Detailed description of router performance indicators

This table mainly compares whether the vro is a modular structure. Vrouters with a modular structure are generally highly scalable and support multiple port types, such as Ethernet interfaces, Fast Ethernet interfaces, and high-speed serial ports. The number of ports of various types is generally optional. It is usually expensive. Vrouters with fixed configurations have poor scalability and are only used for ports of fixed type and quantity, which is generally cheaper.

Router configuration

Interface Type

Lists the types of interfaces supported by routers, reflecting the versatility of routers. Common interfaces include: common serial interfaces are converted to RS 232 DTE/DCE interfaces, V.35 DTE/DCE interfaces, X.21 DTE/DCE interfaces, RS 449 DTE/DCE interfaces, and EIA530 DTE interfaces through cables) 10 M Ethernet interface, Fast Ethernet interface, 10/100 adaptive Ethernet interface, Gigabit Ethernet interface, ATM interface 2 M, 25 M, 155 M, 633M, etc), POS interfaces 155 M, 622M, etc.), card ring interfaces, FDDI interfaces, E1/T1 interfaces, E3/T3 interfaces, and ISDN interfaces.

Number of available user Slots

This indicator refers to the number of slots that can be used by users except CPU board, clock Board, and other necessary system board and/or dedicated slot of system board in a modular router. The maximum number of ports supported by the vro is calculated based on this indicator and the port density of the user board.

CPU

CPU is the heart of a router, both in the Middle-end router and in the high-end router. In low-end and Middle-end routers, the CPU is responsible for exchanging route information, querying route tables, and forwarding data packets. In the preceding vro, the CPU capability directly affects the router's throughput route table query time) and the route computing capability affects the network route convergence time ). In high-end routers, packet forwarding and look-up are usually completed by the ASIC chip, and the CPU only implements the routing protocol, computing routes, and distribution route tables. Due to the development of technology, many work in routers can be implemented by hardware dedicated chips ). The CPU performance does not fully reflect the router performance. The performance of a vro。 is determined by indicators such as the throughput, latency, and route computing capability of the vro.

Memory

A vro may contain multiple types of memory, such as Flash and DRAM. The memory is used for storage configuration, router operating system, and routing protocol software. In the middle and low-end routers, the route table may be stored in the memory. Generally, the larger the vro memory, the better, the lower the price ). However, similar to CPU capabilities, the memory does not directly reflect the performance and capabilities of the router. Because efficient algorithms and excellent software may greatly save memory.

Port Density

This indicator reflects the integration of vro production. Because the vro volume is different, this indicator should be equivalent to the number of ports per inch in the rack. However, for the sake of intuition and convenience, the maximum number of ports supported by the router can be used.

Route Protocol Support

Route information protocol (RIP)

RIP is a distance vector-based routing protocol, which usually uses the number of hops as the Metering Standard. RIP is an internal gateway protocol. Due to its simple implementation, RIP is the most widely used routing protocol. This Protocol is slow to converge and is generally used in small-sized networks. The RIP Protocol is defined in RFC 1058.

Route Information Protocol Version 2 (r00002)

This protocol is an improved version of RIP, which allows more information to be carried and is compatible with RIP. Added the support for CIDR), next hop address, and optional authentication information based on RIP. This version is standardized in RFC 1723.

Open Shortest Path Priority Protocol Version 2 (OSPFv2)

This protocol is a link-based routing protocol. It is developed by the IETF Internal Gateway Protocol Working Group for IP addresses and serves as a successor internal gateway protocol for RIP. OSPF is used for least-cost routing, multi-same path computing, and load balancing. OSPF features openness and SPF algorithm.

"Intermediate System-intermediate system" Protocol (ISIS)

The ISIS protocol is also a link-based routing protocol. This protocol was proposed by ISO and was initially used in the OSI network environment and then modified to be able to run in a dual environment. Similar to OSPF, this protocol can be used as an internal gateway protocol for large-scale IP networks.

