Build a home network. What is smart in smart network switches? What is a hub, switch and router? Select the switch - read the characteristics.

The choice of the router to be used is determined by the Ethernet interfaces that match the switches in the LAN center. It is important to note that routers offer a variety of services and functions for LAN.

Each LAN has a router used as a gateway to connect LAN with other networks. LAN has one or more hubs or switches to connect the end devices with LAN.

Routers are the main devices used to connect networks. Each port on the router is connected to a different network and sends packets between networks. Routers can smoke broadcast domains and domains of collisions.

Routers are also used to connect networks that use different technologies. They may have LAN, and WAN interfaces.

LAN interfaces routers allow them to connect to LAN carriers. Usually, UTP cable connections, but modules can be added to use fiber optics . Depending on the series or model of routers, they can have several types of interfaces for WAN and LAN cable connections.

Devices intraseni.

To create a LAN, we must select the appropriate devices to connect the end nodes with the network. The two most common devices used are hubs and switches.

Concentrator

The hub receives a signal, regenerates it and sends to all ports. The use of hubs creates a logical bus. This means that LAN uses a media media mode. The ports use the bandwidth sharing approach, which often leads to a decrease in performance in LAN due to collisions and recovery. Although you can connect several hubs, the only collision domain will still remain.

Hubs are less expensive than switches. The hub is usually selected as an intermediary device for a very small LAN, which has low capacity requirements, or with limited finance.

Switch

The switch takes the frame and regenerates each frame of the frame to the appropriate destination port. This device is used to segmented the network into several collision domains. Unlike the hub, the switch reduces the number of conflicts in the LAN. Each port on the switch creates a separate domain of collisions. This creates a dot-point logical topology for the device on each port. In addition, the switch provides a highlighted bandwidth on each port, which can increase the LAN performance. The LAN switch can also be used to connect network segments with different speeds.

In general, switches are selected to connect devices to LAN. Although the switch is more expensive than the hub, its improved performance and reliability make it cost-effective.

There is a whole range of available switches with multiple functions that allow you to connect a plurality of computers to a typical installation of the LAN enterprise.

The Ethernet logic topology is a multiple access bus in which all devices use shared access to the same data transfer environment. This logical topology determines how nodes on the network are viewed and processed frames sent and obtained in this network. However, at present, almost all Ethernet networks use a physical topology type "Star" or "Extended Star". This means that in most Ethernet networks, the terminal devices are usually connected to the LAN level 2 switch on the "point-to-point" principle.

The LAN switch 2 switch switches and filter only on the basis of the MAC address of the channel layer of the OSI model. The switch is completely transparent to network protocols and user applications. Level 2 switch creates a MAC address table, which in the future uses for making solutions to ship packages. In the process of transmitting data between independent IP subnets, level 2 switches rely on routers.

Switches use MAC addresses to transfer data over the network through its switching matrix to the corresponding port in the direction of the destination node. The commuting matrix is \u200b\u200bintegrated channels and complementary machine programming tools, which allows you to monitor data paths through the switch. In order for the switch to understand which port must be used to transfer a unicast mailing list, first it needs to know which nodes are available on each of its ports.

The switch determines the method of processing incoming frames using its own table of MAC addresses for this. It creates its own table of MAC addresses by adding the MAC address of the nodes that are connected to each of its ports. After making the MAC address for a node connected to a specific port, the switch will be able to send the traffic intended for this node through the port that is mapped to the node for subsequent gear.

If the switch receives a data frame for which the table does not have the MAC address of the destination, it sends this frame to all ports, except in which this frame was accepted. If a response is received from the destination node, the switch makes the MAC address of the node into the addresses table using the data from the address source address field. In networks with multiple connected switches, several MAC addresses connecting switches that reflect the elements outside the node are included in the MAC address tables. Typically, the switch ports used to connect two switches have several MAC addresses entered in the appropriate table.

In the past, switches were used by one of the following ways to switch data between network ports:

Switching with buffering

Switchless buffering

When switching with buffering when the switch receives a frame, it stores data in the buffer until the entire frame is obtained. During saving, the switch analyzes the frame to get information about its recipient. In this case, the switch also performs an error check using the end part of the Ethernet Ethernet Ethernet Control (CRC).

