According to the type of distribution medium, communication channels are divided into:

wired; acoustic; optical; infrared; radio channels.

4. What is telecommunications and computer telecommunications?

Telecommunications(Greek tele - far away, and lat. communicatio - communication) is the transmission and reception of any information (sound, image, data, text) over a distance through various electromagnetic systems (cable and fiber optic channels, radio channels and other wired and wireless channels connections).

telecommunications network- a system of technical means through which telecommunications are carried out.

Telecommunication networks include:

1. Computer networks (for data transmission)

2. Telephone networks (transmission of voice information)

3. Radio networks (transmission of voice information - broadcast services)

4. Television networks (voice and image transmission - broadcast services)

Computer telecommunications - telecommunications, the terminal devices of which are computers.

The transfer of information from computer to computer is called synchronous communication, and through an intermediate computer, which allows you to accumulate messages and transfer them to personal computers as requested by the user, - asynchronous.

Computer telecommunications are beginning to take root in education. In higher education, they are used to coordinate scientific research, prompt exchange of information between project participants, distance learning, consultations. In system school education- to increase the effectiveness of students' independent activities related to various types of creative work, including educational activities, based on the widespread use of research methods, free access to databases, and the exchange of information with partners both domestically and abroad.

5. What is the bandwidth of the information transmission channel?

Bandwidth - metric characteristic, showing the ratio the limit on the number of passing units ( information, objects, volume ) per unit of time through a channel, system, node.

In computer science, the definition of bandwidth is usually applied to a communication channel and is defined as the maximum amount of information transmitted/received per unit of time.

Bandwidth is one of the most important factors from the user's point of view. It is estimated by the amount of data that the network, in the limit, can transfer per unit of time from one device connected to it to another.

The speed of information transfer depends largely on the speed of its creation (source performance), encoding and decoding methods. The highest possible information transfer rate in a given channel is called its bandwidth. The bandwidth of the channel, by definition, is

information transfer rate when using the “best” (optimal) source, encoder and decoder for a given channel, therefore it characterizes only the channel.

5. In what units is the throughput of information transmission channels measured?

It can be measured in various, sometimes highly specialized, units - pieces, bits / sec, tons, Cubic Meters etc.

6. Classification of computer communication channels (by coding method, by communication method, by signal transmission method)

broadcast networks; networks with transmission from node to node.

7. Characteristics of cable channels for information transmission (coaxial cable, twisted pair, telephone cable, fiber optic cable)

wired - telephone, telegraph (air) communication lines; cable - copper twisted pairs, coaxial, fiber optic;

and also on the basis of electromagnetic radiation:

terrestrial and satellite communications; based on infrared rays.

cables based on twisted (twisted) pairs of copper wires; coaxial cables (central core and copper braid); fiber optic cables.

Based cables twisted pair

Twisted-pair cables are used to transmit digital data and are widely used in computer networks. It is also possible to use them for the transmission of analog signals. Twisting wires reduces the influence of external interference on useful signals and reduces the electromagnetic waves radiated to the outside. Shielding increases the cost of the cable, complicates installation and requires high-quality grounding. On fig. a typical UTP design based on two twisted pairs is presented.

Rice. Unprotected twisted pair cable design.

Depending on the presence of protection - an electrically grounded copper braid or aluminum foil around twisted pairs, the types of cables based on twisted pairs are determined:

unprotected twisted UTP pair(Unshielded twisted pair) - absent protective screen around a separate pair;

foil twisted pair FTP (Foiled twisted pair) - there is one common external shield in the form of foil;

shielded twisted pair STP (Shielded twisted pair) - there is a protective screen for each pair and a common external screen in the form of a grid;

foil shielded twisted pair S / FTP (Screened Foiled twisted pair) - there is a protective shield for each pair in a foil braid and an external shield made of copper braid;

unprotected shielded twisted pair SF/UTP (Screened Foiled Unshielded twisted pair) - double outer shield made of copper braid and foil, each twisted pair is unprotected.

