Table of Contents
Advertisement
S9700 Core Routing Switch
Configuration Guide - SPU
As shown in
the internal network to multiple links. One ISP gateway corresponds to one link. The egress link
load balancing process is as follows:
1.
2.
3.
The egress link load balancing supported by the SPU has the following characteristics:
l
l
l
l
Server Load Balancing
As the Internet and services develop fast, network access data increases rapidly. The access
traffic to data centers, large enterprises, and portal web sites even reaches 10 Gbit/s. In addition,
servers use HTTP, FTP, and SMTP to provide a large amount of information for access users.
The servers are gradually flooded by data. Besides, some web sites especially e-commerce web
sites provide the around-clock non-stop service. Any service interruption or key data loss will
result in commercial loss. These require high performance and high reliability for application
services.
With the development of network technologies, the server processing speed and memory access
speed cannot meet the increasing requirements of network bandwidth and application service.
The network bandwidth increase brings in the increase of users, whereas server resources are
seriously consumed. The servers become the network bottleneck. Simply upgrading server
hardware is expensive and short sighted.
The dynamic load balancing algorithm enables server load balancing to properly allocate
network services to servers in a server group. This reduces the burden of a single server and
improves server reliability. You only need to add a server to a server group, without changing
network structure or stopping services.
In server load balancing, the forwarding mode is DNAT or DMAC. The process of the two
modes is the same. The SPU provides a virtual IP address. After users request services through
the virtual IP address, the SPU allocates the requests to real servers according to the load
balancing algorithm. In DNAT mode, when allocating service requests, the SPU translates the
Issue 01 (2012-03-15)
Figure
6-1, Switch is the load balancing device and distributes the traffic sent from
Users on the internal network send requests to servers on the external network.
When the request packets pass through Switch, Switch selects a link according to the
configured load balancing algorithm, weights, priorities, and inbound/outbound bandwidth
limits, and forwards the request packets to the selected link.
After receiving the response packets of servers on the external network, Switch forwards
the response packets to the users on the internal network.
Load balancing algorithm
It supports the WRR algorithm, least connection algorithm, and IP address-based hash
algorithm.
Link bandwidth threshold
The link bandwidth threshold can be set for the incoming or outgoing traffic. When the
bandwidth threshold or bandwidth percentage on an ISP link exceeds the limit, the load
balancing device does not select the ISP link.
Link health detection
After an Internet Control Message Protocol (ICMP) probe is configured on the SPU, the
SPU periodically sends probing packets to the link gateway to detect the connectivity
between nodes along the link.
Forwarding mode
The forwarding mode for server load balancing is DNAT or DMAC. In egress link load
balancing, the SPU supports the redirection mode.
Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.
6 Load Balancing Configuration
185
Advertisement
Table of Contents