When using switching without buffering, the switch processes the data as they are received even if the transmission is not yet completed. The switch adds to the buffer exactly the amount of frame that is required to read the MAC address of the destination so that it can determine which port to send data. The destination MAC address is specified in 6 bytes of the frame after the preamble. The switch is looking for a MAC address of the destination in its switching table, determines the port of the outgoing interface and sends a frame to its destination node via a dedicated switch port. The switch does not verify the frame for any errors. Since the switch does not need to wait for the entire frame to the buffer all the entire frame, and at the same time it does not check errors, switching without buffering is faster than switching with buffering. However, since the switch does not check the errors, it forwards damaged frames throughout the network. When sending damaged frames reduce bandwidth. Ultimately, the Network Destination deflects damaged frames.

Modular switches offer greater configuration flexibility. As a rule, they are supplied with chassis of various sizes, which allows you to install multiple modular linear boards. Ports are actually located on linear boards. The line card is inserted into the switch chassis like expansion charges installed in the PC. The more chassis, the more modules it supports. As shown in the figure, a variety of different chassis sizes are offered to choose from. If you purchased a modular switch with a 24-port line card, you can easily install another such charge, with the result that the total number of ports will be increased to 48.

The switch is one of the most important devices used when building a local network. In this article we will talk to what switches are and focus on the important characteristics you need to consider when selecting a LAN switch.

To begin with, consider the general structural scheme to understand what place the switch takes on the local network of the enterprise.

Figure above shows the most common block diagram of a small local network. As a rule, access switchboards are used in such local networks.

Access switches are directly connected to end users, providing them with access to the resources of the local network.

However, in large local networks, the switches perform the following functions:

Network Access Level. As mentioned above, the access switches provide the point of connecting the end-user devices. In large local networks, the frame switcher switches do not interact with each other, and are transmitted through distribution switches.

Distribution level. Switches of this level are sent to traffic between access switching switches, but do not interact with end users.

System kernel level. The devices of this type combine data channels from the distribution level switches in large territorial local networks and provide a very high speed of switching data streams.

Switches are:

Unmanaged switches. These are the usual stand-alone devices on the local network that control the data transfer independently and do not have the ability to further adjust. In view of the simplicity of installation and small prices were widespread when installed at home and small business.

Controlled switches. More advanced and expensive devices. Allow the network administrator independently set them up for specified tasks.

Controlled switches can be configured in one of the following ways:

Through the console portVia Web interface

Through Telnet via SNMP protocol

Through ssh.

Switch levels

All switches can be divided into model levels.OSI . How this level is higher than the greatest possibilities of the switch possesses, however, it will be significantly higher than it.

Level 1 Switches (Layer 1). This level includes hubs, repeaters and other devices operating at the physical level. These devices were at the dawn of the development of the Internet and are currently not used on the local network. After receiving a device of this type, simply transmits it further, in all ports except the port of the sender

Level 2 switches (Layaer 2). This level includes uncontrollable and part of the managed switches (switch ) model working on the channel levelOSI . Second-level switches work with frames - frames: the flow of data broken into portions. Having received a frame 2 switch 2 deducts the sender's address from the frame and enters it into its tableMAC addresses, comparing this address of the port on which he received this frame. Thanks to this approach, the second-level switches are sent only to the recipient port, without creating redundant traffic by other ports. Second-level switches do not understandIP addresses located on the third network model modelOSI and work only on the channel level.

Second-level switches support the most common protocols as:

IEEE 802.1. q.or VLAN Virtual local networks. This protocol allows you to create separate logical networks within one physical network.

For example, devices connected to one switch, but are in differentVLAN. You will not see each other and transmit data only in your broadcast domain (devices from the same VLAN). Between themselves, computers in the figure above will be able to transmit data using a device operating at the third levelIP addresses: router.