1.5.2.2. Coaxial cable

Purpose coaxial cable– signal transmission in various fields of technology: communication systems; broadcast networks; computer networks; antenna-feeder systems of communication equipment, etc. This type of cable has an asymmetrical design and consists of an inner copper core and a braid separated from the core by a layer of insulation.

A typical design of a coaxial cable is shown in Fig. 1.22.

Rice. 1.22. Typical coaxial cable design

Thanks to the metal shielding braid, it has high noise immunity. The main advantage of coax over twisted pair is its wide bandwidth, which provides potentially higher data rates than twisted pair cables, which are up to 500 Mbps. In addition, the coaxial provides significantly greater allowable signal transmission distances (up to a kilometer), it is more difficult to mechanically connect to it for unauthorized listening to the network, and it also pollutes the environment much less electromagnetic radiation. However, installation and repair of coaxial cable is more difficult than twisted pair, and the cost is higher.

Conventional LED transceivers are used here, which reduces the cost and increases the service life compared to single-mode cable. In Figure 1.24. the characteristic of attenuation of signals in an optical fiber is given. Compared to other types of cables used for communication lines, this type of cable has significantly lower signal attenuation values, which are usually in the range of 0.2 to 5 dB per 1000 m of length. Multimode fiber is characterized by attenuation transparency windows in the wavelength ranges 380-850, 850-1310 (nm), and single-mode fiber, respectively, 850-1310, 1310-1550 (nm).

Figure 1.24. Optical fiber transparency windows.

Advantages of fiber optic communication type:

Wide bandwidth.

Extremely conditioned high frequency carrier vibration. When applying the technology of spectral multiplexing of communication channels by the method of wave

multiplexing in 2009, the signals of 155 communication channels with a transmission rate of 100 Gbit / s in each were transmitted over a distance of 7000 kilometers. Thus, the total data transmission rate over the fiber was 15.5 Tbps. (Tera = 1000 Giga);

Low attenuation of the light signal in the fiber.

Allows you to build fiber-optic communication lines of great length without intermediate amplification of signals;

Low noise level in the fiber optic cable.

Allows you to increase the bandwidth by transmitting various signal modulations with low code redundancy;

High noise immunity and security from unauthorized access.

Provided by the absolute security of the fiber from electrical interference, interference and the complete absence of radiation to the external environment. This is due to the nature of the light vibration, which does not interact with electromagnetic fields of other frequency ranges, like the fiber itself, which is a dielectric. Using a number of properties of light propagation in optical fiber, optical link integrity monitoring systems can instantly disable a “hacked” communication link and raise an alarm. Such systems are especially needed when creating communication lines in government, banking and some other special services that place high demands on data protection;

No need for galvanic isolation of network nodes.

Fiber optic networks fundamentally cannot have electrical "ground" loops, which occur when two network devices have grounds at different points in a building;

 High explosion and fire safety, resistance to aggressive environments.

Due to the absence of the possibility of sparking, optical fiber increases the safety of the network in chemical, oil refineries, maintenance technological processes increased risk;

 Light weight, volume, cost-effective fiber optic cable.

The basis of the fiber is quartz (silicon dioxide), which is a widely used inexpensive material. Currently, the cost of fiber versus copper pair is 2:5. The cost of the fiber optic cable itself is constantly decreasing, however, the use of special optical receivers and transmitters (fiber optic modems), which convert light signals into electrical signals and vice versa, significantly increases the cost of the network as a whole;

 Long service life.

The service life of optical fiber is at least 25 years. Fiber optic cable also has some disadvantages. The main one is the high complexity of installation. When connecting the ends of the cable, it is necessary to ensure the high accuracy of the cross-section of the glass fiber, the subsequent polishing of the cut and the centering of the glass fiber when installed in the connector. Connectors are installed by seam welding or by gluing using a special gel that has the same light refractive index as fiberglass. In any case, this requires highly qualified personnel and special tools. In addition, fiber optic cable is less durable and less flexible than electrical cable, and is sensitive to mechanical stress. It is also sensitive to ionizing radiation, due to which the transparency of glass fiber decreases, that is, the signal attenuation in the cable increases. Sudden temperature changes can cause fiberglass to crack. To reduce the influence of these factors, various design solutions are used, which affects the cost of the cable.

