No relevant resource is found in the selected language.

This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy>Search


To have a better experience, please upgrade your IE browser.


Configuration Guide - Network Management and Monitoring

CloudEngine 8800, 7800, 6800, and 5800 V200R005C10

This document describes the configurations of Network Management and Monitoring, including SNMP, RMON, NETCONF, OpenFlow, LLDP, NQA, Mirroring, Packet Capture, Packet Trace, Path and Connectivity Detection Configuration, NetStream, sFlow, and iPCA.
Rate and give feedback:
Huawei uses machine translation combined with human proofreading to translate this document to different languages in order to help you better understand the content of this document. Note: Even the most advanced machine translation cannot match the quality of professional translators. Huawei shall not bear any responsibility for translation accuracy and it is recommended that you refer to the English document (a link for which has been provided).
End-to-End Performance Measurement

End-to-End Performance Measurement

IP data communication is a mainstream communication method. It provides various access methods, supports large network size, and allows timely and accurate network fault locating, facilitating network Service Level Agreement (SLA) evaluation. Based on IP FPM, iPCA implements end-to-end network performance measurement.

End-to-end performance measurement is applicable to the following scenarios:
  • Packet loss and delay measurement for a specialized service flow, such as a voice flow and a video flow, on an enterprise network
  • Packet loss and delay measurement on a WAN when an enterprise has multiple networks connected through the WAN or on an enterprise campus network consisting of devices that do not support iPCA
  • Network routine maintenance performance and SLA monitoring
In end-to-end network performance measurement, continual or on-demand performance measurement needs to be configured, as shown in Table 16-2.
Table 16-2 End-to-end performance measurement



End-to-end continual measurement

To detect network performance degradation in a timely manner, use this mode. This mode displays continual network running status.

End-to-end on-demand measurement

If you detect that network performance degrades or you want to monitor the performance of a specified service flow, use this mode. This mode displays detailed performance statistics in recent time.

End-to-End Packet Loss Measurement

End-to-end networks include point-to-point, multipoint-to-multipoint, and across-WAN networks. Configure iPCA on two ends of the network to perform end-to-end packet loss measurement, including the number of lost packets and packet loss rate, for an entire link or network.

Point-to-point packet loss measurement

In Figure 16-9, the target flow enters the network through switch 1 and leaves the network through switch 2. The target flow enters the network through one interface and leaves through another interface. After IP FPM packet loss measurement is configured on switch 1 and switch 2, the packet loss statistics on the network can be collected.

Figure 16-9 Point-to-point network

Multipoint-to-multipoint packet loss measurement

In Figure 16-10, all Huawei devices form a multipoint-to-multipoint network. The ingress and egress interfaces of the target flow are located on multiple edge devices. The number of all incoming packets on the network should be the same as the number of all outgoing packets. After IP FPM packet loss measurement is configured on network edge devices switch 1, switch 2, switch 3, and switch 4, the packet loss statistics on the network can be collected.

Figure 16-10 Multipoint-to-multipoint network

Across-WAN packet loss measurement

In Figure 16-11, an enterprise leases WAN links from a carrier to construct a regional network. To measure the performance of the regional network, IP FPM can be configured on the Huawei egress devices on the branch and headquarters networks. All traffic sent by egress devices to the WAN is the input traffic of the regional network, and all traffic received by Huawei devices from the WAN is the output traffic of the regional network. If the input traffic volume is higher than the output traffic volume, packet loss occurs on the regional network.

Figure 16-11 Across-WAN network

VXLAN packet loss measurement

Virtual eXtensible Local Area Network (VXLAN) uses the MAC in UDP encapsulation method to build a Layer 3 virtual network, and is widely used in the scenarios such as public cloud, private cloud, campus, and carrier network. However, the VXLAN packet encapsulation has a large size. This restricts network maintenance and makes end-to-end packet loss, delay, and jitter measurement difficult. To monitor VXLAN tunnel performance, you can configure both IP FPM and VXLAN. That is, configure IP FPM in the VXLAN tunnel to monitor the tunnel running status. This facilitates fault locating.

On a VXLAN shown in Figure 16-12, VXLAN tunnels are set up between Spine and Leaf 1, Spine and Leaf 2, and Leaf 1 and Leaf 2. VXLAN packets are used on the VXLAN tunnels, and IP packets are used on other paths.

After IP FPM is configured on two ends of a VXLAN flow, the packet loss statistics on the VXLAN tunnel can be collected. For example, after interfaces B and C are selected as TLPs, packet loss statistics on the VXLAN tunnel between Spine and Leaf2 are collected. After interfaces A and C are selected as TLPs, packet loss statistics on the VXLAN tunnel between Leaf1 and Leaf2 are collected. The user-side interfaces on Leaf1 and Leaf2 process only IP packets, but not VXLAN packets, so these interfaces cannot be configured as TLPs. (If they are configured as TLPs, only the statistics on IP packets are collected.)

Figure 16-12 VNI-based VXLAN performance measurement
  • Both ends of a VXLAN tunnel must be configured as the TLPs when you collect packet statistics on this VXLAN tunnel.

  • A measurement instance cannot cover different IP or VXLAN networks.

End-to-End Delay Measurement

By periodically coloring packets, iPCA can measure forwarding delays (unidirectional and bidirectional) in addition to measuring the lost packets. iPCA delay measurement is applicable to the following scenarios:

Delay measurement cannot be applied to point-to-multipoint, multipoint-to-point, and multipoint-to-multipoint networks.

For a unidirectional service flow, you can measure the unidirectional forwarding delay of service packets between two devices.

For bidirectional service flows, you can measure the round trip delay of packets between two devices.

Updated: 2019-04-20

Document ID: EDOC1100075365

Views: 41403

Downloads: 127

Average rating:
This Document Applies to these Products
Related Version
Related Documents
Previous Next