IEEE. 802.1p (Priority Tags ). This protocol is initially present in the protocol.IEEE 802.1 Q. and represents a 3 bit field from 0 to 7. This protocol allows you to label and sort the entire traffic according to the degree of importance to expose priorities (maximum priority 7). Frames with a big priority will be sent primarily.

IEEE 802.1D Spanning Tree Protocol (STP).This protocol builds a local network in the form of a tree structure to avoid networking network and prevent the formation of a network storm.

Suppose the installation of the local network is made in the form of a ring to increase system fault tolerance. The switch with the highest priority on the network is selected root (root).In the example above, SW3 is root. Do not deepen in the protocol execution algorithms, the switches calculate the path with the maximum price and block it. For example, in our case, the crosted path from SW3 to SW1 and SW2 will be through its own selected interfaces (DP) Fa 0/1 and Fa 0/2. In this case, the price of the default path for the 100 Mbit / c interface will be 19. The FA 0/1 switch of the LAN switch is blocked because the general price of the path will be folded from two transitions between 100 Mbps with interfaces 19 + 19 \u003d 38.

If the working route is damaged, the switches will execute the path recalculation and unlock this port

IEEE 802.1W Rapid Spanning Tree Protocol (RSTP).Advanced standard 802.1.d. which has a higher stability and less communication lifting time.

IEEE 802.1s Multiple Spanning Tree Protocol.The latest version, taking into account all the shortcomings of the protocolsSTP and RSTP.

IEEE 802.3ad Link Aggregation for Parallel Link.This protocol allows you to combine ports to the group. The total speed of this port of the aggregation will be folded out of the amount of the speeds of each port in it.The maximum speed is determined by the IEEE 802.3ad standard and is 8 Gb / s.

Switches 3 levels (Layer 3). These devices are also called multiswichs as they combine the capabilities of switches operating at the second level and routers working withIP packages at the third level. Level 3 switches Fully support all functions and standards of level 2 switches. Network devices can work on IP addresses. The level switch 3 supports the installation of various connections:l 2 TP, PPTP, PPPOE, VPN, etc.

Layer Switches 4 (Layer 4) . L4 level devices working at the transport level modelOSI . Respond to ensuring the reliability of data transfer. These switches may, on the basis of information from the packet headers, understand the traffic belonging to different applications and make decisions on the redirection of such traffic on the basis of this information. The name of such devices was not subject to, sometimes they are called intelligent switches, or L4 switches.

Main characteristics of switches

Number of ports. Currently, there are switches with the number of ports from 5 to 48. The number of network devices that can be connected to this switch depends on this parameter.

For example, when building a small local network of 15 computers, we will need a switch with 16 ports: 15 to connect the terminal devices and one to install and connect the router to access the Internet.

Data transfer rate. This is the speed on which each switch port works. Typically, speeds are indicated as follows: 10/100/1000 Mbps. The speed of the port is determined in the process of auto agreement with the final device. In managed switches, this parameter can be configured manually.

For example : PC client device with a network board 1 Gb / s connected to the port of the switch at a speed of 10/100 Mbpsc. . As a result, the auto coordination of the device agrees to use the maximum possible speed of 100 Mbps.

Auto Port Agreement betweenFull - Duplex and Half - Duplex. Full - Duplex: data transmission is simultaneously carried out in two directions.Half - Duplex Data transmission is carried out first in one, then in the other direction sequentially.

The internal bandwidth of the switching matrix. This parameter shows from what common speed switch can process data from all ports.

For example: in the local network there is a switch in which 5 ports operating at a speed of 10/100 Mbps. In the specifications, the switching matrix parameter is 1 Gbit /c. . This means that each port is in modeFull - duplex can work at a speed of 200 Mbpsc. (100 Mbps Reception and 100 Mbps Transmission). Suppose the parameter of this switching matrix is \u200b\u200bless than specified. This means that at the time of peak loads, the ports will not be able to work with the claimed rate of 100 Mbps.

Auto Correction Type MDI / MDI-X Cable. This feature allows you to determine which of the two methods was compressed twisted steam EIA / TIA-568A or EIA / TIA-568B. When installing local networks, the EIA / TIA-568B scheme obtained the greatest distribution.