Given the unique properties of optical fiber, telecommunication based on it is increasingly being used in all areas of technology. These are computer networks, city, regional, federal, as well as intercontinental underwater primary communication networks, and much more. With the help of fiber-optic communication channels, the following are carried out: cable TV, remote video surveillance, video conferencing and video broadcasting, telemetry and other information systems.

8. Characteristics of wireless information transmission channels (satellite,

radio channels, Wi-Fi, Bluetooth)

Wireless technologies- subclass information technologies, serve to transmit information over a distance between two or more points, without requiring their connection by wires. can be used to convey informationinfrared radiation, radio waves , optical or laser radiation.

Many wireless technologies currently exist, most commonly known to users by their marketing names such as Wi-Fi, WiMAX, Bluetooth. Each technology has certain characteristics that determine its scope.

There are various approaches to the classification of wireless technologies.

By range:

o Wireless personal networks ( WPAN - Wireless Personal Area Networks). Technology examples are Bluetooth.

o Wireless local networks ( WLAN - Wireless Local Area Networks).

Technology examples are Wi-Fi.

o City-wide wireless networks ( WMAN - Wireless Metropolitan Area Networks). Technology examples are WiMAX.

o Wireless global networks ( WWAN - Wireless Wide Area Network).

Technology examples are CSD, GPRS, EDGE, EV-DO, HSPA.

Topology:

o Point-to-Point.

o Point-to-multipoint.

By area of ​​application:

o Corporate (departmental) wireless networks- created by companies for their own needs.

o Operator wireless networks - created by telecom operators for paid provision of services.

A short but concise way to classify is to display two of the most significant characteristics of wireless technologies simultaneously on two axes: maximum speed transmission of information and maximum distance.

Tasks Task 1 . For 10 s, 500 bytes of information were transmitted over the communication channel. What is equal to

channel capacity? (500/10=50 bytes/s=400bps)

Task 2 . How much information can be transmitted over a channel with a bandwidth of 10 kbps in 1 minute? (10 kbps*60 s = 600 kbps)

Problem 3. The average data transfer rate using a modem is 36864 bps. How many seconds will the modem need to transmit 4 pages of text in KOI-8 encoding, if we assume that each page has an average of 2304 characters.

Solution: Number of characters in the text: 2304*4 = 9216 characters.

In the KOI-8 encoding, each character is encoded with one byte, then the information volume of the text is 9216 * 8 = 73,728 bits.

Time = volume / speed. 73728: 36864 = 2 s

In modern IP networks, with the emergence of many new network applications, it becomes increasingly difficult to estimate the required bandwidth: as a rule, you need to know which applications you plan to use, what data transfer protocols they use, and how they will communicate

Ilya Nazarov
System engineer of "INTELCOM line" company

After evaluating the required bandwidth on each of the sections of the IP network, it is necessary to decide on the choice of technologies for the OSI network and link layers. In accordance with the selected technologies, the most suitable models of network equipment are determined. This question is also not easy, since throughput directly depends on the performance of the hardware, and performance, in turn, depends on the software and hardware architecture. Let us consider in more detail the criteria and methods for assessing the throughput of channels and equipment in IP networks.

Throughput Evaluation Criteria

Since the advent of teletraffic theory, many methods have been developed to calculate channel capacities. However, unlike the calculation methods applied to circuit-switched networks, the calculation of the required throughput in packet networks is quite complex and is unlikely to provide accurate results. First of all, this is due to a huge number of factors (especially inherent in modern multiservice networks), which are quite difficult to predict. In IP networks, a common infrastructure is typically shared by multiple applications, each of which may use its own distinct traffic model. Moreover, within one session, the traffic transmitted in the forward direction may differ from the traffic passing in the opposite direction. In addition, calculations are complicated by the fact that the speed of traffic between individual network nodes can change. Therefore, in most cases, when building networks, the assessment of throughput is actually determined by the general recommendations of manufacturers, statistical studies and the experience of other organizations.

