The LVS of Linux load Balancing technology

I. Introduction of LVS
LVS is the short name of Linux virtual server, which is a free software project initiated by Dr. Zhangwensong, and its official site is www.linuxvirtualserver.org. Now that LVS is already part of the Linux standard kernel, the LVS must be recompiled to support the LVS function module before the Linux2.4 kernel is used. However, since the Linux2.4 kernel, the various functions of LVS have been built-in, without any patches to the kernel, which can be directly used by LVs .

lvs has been a relatively mature technology project since the beginning of 1998. For example: The Linux portal (www.linux.com), The real Company (Www.real.com) known for providing audio and video services to RealPlayer, The world's largest open source website (sourceforge.net).

II. structure of the LVS system
The server cluster system with LVS is composed of three parts: the most front-end load balancing layer ( one cluster, multiple servers). Multiple sets of words will be used to keepalived or heartbeat and other high-availability software , with load balancer, the middle of the server group layer, with server array, the bottom of the data sharing storage layer, with shared Storage said that in the eyes of users, all internal applications are transparent, users are only using a virtual server to provide high-performance services.
The LVS architecture is shown in 1:

 

? load Balancer Layer: Located at the forefront of the entire cluster system, There is one or more load scheduler (Director server), the LVS module is installed on the director server, The director's main role is similar to a router, which contains the completion of the LVS function set by the routing table, through these routing tables to distribute the user's request to the server array Layer application server (Real server). Also, on the director server, you install the Monitoring module Ldirectord for the real Server service, which is used to monitor the health status of each real Server service. The removes it from the LVS routing table when real server is unavailable and re-joins when it resumes.
? server array layer: Consists of a set of machines that actually run the application service, real server can be a Web server, a mail server, an FTP server , a DNS server, one or more of the video servers, and each real server is connected through a high-speed LAN or across a WAN. In a real-world application, Director server can also be the role of real server concurrently.

As can be seen from the entire LVS structure, director server is the core of the entire LVS, currently, the operating system for director server can only be Linux and FreeBSD, The linux2.6 kernel can support LVS without any setup, and FreeBSD as a director server is not a lot of applications, performance is not very good.
For real Server, almost all system platforms, Linux, Windows, Solaris, AIX, BSD series can be very well supported.

Three,   LVS cluster features
3.1  IP load Balancing and load scheduling algorithm

1. IP load Balancing Technology
lvs IP load Balancing technology is implemented by Ipvs module, Ipvs is the core software of LVS cluster system, its main function is: installed on the director server and virtual an IP address on the director server, the user must access the service through this virtual IP address. This virtual IP is generally called the LVS VIP, namely virtual IP. The requests that are accessed first go through the VIP to the load scheduler, and then the load Scheduler picks a service node from the real server list to respond to the user's request.

 
?  vs/nat: That is (virtual server via network address translation)

(Virtual Server via IP tunneling)  
therefore, in the Tun mode, the scheduler will only process the user's message request, the cluster system throughput greatly increased.

? VS/DR: that is (Virtual Server via Direct Routing)

That is, the use of direct routing technology to implement virtual server. Its connection scheduling and management is the same as in Vs/nat and Vs/tun, but its message forwarding method is different , vs/dr by overwriting the request message's MAC address, sends the request to real server, and real server returns the response directly to the customer, eliminating the VS The IP tunneling overhead in the/tun. This is the best performance in three load scheduling mechanisms, but it must be required that both the Director server and the real server have a NIC attached to the same physical network segment .

2. Load Scheduling algorithm
As we mentioned above, the load scheduler is based on the load situation of each server, dynamically select a real server to respond to user requests, then the dynamic selection is how to implement, in fact, we are here to say the load scheduling algorithm, according to different network service requirements and server configuration, Ipvs implements the following eight kinds of load scheduling algorithms, here we detail the most commonly used four scheduling algorithms, the remaining four scheduling algorithms please refer to other information.
round call scheduling (Round Robin)
"Round Call" dispatch is also called 1:1 scheduling, the scheduler through the "round call" scheduling algorithm to the external user request in order 1:1 to each real server in the cluster, the algorithm treats each real server equally, regardless of the actual load status and connection status on the server.
?weighted round call scheduling (Weighted Round Robin)
The "Weighted round call" scheduling algorithm dispatches access requests based on the different processing capabilities of real server. You can set different scheduling weights for each real server, and for a relatively good real server, you can set a higher weight, and for a less powerful real server, you can set a lower weight value, which ensures that the processing power of the server handles more traffic. The server resources are utilized fully and rationally. At the same time, the scheduler can automatically query the real server load situation, and dynamically adjust its weight value.
? Minimum link scheduling (Least Connections)
The "least connection" scheduling algorithm dynamically dispatches network requests to the server with the fewest number of established links. If the real server of the cluster system has similar system performance, the "Minimum connection" scheduling algorithm can be used to balance the load well.
? weighted minimum link scheduling (Weighted Least Connections)
"Weighted least link scheduling" is a superset of "least connection scheduling", each service node can use the corresponding weights to represent its processing power, and the system administrator can dynamically set the corresponding weights, the default weight is 1, the weighted minimum connection scheduling when allocating new connection requests as far as possible to make the service node's established connection number and its weight is proportional.
The other four scheduling algorithms are: local-based least-link (locality-based Least Connections), local-based least-link with replication (locality-based Least Connections with Replication), the destination address hash (Destination Hashing), and the source address hash (sources Hashing), the meaning of these four scheduling algorithms, this article no longer described, if you want to learn more about the remaining four scheduling strategies, You can log in to the LVS Chinese site zh.linuxvirtualserver.org for more detailed information.

3.2 High Availability
LVS is a kernel-level application software, so has high processing performance, with the LVS framework of load Balancing cluster system has excellent processing capacity, each service node failure will not affect the normal use of the entire system, while achieving a reasonable load balance, so that the application has an ultra-high load of service capacity, Millions of concurrent connection requests can be supported. If you configure a Gigabit network card, using Vs/tun or VS/DR scheduling technology, the entire cluster system throughput can be as high as 1gbits/s, such as the configuration of gigabit network cards, the maximum throughput of the system is close to 10gbits/s.

3.3 High Reliability
LVS load Balancing cluster software has been widely used in enterprises, schools and other industries, many large, critical Web sites have also adopted the LVS cluster software, so its reliability in practice has been well confirmed. There are a lot of LVS-made load-balancing systems that run for a long time and have never been restarted. These demonstrate the high stability and high reliability of LVS.

3.4 Applicable environment
LVS for front-end director server currently supports only Linux and FreeBSD systems, but supports most TCP and UDP protocols, and applications that support TCP protocols are: Http,https, Ftp,smtp,,pop3,imap4,proxy, Ldap,ssmtp and so on. The applications that support UDP protocol are: DNS,NTP,ICP, video, audio streaming protocol and so on.

LVS has no limitations on real server's operating system, and real server can run on any TCP/IP-enabled operating system, including Linux, Unix (such as FreeBSD, Sun Solaris, HP UNIX, etc.), mac/ OS, Windows, and more.

3.5 Open Source Software
The LVS cluster software is a free software issued under the GPL (GNU public License) license, so the user can obtain the source code of the software, and can make various changes according to their own needs, but the modification must be distributed in GPL mode.

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

The LVS of Linux load Balancing technology