Stacking - This is a combination of several switches to one single logical device. Different switch manufacturers use their stacking technologies, such asc. iSCO uses Stack Wise Stack Wise technology with a bus between 32 Gb / s switches and Stack Wise Plus with a bus between Switches 64 Gb / s.

For example, this technology is relevant in large local networks, where it is required on the basis of one device to connect more than 48 ports.

Fasteners for 19 "Racks. At home and small local networks, the switches are often installed on smooth surfaces or attached to the wall, however, the presence of so-called "ears" is needed in larger local networks where the active equipment is placed in server cabinets.

MAC table sizeaddresses. Switch (Switch) This device operating at 2 modelsOSI . Unlike a hub, which simply redirects the received frame to all ports except the port of the sender, the switch is learning: remembersMAC sender's device address, enjoying it, port number and lifetime to the table. Using this table, the switch redirects the frame not to all ports, but only to the port of the recipient. If on the local network the number of network devices is considerably and the size of the table is full, the switch starts to rub the older entries in the table and writes new, which significantly reduces the speed of the switch.

Jumboframe. . This feature allows the switch to work with a large package size than is defined by the Ethernet standard. After receiving each package, spends some time on its processing. When using the increased size of the package using the Jumbo Frame technology, you can save on the packet processing time in networks where data transfer rates from 1 GB / s and above are used. With lower speed of big win no

Switching modes.In order to understand the principle of operation of switching modes, first consider the frame structure of the frame transmitted on the channel between the network device and the switch on the local network:

As can be seen from the picture:

• First there is a preamble signaling the start of the frame transfer,
• Then MAC destination address (DA) and Mac sender's address (SA)
• Identifier third level:IPv 4 or IPv 6 is used
PAYLOAD)
• And at the end checksumFCS: 4 CRC Speed \u200b\u200bValue Used to detect transmission errors. It is calculated by the sending side, and is placed in the FCS field. The receiving Party calculates this value independently and compares with the value obtained.

Now consider the switching modes:

Store - and - Forward. This switching mode saves the frame to the whole buffer and checks the fieldFCS. which is at the very end of the frame and if the checksum of this field does not coincide, discards the entire frame. As a result, the probability of overloading overloads is reduced, since it is possible to discard frames with an error and set off the packet transmission time. This technology is present in more expensive switches.

Cut -through. More simple technology. In this case, frames can be processed faster, as they are not saved in the buffer completely. For analysis, the buffer saves data from the start of the frame to the MAC destination address (DA) inclusive. The switch subtracts this MAC address and redirects it to the addressee. The disadvantage of this technology is that the switch is forwarding in this case both dwarf, less than 512 bitted intervals and damaged packets, increasing the load on the local network.

POE technology support

Pover Over Ethernet technology allows you to pave a network device on the same cable. This decision reduces the cash costs for additional installation of the supply lines.

There are the following PoE standards:

PoE 802.3AF supports equipment with a capacity of up to 15.4 W

PoE 802.3at supports equipment with a capacity of up to 30 W

Passiv PoE

POE 802.3 AF / AT has intelligent voltage control circuits to the device: Before applying for a power to the PoE device, the AF / AT source is negotiated to avoid damage to the device. Passiv PoE is significantly cheaper than the first two standards, the power is powered directly to the device via free pairs of the network cable without any coordination.

Characteristics of standards

The PoE 802.3AF standard is supported by most inexpensive IP video cameras, IP phones and access points.

The PoE 802.3at standard is present in the more expensive video surveillance IP camera models, where it is not possible to meet at 15.4 W. In this case, both the IP camcorder and the PoE source (switch) must support this standard.

Expansion slots. Switches may have additional expansion slots. The most common are SFP modules (Small Form-Factor Pluggable). Modular, compact transceivers are used to transmit data in a telecommunications environment.

SFP modules are inserted into the free SFP port of the router, switch, multiplexer or media converter. Although there are SFP Ethernet modules, most oftenfiber optic modules are used for the solar channel soil when transmitting data over long distances are not available to the Ethernet standard. SFP modules are selected depending on the distance, data rate. The most common are two-wheeled SFP modules using one fiber for receiving, another for data transmission. However, WDM technology allows data transmission on different wavelengths by one optical cable.

