弹性负载均衡好在哪，讲解6种负载均衡算法

弹性负载均衡（Elastic Load Balance）是将访问流量根据转发策略分发到后端多台弹性云服务器的流量分发控制服务，可以通过流量分发扩展应用系统对外的服务能力，提高应用程序的容错能力。弹性负载均衡具有以下优势：

1.高性能

集群支持最高1亿并发连接，满足用户的海量业务访问需求。

2.高可用

采用集群化部署，支持多可用区的同城双活容灾，无缝实时切换。

图1 高可用               

3.灵活扩展

根据应用流量自动完成分发，与弹性伸缩服务无缝集成，灵活扩展用户应用的对外服务能力。

图2 灵活扩展               

4.简单易用

快速部署ELB，实时生效，支持多种协议、多种调度算法可选，用户可以高效地管理和调整分发策略。

增强型负载均衡算法，支持以下三种调度算法：

加权轮询算法：根据后端服务器的权重，按顺序依次将请求分发给不同的服务器。它用相应的权重表示服务器的处理性能，按照权重的高低以及轮询方式将请求分配给各服务器，相同权重的服务器处理相同数目的连接数。常用于短连接服务，例如HTTP等服务。

加权最少连接：最少连接是通过当前活跃的连接数来估计服务器负载情况的一种动态调度算法。加权最少连接就是在最少连接数的基础上，根据服务器的不同处理能力，给每个服务器分配不同的权重，使其能够接受相应权值数的服务请求。常用于长连接服务，例如数据库连接等服务。

源IP算法：将请求的源IP地址进行Hash运算，得到一个具体的数值，同时对后端服务器进行编号，按照运算结果将请求分发到对应编号的服务器上。这可以使得对不同源IP的访问进行负载分发，同时使得同一个客户端IP的请求始终被派发至某特定的服务器。该方式适合负载均衡无cookie功能的TCP协议。

经典型负载均衡算法，支持以下三种调度算法：

轮询算法：按顺序把每个新的连接请求分配给下一个服务器，最终把所有请求平分给所有的服务器。常用于短连接服务，例如HTTP等服务。

最少连接：通过当前活跃的连接数来估计服务器负载情况的一种动态调度算法，系统把新的连接请求分配给当前连接数目最少的服务器。常用于长连接服务，例如数据库连接等服务。

源IP算法：将请求的源IP地址进行Hash运算，得到一个具体的数值，同时对后端服务器进行编号，按照运算结果将请求分发到对应编号的服务器上。这可以使得对不同源IP的访问进行负载分发，同时使得同一个客户端IP的请求始终被派发至某特定的服务器。该方式适合负载均衡无cookie功能的TCP协议。

Elastic load balance is a traffic distribution control service that distributes the access traffic to multiple elastic cloud servers at the back end according to the forwarding policy. It can expand the external service capability of the application system through traffic distribution, and improve the fault tolerance capability of the application program. Elastic load balancing has the following advantages:

1. high performance

The cluster supports up to 100 million concurrent connections to meet users' massive business access needs.

2. high availability

The cluster deployment is adopted to support the disaster recovery of the same city and double activities in multiple zones, and seamless real-time switching.

Figure 1 high availability

3. Flexible expansion

According to the application traffic, it can automatically complete the distribution, integrate with the elastic scaling service seamlessly, and expand the external service ability of the user application flexibly.

Figure 2 flexible expansion

4. Easy to use

Rapid deployment of ELB, real-time effective, support for a variety of protocols, a variety of scheduling algorithm options, users can effectively manage and adjust the distribution strategy.

The enhanced load balancing algorithm supports the following three scheduling algorithms:

Weighted polling algorithm: according to the weight of back-end servers, requests are distributed to different servers in order. It uses the corresponding weight to represent the processing performance of the server. According to the weight and polling method, requests are allocated to each server. The servers with the same weight process the same number of connections. Commonly used for short connection services, such as HTTP and other services.

Weighted least connection: least connection is a dynamic scheduling algorithm to estimate the server load by the number of active connections. Weighted least connection is to assign different weights to each server based on the minimum number of connections, according to the different processing capabilities of the server, so that it can accept the service request with the corresponding number of weights. It is often used for long connection services, such as database connection and other services.

Source IP algorithm: hash the requested source IP address to get a specific value, number the back-end server, and distribute the request to the corresponding numbered server according to the calculation result. This can make the access of different source IP be load distributed, and at the same time make the request of the same client IP always be distributed to a specific server. This method is suitable for TCP protocol without cookie function.

The classic load balancing algorithm supports the following three scheduling algorithms:

Polling algorithm: allocate each new connection request to the next server in order, and finally distribute all requests equally to all servers. Commonly used for short connection services, such as HTTP and other services.

Minimum connections: a dynamic scheduling algorithm that estimates the server load by the number of currently active connections. The system allocates new connection requests to the servers with the minimum number of connections. It is often used for long connection services, such as database connection and other services.

Source IP algorithm: hash the requested source IP address to get a specific value, number the back-end server, and distribute the request to the corresponding numbered server according to the calculation result. This can make the access of different source IP be load distributed, and at the same time make the request of the same client IP always be distributed to a specific server. This method is suitable for TCP protocol without cookie function.

--------------------------------------------------------------
主机参考，收集国内外VPS，VPS测评，主机测评，云服务器，虚拟主机，独立服务器，国内外服务器，高性价比建站主机相关优惠信息@zhujicankao.com
详细介绍和测评国外VPS主机,云服务器,国外服务器,国外主机的相关优惠信息,商家背景,网络带宽等等,也是目前国内最好的主机云服务器VPS参考测评资讯优惠信息分享平台

这几篇文章你可能也喜欢：

• 1G的云服务器可以做什么？好不好
• WNT虫洞传输实现内网安全、高速、稳定传输
• 云服务器ecs怎么选购，有哪些技巧
• WNT高速直连公有云、私有云，解决丢包，降低延迟
• Exchange 2013安装部署及补充

本文由主机参考刊发，转载请注明：弹性负载均衡好在哪，讲解6种负载均衡算法 https://zhujicankao.com/12721.html