In order to more or less accurately determine how much bandwidth is required for the network being designed, it is necessary first of all to know which applications will be used. Further, for each application, it is necessary to analyze how the data transfer will take place during the selected periods of time, which protocols are used for this.

For a simple example, consider the applications of a small corporate network.

Throughput Calculation Example

Suppose there are 300 working computers and the same number of IP phones on the network. It is planned to use the following services: e-mail, IP-telephony, video surveillance (Fig. 1). For video surveillance, 20 cameras are used, from which video streams are transmitted to the server. Let's try to estimate what maximum bandwidth is required for all services on the channels between the network core switches and at the junctions with each of the servers.

It should be noted right away that all calculations must be carried out for the time of the greatest network activity of users (in the theory of teletraffic - CNN, peak hours), since usually during such periods network performance is most important and the resulting delays and application failures associated with lack of bandwidth , are unacceptable. In organizations, the highest load on the network may occur, for example, at the end of the reporting period or during the seasonal influx of customers, when the most phone calls are made and most mail messages are sent.

Email
Returning to our example, consider an email service. It uses protocols that run on top of TCP, that is, the data transfer rate is constantly adjusted, trying to take up all the available bandwidth. Thus, we will start from the maximum value of the delay in sending a message - suppose 1 second will be enough for the user to be comfortable. Next, you need to estimate the average volume of the message sent. Let's assume that during the peaks of activity, mail messages will often contain various attachments (copies of invoices, reports, etc.), so for our example, we will take the average message size of 500 kb. And finally, the last parameter that we need to choose is the maximum number of employees who simultaneously send messages. Let's say that half of the employees simultaneously press the "Send" button in the email client during the rush job. The required maximum throughput for email traffic would then be (500 kb x 150 hosts)/1 s = 75,000 kb/s or 600 Mbps. From this we can immediately conclude that to connect the mail server to the network, you must use a Gigabit Ethernet channel. In the core of the network, this value will be one of the terms that make up the total required bandwidth.

Telephony and video surveillance
Other applications - telephony and video surveillance - are similar in their streaming structure: both types of traffic are transmitted using the UDP protocol and have a more or less fixed transmission rate. The main differences are that for telephony the streams are bidirectional and limited by the call time, for video surveillance the streams are transmitted in one direction and, as a rule, are continuous.

To estimate the required throughput for telephony traffic, let's assume that during the peaks of activity the number of simultaneous connections passing through the gateway can reach 100. When using the G.711 codec in Ethernet networks, the speed of a single stream, including headers and service packets, is approximately 100 kbps. With. Thus, during periods of the highest user activity, the required bandwidth in the network core will be 10 Mbps.

Video surveillance traffic is calculated quite simply and accurately. Suppose, in our case, video cameras transmit streams of 4 Mbps each. The required bandwidth will be equal to the sum of the speeds of all video streams: 4 Mbps x 20 cameras = 80 Mbps.

As a result, it remains to add the obtained peak values ​​​​for each of the network services: 600 + 10 + 80 = 690 Mbps. This will be the required bandwidth in the network core. The design should also consider the possibility of scaling so that the communication links can serve the traffic of a growing network for as long as possible. In our example, the use of Gigabit Ethernet will be enough to satisfy the requirements of the services and at the same time be able to seamlessly grow the network by connecting more nodes.

Of course, the given example is far from being a reference one - each case must be considered separately. In reality, the network topology can be much more complex (Fig. 2), and throughput needs to be estimated for each of the network sections.

It should be taken into account that VoIP traffic (IP telephony) is distributed not only from phones to the server, but also directly between phones. In addition, network activity can vary across departments in an organization: the help desk makes more phone calls, the project department uses email more than others, the engineering department consumes Internet traffic more than others, and so on. As a result, some parts of the network may require more bandwidth than others.