SFP modules are:

• SX - 850 Nm is used with multimode optical cable at a distance of up to 550m
• LX - 1310 Nm is used with both types of optical cable (SM and MM) at a distance of up to 10 km
• BX - 1310/1550 Nm Used with both types of optical cable (SM and MM) at a distance of up to 10 km
• XD - 1550 nm is used with a single-mode cable to 40km, ZX to 80km, EZ or EZX to 120 km and DWDM

The SFP standard itself provides for data transmission at a rate of 1Gbit / s, or at a speed of 100 Mbps. For faster data transfer, SFP + modules were developed:

• SFP + data transmission with a speed of 10 Gb / s
• XFP data transmission with a speed of 10 Gb / s
• QSFP + data transmission with a speed of 40 Gb / s
• CFP data transmission with a speed of 100 Gb / s

However, at higher speeds, signals are processed at high frequencies. This requires a larger heat sink and, accordingly, large dimensions. Therefore, in fact, the SFP form factor has been preserved only in SFP + modules.

Conclusion

Many readers probably faced uncontrollable switches and budget managed second-level switches in small local networks. However, the choice of switches for building larger and technically complex local networks is better to provide professionals.

Safe Kuban When installing local networks uses switches of the following brands:

Professional solution:

Cisco.

QTech

Budget decision

D-Link

TP-LINK.

Tenda.

Safe Kuban performs installation, commissioning and maintenance of local networks in Krasnodar and the south of Russia.

The issues of building local networks are represented by non-specialist users are very complex due to an extensive terminological dictionary. Habs and switches are drawn in the imagination with complex equipment resembling telephone PBXs, and the creation of a local home network becomes a reason for accessing specialists. In fact, the Svitch is not so worried as its name: both devices are elementary network nodes that have minimal functionality that do not require installation and operational knowledge and quite accessible to everyone.

Definition

Hub - Network hub, designed to combine computers to a single local network by connecting Ethernet cables.

Svitche (Switch - Switch) - a network switch designed to combine into a local network of multiple computers via an Ethernet interface.

Comparison

As we see from the definition, the difference between the hub and the switch is associated with the type of devices: a hub and switch. Despite one task - the organization of the local network via Ethernet is to solve it in different ways. The hub is the simplest splitter, providing a direct connection between network clients. Svitch - more "smart" device that distribute data packets between clients in accordance with the request.

Hub, receiving a signal from one node, transmits it to all connected devices, and the receipt of the whole depends on the destination: the computer must recognize, whether it is designed for a package. Naturally, the answer implies the same scheme. The signal comes in all segments of the network until it finds the one that will accept it. This circumstance reduces network bandwidth (and data exchange rate, respectively). Svitch, receiving a data packet from a computer, directs it according to the address that was set by the sender, eliminating the network from the load. The network, organized by the switch, is considered more secure: the exchange of traffic takes place directly between the two clients, and others cannot process the signal, designed not to them. Unlike a hub, the switch provides a high bandwidth of the created network.

Hub Logitec LAN-SW / PS

The switch requires proper configuration of the client's computer network card: the IP address and subnet mask must match each other (the subnet mask specifies the part of the IP address as the network address, and the other part is as the client addresses). Hub settings does not require, because it works on the physical level of the OSI network model, translating the signal. The switch works at the channel level, carrying out the exchange of data packets. Another feature of the hub is the adjustment of nodes with respect to the transfer rate, focusing on the lowest indicators.

Switch Compex PS2208B.

Conclusions Site

1. Hub - hub, switch - switch.
2. Hub device is the simplest, switch - more "intellectual".
3. The hub transmits a signal to all network clients, switch - only addressee.
4. The network performance organized through the switch is higher.
5. Svortch provides a higher level of data transfer.
6. The hub works on the physical level of the OSI network model, the switch on the channel.
7. The switch requires the correct setup of network client network cards.