Edge Gateway Protocol (BGP4)

BGP is an Inter-Domain Routing protocol used to replace EGP. BGP4 is currently the most popular and only optional Inter-Domain Routing Protocol on the IP network. This version of the protocol supports CIDR, and can use the Routing aggregation mechanism to greatly reduce the route table. The BGP4 protocol can use multiple attributes to flexibly control routing policies.

802.3. 802.1Q support

802.3 is the IEEE Standard for Ethernet. Vrouters that support Ethernet interfaces must comply with the 802.3 protocol. 802.1Q is the IEEE Standard for virtual networks. Vrovlan interfaces conforming to 802.1Q can support multiple VLANs on the same physical interface.

IPv6 support

In the future, the IP network may be an IPv6 network. Due to the explosive development of the Internet caused by the emergence of the Web, the number of users of the IP network rapidly increased, and the IP address was unprecedentedly tight, so IPv6 was proposed. IPv6 expands the address space and adds authentication and encryption security measures at the IP layer. It defines Flow labels for real-time business applications ). However, due to the huge inertia of the market and the effective application of classless addressing (CIDR), the IP address depletion time has been greatly postponed. IPv6 has not been widely used yet. However, with the increase of services and the further development of the Internet, IPv6 is inevitable.

Support for protocols other than IP addresses

In addition to the IP protocol, router devices can also support IPX, DECNet, AppleTalk, and other protocols. These protocols have some application in foreign countries and are rarely used in China.
Supports source address routing and transparent bridging

Address routing means that when a router selects a route for a data packet, it selects a route based on the source address of the IP packet instead of the destination address of the IP packet. Source Address routing is a type of policy routing. Generally, vrouters should support this function. Transparent bridging means that the router port operates in the form of a transparent bridge and implements the bridge function. No route check and forwarding is performed for data packets, and only MAC frame bridging is performed.

Policy Routing Mode

In addition to using the destination address as the basis for route selection, the router can also select a path for data packets based on the TOS field, source, and destination port number high-level application protocol. Policy routing can implement traffic engineering to a certain extent, so that different service quality streams, data voices of different types, and FTP can go through different paths.

PPP and MLPPP

The PPP protocol is an important protocol in Internet protocols. Earlier networks were connected by routers using the PPP protocol to point-to-point connections, and most users used PPP access. Therefore, all vrouters with serial ports should support the PPP protocol and act as the first choice. MLPPP refers to bundling multiple PPP links.

PPPOE support

PPP Over Ethernet is a new protocol used to authenticate and charge Ethernet Access Users. Similar to the PPP Over ATM protocol, a router device using this protocol can terminate the access service. Currently, the PPPOE and PPPOA protocols have capacity problems. Most vrouters that support this Protocol can only process thousands of active sessions.

Supports the list protocol for multicast)

Internet connection Group Management Protocol (IGMP)

IGMPInternet Group Management Protocol) is an IP host used to report Group members to an adjacent multi-view router. A Multicast router sends an IGMP query to a local network. The host sends an IGMP report to respond to the query. The multicast router is responsible for forwarding multicast packets to all multicast members in the network.

Distance Vector Multicast Routing Protocol (DVMRP)

DVMRP is a multicast routing protocol based on distance vectors, which is basically developed based on RIP. DVMRP uses IGMP to exchange route data packets with neighbors. Protocol-Independent Multicast Protocol (PIM)

PIM is a multicast transmission protocol that can transmit multicast data over existing IP addresses. PIM is a multicast protocol independent of the routing protocol. It can work in two modes: intensive mode and loose mode. In PIM intensive mode, the packet group forwards packets to all ports by default until the packets are cut and removed. In dense mode, if the devices on all ports are multicast members, multicast packets may be used. The loose mode is opposite to the dense mode. Only multicast data is sent to the requested port.

VPN support

The VPN on the IP address is described in the router technology above. Possible protocols include L2TP, GRE, IP Over IP, and IPSec. The VPN support capability should also be concerned.

Encryption Method

The router may use the encryptor mechanism in VPN implementation or other conditions to ensure security. A router uses a CPU to execute software algorithms, which usually affects forwarding efficiency. Some routers adopt hardware encryption to improve forwarding efficiency.