Useful and total bandwidth

In our example, when calculating the IP telephony stream rate, we took into account the codec used and the size of the packet header. This is an important detail to keep in mind. Depending on the encoding method (codecs used), the amount of data transmitted in each packet, and the link layer protocols used, the total bandwidth of the stream is formed. It is the total bandwidth that should be taken into account when estimating the required network bandwidth. This is most relevant for IP telephony and other applications that use low-rate real-time streaming, in which the size of the packet headers is a significant fraction of the packet size. For clarity, let's compare two VoIP streams (see table). These streams use the same compression, but different payload sizes (the actual digital audio stream) and different link layer protocols.

The data transfer rate in its purest form, without taking into account the headers of network protocols (in our case, the digital audio stream), is a useful bandwidth. As can be seen from the table, with the same useful throughput of streams, their total throughput can vary greatly. Thus, when calculating the required network bandwidth for telephone calls during peak loads, especially for telecom operators, the choice of channel protocols and flow parameters plays a significant role.

Equipment selection

The choice of link-layer protocols is usually not a problem (today the question is more often about how much bandwidth an Ethernet channel should have), but even an experienced engineer can find it difficult to choose the right equipment.

The development of network technologies along with the growing demands of applications for network bandwidth is forcing manufacturers of network equipment to develop new software and hardware architectures. Often, at a single manufacturer, there are at first glance similar models of equipment, but designed to solve different network problems. Take, for example, Ethernet switches: along with the usual switches used in enterprises, most manufacturers have switches for building storage networks, for organizing operator services, etc. Models of the same price category differ in their architecture, "sharpened" for certain tasks.

In addition to overall performance, the choice of hardware should also be driven by the supported technologies. Depending on the type of equipment, a certain set of functions and types of traffic can be processed at the hardware level, without using CPU and memory resources. In this case, the traffic of other applications will be processed at the software level, which greatly reduces the overall performance and, as a result, the maximum throughput. For example, multilayer switches, due to their sophisticated hardware architecture, are capable of transmitting IP packets without performance degradation when all ports are fully loaded. Moreover, if we want to use more complex encapsulation (GRE, MPLS), then such switches (at least inexpensive models) are unlikely to suit us, since their architecture does not support the corresponding protocols, and at best such encapsulation will occur at the expense of the central processor low performance. Therefore, to solve such problems, one can consider, for example, routers whose architecture is based on a high-performance central processor and depends more on software than hardware implementation. In this case, at the expense of maximum throughput, we get a huge set of supported protocols and technologies that are not supported by switches of the same price category.

Overall equipment performance

In the documentation for their equipment, manufacturers often indicate two values ​​\u200b\u200bof the maximum throughput: one is expressed in packets per second, the other in bits per second. This is due to the fact that most of the performance of network equipment is usually spent on processing packet headers. Roughly speaking, the equipment must accept the packet, find a suitable switching path for it, form a new header (if necessary) and pass it on. Obviously, in this case, it is not the amount of data transmitted per unit of time that plays a role, but the number of packets.

If we compare two streams transmitted at the same rate but with different packet sizes, then the stream with a smaller packet size will require more performance. This fact should be taken into account if the network is supposed to use, for example, a large number of IP telephony streams - the maximum throughput in bits per second here will be much less than the declared one.

It is clear that with mixed traffic, and even taking into account additional services (NAT, VPN), as is the case in the vast majority of cases, it is very difficult to calculate the load on equipment resources. Often, equipment manufacturers or their partners perform load testing of different models under different conditions and publish the results on the Internet in the form of comparative tables. Familiarization with these results greatly simplifies the task of choosing an appropriate model.

Pitfalls of modular equipment

If the selected network equipment is modular, then, in addition to the flexible configuration and scalability promised by the manufacturer, you can get a lot of "pitfalls".

When choosing modules, you should carefully read their description or consult the manufacturer. It is not enough to be guided only by the type of interfaces and their number - you also need to familiarize yourself with the architecture of the module itself. It is not uncommon for similar modules when, when transmitting traffic, some are able to process packets autonomously, while others simply forward packets to the central processing unit for further processing (accordingly, for the same external modules, the price for them can vary several times). In the first case, the overall performance of the equipment and, as a result, its maximum throughput are higher than in the second, since the central processor shifts part of its work to the module processors.

In addition, modular equipment often has a blocking architecture (when the maximum throughput is lower than the total speed of all ports). This is due to the limited bandwidth of the internal bus through which the modules exchange traffic between themselves. For example, if a modular switch has a 20 Gbps internal bus, then only 20 ports can be used for its line card with 48 Gigabit Ethernet ports when fully loaded. Such details should also be kept in mind and when choosing equipment, carefully read the documentation.

When designing IP networks, bandwidth is a key parameter that will affect the architecture of the network as a whole. For a more accurate estimate of throughput, you can follow these guidelines:

1. Study the applications that you plan to use on the network, the technologies they use, and the volumes of transmitted traffic. Use the advice of developers and the experience of colleagues to take into account all the nuances of these applications when building networks.
2. Learn in detail the network protocols and technologies used by these applications.
3. Read the documentation carefully when choosing equipment. To have some stock of ready-made solutions, check out the product lines of different manufacturers.

As a result, with the right choice of technologies and equipment, you can be sure that the network will fully meet the requirements of all applications and, being flexible and scalable enough, will last for a long time.

throughput - important parameter for any pipes, channels and other heirs of the Roman aqueduct. However, the throughput is not always indicated on the pipe packaging (or on the product itself). In addition, it also depends on the pipeline scheme how much liquid the pipe passes through the section. How to correctly calculate the throughput of pipelines?

Methods for calculating the throughput of pipelines

There are several methods for calculating this parameter, each of which is suitable for a particular case. Some notations that are important in determining the throughput of a pipe:

Outer diameter - the physical size of the pipe section from one edge of the outer wall to the other. In calculations, it is designated as Dn or Dn. This parameter is indicated in the marking.

Nominal diameter is the approximate value of the diameter of the internal section of the pipe, rounded up to a whole number. In calculations, it is designated as Du or Du.

Physical methods for calculating the throughput of pipes

Pipe throughput values ​​are determined by special formulas. For each type of product - for gas, water supply, sewerage - the methods of calculation are different.

Tabular calculation methods

There is a table of approximate values ​​\u200b\u200bcreated to facilitate the determination of the throughput of pipes for intra-apartment wiring. In most cases high accuracy is not required, so the values ​​can be applied without complex calculations. But this table does not take into account the decrease in throughput due to the appearance of sedimentary growths inside the pipe, which is typical for old highways.

There is an exact capacity calculation table, called the Shevelev table, which takes into account the pipe material and many other factors. These tables are rarely used when laying water pipes around the apartment, but in a private house with several non-standard risers they can come in handy.

Calculation using programs

At the disposal of modern plumbing companies there are special computer programs for calculating the throughput of pipes, as well as many other similar parameters. In addition, online calculators have been developed that, although less accurate, are free and do not require installation on a PC. One of the stationary programs "TAScope" is a creation of Western engineers, which is shareware. Large companies use "Hydrosystem" - this is a domestic program that calculates pipes according to criteria that affect their operation in the regions of the Russian Federation. In addition to hydraulic calculation, it allows you to calculate other parameters of pipelines. The average price is 150,000 rubles.

How to calculate the throughput of a gas pipe

Gas is one of the most difficult materials to transport, in particular because it tends to compress and therefore can flow through the smallest gaps in pipes. To the calculation of the throughput of gas pipes (as well as to the design gas system in general) have special requirements.

The formula for calculating the throughput of a gas pipe

The maximum capacity of gas pipelines is determined by the formula:

Qmax = 0.67 DN2 * p

where p is equal to the working pressure in the gas pipeline system + 0.10 MPa or the absolute pressure of the gas;

Du - conditional passage of the pipe.

Exists complex formula to calculate the throughput of a gas pipe. When carrying out preliminary calculations, as well as when calculating a domestic gas pipeline, it is usually not used.

Qmax = 196.386 Du2 * p/z*T

where z is the compressibility factor;

T is the temperature of the transported gas, K;

According to this formula, the direct dependence of the temperature of the transported medium on pressure is determined. The higher the T value, the more the gas expands and presses against the walls. Therefore, when calculating large highways, engineers take into account possible weather conditions in the area where the pipeline passes. If the nominal value of the DN pipe is less than the gas pressure generated at high temperatures in summer (for example, at + 38 ... + 45 degrees Celsius), then the line is likely to be damaged. This entails the leakage of valuable raw materials, and creates the possibility of an explosion of the pipe section.