MPLS

MPLS technology is described in the router technology above. In addition to label switching, MPLS also includes advanced applications such as fast rerunning, VPN in MPLS, and traffic engineering. As the MPLS standard is not yet mature, we should also pay attention to MPLS intercommunication.

Router performance

Full-duplex line rate forwarding capability

The most basic and important function of a router is packet forwarding. Forwarding packets at the same port rate is the greatest test of the router packet forwarding capability. Full-duplex line rate Forwarding is based on the minimum packet length Ethernet 64 bytes and POS port 40 bytes. The minimum packet interval complies with the protocol.) bidirectional transmission on the router port does not cause packet loss. This indicator is an important indicator of vro performance.

Device Throughput

The packet forwarding capability of the entire device, which is an important indicator of the device performance. The router selects routes based on IP headers or MPLS labels, so the performance indicator is the number of forwarding packets per second. The device throughput is generally less than the sum of the throughput of all ports on the vro.

Port throughput

Port throughput refers to the port packet forwarding capability, which is usually measured by pps: Packets per second. It is the packet forwarding capability of the router on a port. Generally, two interfaces with the same rate are used for testing. However, the test interface may be related to the interface location and relationship. For example, the throughput tested between ports on the same plug-in card may be different from the throughput value between ports on different plug-in cards.

Number of back-to-back frames

The number of back-to-back frames refers to the number of data packets when the maximum number of data packets sent at the minimum frame interval does not cause packet loss. This indicator is used to test the router cache capability. Vrouters with full-duplex forwarding capability of Wired speed have an unlimited value.

Route table capability

A Router usually depends on the route table created and maintained to determine how to forward data. The route table capability refers to the maximum number of route table entries in a route table. Because the number of routers that execute the BGP protocol on the Internet usually has hundreds of thousands of Route entries, this project is also an important embodiment of the router capability.

Backplane capability

The backplane capability is the internal implementation of the router. The backplane capability can be reflected in the router throughput: The backplane capability is generally greater than the value calculated based on the throughput and test package location. However, the backplane capability can only be reflected in the design and cannot be tested.

Packet Loss Rate

Packet Loss Rate refers to the ratio of the number of lost data packets in the test to the number of sent data packets, which is usually tested within the throughput range. Packet loss rate is related to the packet length and packet sending frequency. In some environments, route jitter and a large number of routes can be added for testing.

Latency

Latency refers to the time interval between the first bit of a data packet entering the router and the last bit output from the router. In testing, a test instrument is usually used to issue a test package to the time interval at which the data packet is received. Latency is related to the length of data packets. Generally, it is tested within the throughput range of the router port. If the throughput is exceeded, this indicator is meaningless.

Latency Jitter

Latency jitter refers to latency changes. The data service is not sensitive to latency jitter, so this indicator is not used in the Benchmarking test. This indicator is necessary only when multiple services on the IP address, including voice and video services, appear.

VPN support

Generally, vrouters support VPN. The performance difference is generally reflected in the number of VPN supported. Vrouters generally support a large number of VPNs. No-fault working time

This indicator indicates the time when the device has no fault according to the statistical method. Generally, it cannot be tested. It can be calculated based on the fault-free working time of the main device or the working conditions of a large number of identical devices.

Internal clock precision

Vro interconnection with ATM ports for circuit simulation or POS ports usually needs to be synchronized. If an internal clock is used, its precision will affect the bit error rate. For the definition of the internal clock precision level and the test method, see the corresponding synchronization standard.

QoS capability

Queue Management Mechanism

The queue management control mechanism usually refers to the router congestion management mechanism and queue scheduling algorithm. Common methods include RED, WRED, WRR, DRR, WFQ, and WF2Q.

Number of port hardware queues

Generally, the priority supported by routers is guaranteed by the port hardware queue. The priority of each queue is controlled by the queue scheduling algorithm.

QoS classification method

Information on which a router can differentiate QoS. The simplest QoS classification can be based on ports. Likewise, a router can distinguish packet priority based on the link layer priority 802.1Q), the upper layer content TOS field, source address, Destination Address, source port, destination port, and other information.