Table of capacities of gas pipes depending on pressure

There is a table for calculating the throughput of a gas pipeline for commonly used diameters and nominal working pressure of pipes. To determine the characteristics of the gas pipeline custom sizes and pressure will require engineering calculations. Also, the pressure, speed of movement and volume of gas is affected by the temperature of the outside air.

The maximum velocity (W) of the gas in the table is 25 m/s and z (compressibility factor) is 1. The temperature (T) is 20 degrees Celsius or 293 Kelvin.

Capacity of the sewer pipe

The capacity of the sewer pipe is an important parameter that depends on the type of pipeline (pressure or non-pressure). The calculation formula is based on the laws of hydraulics. In addition to the laborious calculation, tables are used to determine the capacity of the sewer.

For the hydraulic calculation of sewerage, it is required to determine the unknowns:

1. pipeline diameter Du;
2. average flow velocity v;
3. hydraulic slope l;
4. degree of filling h / Du (in calculations, they are repelled from the hydraulic radius, which is associated with this value).

In practice, they are limited to calculating the value of l or h / d, since the remaining parameters are easy to calculate. The hydraulic slope in preliminary calculations is considered to be equal to the slope of the earth's surface, at which the movement of wastewater will not be lower than the self-cleaning speed. The speed values ​​as well as the maximum h/Dn values ​​for domestic networks can be found in Table 3.

Yulia Petrichenko, expert

In addition, there is a normalized value for the minimum slope for pipes with a small diameter: 150 mm

(i=0.008) and 200 (i=0.007) mm.

The formula for the volumetric flow rate of a liquid looks like this:

where a is the free area of ​​the flow,

v is the flow velocity, m/s.

The speed is calculated by the formula:

where R is the hydraulic radius;

C is the wetting coefficient;

From this we can derive the formula for the hydraulic slope:

According to it, this parameter is determined if calculation is necessary.

where n is the roughness factor, ranging from 0.012 to 0.015 depending on the pipe material.

The hydraulic radius is considered equal to the usual radius, but only when the pipe is completely filled. In other cases, use the formula:

where A is the area of ​​the transverse fluid flow,

P is the wetted perimeter, or the transverse length of the inner surface of the pipe that touches the liquid.

Capacity tables for non-pressure sewer pipes

The table takes into account all the parameters used to perform the hydraulic calculation. The data is selected according to the value of the pipe diameter and substituted into the formula. Here, the volumetric flow rate q of the liquid passing through the pipe section has already been calculated, which can be taken as the throughput of the pipeline.

In addition, there are more detailed Lukin tables containing ready-made throughput values ​​for pipes of different diameters from 50 to 2000 mm.

Capacity tables for pressurized sewer systems

In the capacity tables for sewer pressure pipes, the values ​​depend on the maximum degree of filling and the estimated average flow rate of the waste water.

Capacity of the water pipe

Water pipes in the house are used most often. And since they are subjected to a large load, the calculation of the throughput of the water main becomes an important condition for reliable operation.

Passability of the pipe depending on the diameter

Diameter is not the most important parameter when calculating pipe patency, but it also affects its value. The larger the inner diameter of the pipe, the higher the permeability, as well as the lower the chance of blockages and plugs. However, in addition to the diameter, it is necessary to take into account the coefficient of friction of water on the pipe walls (table value for each material), the length of the line and the difference in fluid pressure at the inlet and outlet. In addition, the number of bends and fittings in the pipeline will greatly affect the patency.

Table of pipe capacity by coolant temperature

The higher the temperature in the pipe, the lower its capacity, as the water expands and thus creates additional friction. For plumbing, this is not important, but in heating systems it is a key parameter.

There is a table for calculations of heat and coolant.

Pipe capacity table depending on the coolant pressure

There is a table describing the throughput of pipes depending on the pressure.