Classified service bandwidth guarantee

Whether the router can guarantee the bandwidth of various service levels. This indicator can be implemented by queue scheduling algorithms.

RSVP

RSVP is a Resource Reservation Protocol used to reserve resources on end-to-end paths. Refreshing with soft status is a stream-driven way of working. Generally, this Protocol cannot run on large-scale nationwide networks. However, routers usually support this protocol, which is used by some famous vendors for MPLS.

IP Diff Serv

Service differentiation is a classification of IP service quality and a simplification of QoS.

CAR support

A car is a type of access control that guarantees the access rate. According to the agreement signed with the user, the packet exceeding the promised rate is processed differently: discarded or marked; also known as the marked color.

Redundancy

Redundancy can include interface redundancy, plug-in Card Redundancy, power supply redundancy, system board redundancy, clock board redundancy, and equipment redundancy. Redundancy is used to ensure the reliability and availability of devices. Redundancy should be designed to compromise between equipment reliability requirements and investment.

Hot swapping Components

Because a vro usually requires 24 hours of work, changing parts should not affect the work of the vro. Hot swapping of components ensures the 24-hour operation of routers.

Router redundancy protocol

Vrouters can use VRRP and other protocols to ensure router redundancy.

Network Management

Network administrators perform centralized management of network resources through network management programs. Including configuration management, accounting management, performance management, error management, and security management. The degree of network management supported by devices reflects the manageability and maintainability of devices.

Web-based management

Indicates whether devices can be managed on the Web. Web management is more convenient, but the security is poor. Web browsing is usually allowed, and Web modification is not allowed.

Network management type

Indicates the types supported by network management. SNMP is usually used for management.

Support for out-of-band connections

The support of the Internet-Band Channel indicates whether the router can be managed through the out-of-band channel.

Network management granularity

Indicates the granularity of vro management: to manage ports, to CIDR blocks, to IP addresses, to MAC addresses, and so on. The management granularity may affect the router forwarding capability.

Billing capability/protocol

As a router enters the carrier network, billing becomes an essential part. A vro must support certain billing capabilities and protocols.

Group speech capability

Supported group speech

In enterprises, the voice bearer capability of vro groups is very important. Between the remote office and the headquarters, vrouters Supporting Group speech can integrate telephone communication and data communication, effectively saving long-distance calls.

In the current technical environment, group speech can be divided into three types: Using IP to carry group speech, using ATM to carry speech, and using frame relay to carry speech. When using ATM to carry voice, there are two types: AAL1 and AAL. AAL1 is circuit simulation. The technology is very mature, but the relative cost is high. The AAL technology is advanced, but the current ATM interface is usually not supported. Frame Relay speech is also relatively mature and relatively cost-effective. Currently, IP-bearer speech is popular. The above technology has the lowest cost, but it is difficult to ensure the QoS of the current IP network and ensure the quality of calls.

Protocol Support

H.323 is the ITU Standard and is the most common protocol stack for IP Phone networks. As an IETF standard, SIP aims to simplify network devices and implement complex functions on user terminals. From the perspective of the nature of the IP network, the router has nothing to do with the services it carries, but the support of the router port on the IP Phone protocol can save costs.

Speech compression capability

Voice compression is one of the key to saving costs for IP phones. G.723 and G.729 are usually used. G.723 is recommended in the ITU-T G.723.1 (1996), which is specified in the 5.3 and 6.3Kbps dual-rate voice encoder for multimedia communication transmission. Relatively high compression, large compression latency. G.729 is specified in the ITU-T recommendation G.729 (1996), 8 kbps co-structured representative digital incentive Linear Prediction (CS-ACELP) Speech Encoding. Low compression ratio and good call quality.

Port Density

The ability of a router to support IP phones. Generally, it is calculated as E1. Generally, an E1 instance supports 30 calls.

Signaling support

Vroe1 E1 port may support multiple types of signaling: ISUP, TUP, China 1 signaling, and dss1. Vrouters that support ISUP, TUP, or DSS1 signaling can effectively reduce the connection time. Generally, carrier-level IP phone network devices require support for signaling 7. However, as a low-end router, only DSS1 and China 1 signaling are supported.