Pipe capacity table depending on diameter (according to Shevelev)

The tables of F.A. and A.F. Shevelev are one of the most accurate tabular methods for calculating the throughput of a water supply system. In addition, they contain all the necessary calculation formulas for each specific material. This is a voluminous informative material used by hydraulic engineers most often.

The tables take into account:

1. pipe diameters - internal and external;
2. wall thickness;
3. service life of the pipeline;
4. line length;
5. pipe assignment.

Hydraulic Calculation Formula

For water pipes, the following calculation formula applies:

Online calculator: pipe capacity calculation

If you have any questions, or if you have any guides that use methods not mentioned here, write in the comments.

Bandwidth

Bandwidth- a metric characteristic showing the ratio of the maximum number of passing units (information, objects, volume) per unit of time through a channel, system, node.

Used in various fields:

• in communications and computer science PS - the maximum achievable amount of passing information;
• in transport P. S. - the number of units of transport;
• in mechanical engineering - the volume of passing air (oils, lubricants).

It can be measured in various, sometimes highly specialized, units - pieces, bit / sec, tons, cubic meters, etc.

In computer science, the definition of bandwidth is usually applied to a communication channel and is defined as the maximum amount of information transmitted or received per unit of time.
Bandwidth is one of the most important factors from the user's point of view. It is estimated by the amount of data that the network, in the limit, can transfer per unit of time from one device connected to it to another.

Channel capacity

The highest possible information transfer rate in a given channel is called its bandwidth. The channel capacity is the information transfer rate when using the “best” (optimal) source, encoder and decoder for a given channel, therefore it characterizes only the channel.

Bandwidth of a discrete (digital) channel without interference

C = log(m) bits/symbol

where m is the base of the signal code used in the channel. Information transfer rate in discrete channel without noise (ideal channel) is equal to its capacity, when the characters in the channel are independent, and all m characters of the alphabet are equally likely (used equally often).

Neural Network Bandwidth

The bandwidth of the neural network is the arithmetic mean between the volumes of processed and generated information neural network per unit of time.

see also

• List of bandwidths of data interfaces

Wikimedia Foundation. 2010 .

• Gareev, Musa Gaisinovich
• Borkolabovskaya Icon of the Mother of God

See what "Bandwidth" is in other dictionaries:

Bandwidth- water flow through the spillway fittings with an unflooded outlet funnel. Source: GOST 23289 94: Sanitary weir fittings. Specifications original doc … Dictionary-reference book of terms of normative and technical documentation

Bandwidth- the total amount of oil products that can be pumped through the pipeline (through the terminal) per unit of time. Storage capacity of a tank (tank farm) is the total amount of oil products that can be stored in ... ... Financial vocabulary

throughput- Weight flow of the working medium through the valve. [GOST R 12.2.085 2002] Capacity KV Fluid flow rate (m3/h), with a density equal to 1000 kg/m3, passed by the regulatory body at a pressure drop of 1 kgf/cm2. Note. Current… … Technical Translator's Handbook

Bandwidth- the maximum amount of information that can be processed per unit of time, measured in bits / s ... Psychological Dictionary

throughput- performance, power, output, capacity Dictionary of Russian synonyms ... Synonym dictionary

Bandwidth- - see maintenance mechanism ... Economic and Mathematical Dictionary

throughput- Category. Ergonomic characteristic. Specificity. Maximum amount information that can be processed per unit of time, measured in bits/s. Psychological dictionary. THEM. Kondakov. 2000... Great Psychological Encyclopedia

throughput- The maximum number of vehicles that can pass on a given section of the road in a specific time ... Geography Dictionary

CAPACITY- (1) roads the largest number of ground transport units (million pairs of trains) that this road can pass per unit of time (hour, day); (2) P. s. communication channel maximum error-free transmission rate (cm.) this channel… … Great Polytechnic Encyclopedia

CAPACITY- the highest data transfer rate of the equipment with which information enters the storage device without loss while maintaining the sampling rate and analog-to-digital conversion. for devices with architecture on a parallel bus throughput ... ... Glossary of concepts and terms formulated in the normative documents of Russian legislation