13- clustering and load balancing

8/2/2019 13- Clustering and Load Balancing

1/33

Clustering and Load Balancing


2/33

Outline

Introduction

Linux Virtual Server

Microsoft load balancing solution


3/33

Introduction

Explosive Growth of the Internet

100% annual growth rate

Sites receiving unprecedented workloadYahoo! 625 million views per day

AOL Web cache system receiving 5 billion

requests per day


4/33

Introduction

load balancing is a technique to spread work between manycomputers, processes, disks or other resources in order to get optimalresource utilization and decrease computing time.

A load balancer can be used to increase the capacity of a server farm

beyond that of a single server. It can also allow the service to continue even in the face of server

down time due to server failure or server maintenance. A load balancer consists of a virtual server (also referred to as vserver

orVIP) which, in turn, consists of an IP address andport.

virtual serveris bound to a number of physical services running on the

physical servers in a server farm. A client sends a request to the virtual server, which in turn selects a

physical server in the server farm and directs this request to theselected physical server.
http://en.wikipedia.org/wiki/Server_farmhttp://en.wikipedia.org/wiki/Server_(computing)http://en.wikipedia.org/wiki/VIPhttp://en.wikipedia.org/wiki/IP_addresshttp://en.wikipedia.org/wiki/Porthttp://en.wikipedia.org/wiki/Virtual_serverhttp://en.wikipedia.org/wiki/Virtual_serverhttp://en.wikipedia.org/wiki/Porthttp://en.wikipedia.org/wiki/IP_addresshttp://en.wikipedia.org/wiki/VIPhttp://en.wikipedia.org/wiki/Server_(computing)http://en.wikipedia.org/wiki/Server_farm


5/33

Introduction (cont.)

Different virtual servers can be configured for different sets ofphysical services, such as TCP and UDP services in general.

Application specific virtual server may exist to support HTTP, FTP,SSL, DNS, etc.

The load balancing methods manage the selection of an appropriatephysical server in a server farm.

Persistence can be configured on a virtual server; once a server isselected, subsequent requests from the client are directed to the sameserver.

Persistence is sometimes necessary in applications where client state is

maintained on the server, but the use of persistence can causeproblems in failure and other situations.

A more common method of managing persistence is to store stateinformation in a shared database, which can be accessed by all realservers, and to link this information to a client with a small token suchas a cookie, which is sent in every client request.
http://en.wikipedia.org/wiki/Transmission_Control_Protocolhttp://en.wikipedia.org/wiki/User_Datagram_Protocolhttp://en.wikipedia.org/wiki/Statehttp://en.wikipedia.org/wiki/Statehttp://en.wikipedia.org/wiki/User_Datagram_Protocolhttp://en.wikipedia.org/wiki/Transmission_Control_Protocol


6/33


Load balancers also perform server monitoring of servicesin a web serverfarm.

case of failure of a service, the load balancer continues to

perform load balancing across the remaining services thatare UP.

In case of failure of all the servers bound to a virtual server,requests may be sent to a backup virtual server (ifconfigured) or optionally redirected to a configured URL.

In Global Server Load Balancing(GSLB) the load balancerdistributes load to a geographically distributed set of serverfarms based on health, server load or proximity.
http://en.wikipedia.org/wiki/Web_serverhttp://en.wikipedia.org/wiki/Web_serverhttp://en.wikipedia.org/wiki/Gslbhttp://en.wikipedia.org/wiki/Gslbhttp://en.wikipedia.org/wiki/Gslbhttp://en.wikipedia.org/wiki/Web_server


7/33


Load balancing methods: Least connections

Round robin

Least response time Leastbandwidth

Leastpackets

URLhashing

Domain name hashing

Source IP address

Destination IP address

Source IP - destination

Staticproximity, used for GSLB
http://en.wikipedia.org/wiki/Connectionhttp://en.wikipedia.org/wiki/Round_robinhttp://en.wikipedia.org/wiki/Response_timehttp://en.wikipedia.org/wiki/Bandwidthhttp://en.wikipedia.org/wiki/Packetshttp://en.wikipedia.org/wiki/URLhttp://en.wikipedia.org/wiki/Hashinghttp://en.wikipedia.org/wiki/Domainhttp://en.wikipedia.org/wiki/IP_addresshttp://en.wikipedia.org/wiki/Proximityhttp://en.wikipedia.org/wiki/Proximityhttp://en.wikipedia.org/wiki/IP_addresshttp://en.wikipedia.org/wiki/Domainhttp://en.wikipedia.org/wiki/Hashinghttp://en.wikipedia.org/wiki/URLhttp://en.wikipedia.org/wiki/Packetshttp://en.wikipedia.org/wiki/Bandwidthhttp://en.wikipedia.org/wiki/Response_timehttp://en.wikipedia.org/wiki/Round_robinhttp://en.wikipedia.org/wiki/Connection


8/33

Web Server Load Balancing

One major issue for large Internet sites is how to handle the load of thelarge number of visitors they get.

This is routinely encountered as a scalability problem as a site grows.

There are several ways to accomplish load balancing

For example in WikiMedia load is balanced as: Round robin DNS distributed page requests evenly to one of three Squid

Cache servers

Squid cache servers used response time measurements to distribute pagerequests between seven web servers.

In addition, the Squid servers cached pages and delivered about 75% of all

pages without ever asking a web server for help. The PHP scripts which run the web servers distribute load to one of several

database servers depending on the type of request, with updates going to amaster database server and some database queries going to one or moreslave database servers.


9/33

Server Load Balancing and

redundancy Alternative methods include use of Layer 4 Router

Linux virtual server, which is an advanced opensource load balancing solution for networkservices.

Network Load Balancing Services, which is anadvanced open source load balancing solution fornetwork services.

Many sites are turning to the multi-homed scenario;having multiple connections to the Internet viamultiple providers to provide a reliable and highthroughput service.


10/33

Linux Virtual Server

Started in 1998, the Linux Virtual Server (LVS) projectcombines multiple physical servers into one virtual server,eliminating single points of failure (SPOF).

Built with off-the-shelf components, LVS is already in usein some of the highest-trafficked sites on the Web.

Requirements for LVS: The service must scale: when the service workload increases, the

system must scale up to meet the requirements.

The service must always be on and available, despite transientpartial hardware and software failures.

The system must be cost-effective: the whole system must beeconomical to build and expand.

Although the whole system may be big in physical size, it should

be easy to manage.


11/33

LVS In LVS, a cluster of Linux servers appear as a single

(virtual) server on a single IP address.

Client applications interact with the cluster as if it were asingle, high-performance, and highly-available server.

Inside the virtual server, LVS directs incoming networkconnections to the different servers according to schedulingalgorithms.

Scalability is achieved by transparently adding or removingnodes in the cluster.

High availability is provided by detecting node or daemonfailures and reconfiguring the system accordingly, on-the-fly.

For transparency, scalability, availability and manageability,LVS is designed around a three-tier architecture, asillustrated in next figure


12/33

LVS architecture

The load balancer,servers, and sharedstorage are usually

connected by a high-speed network, suchas 100 Mbps Ethernetor Gigabit Ethernet,so that the

intranetwork does notbecome a bottleneckof the system as thecluster grows.


13/33

IPVS

IPVS modifies the TCP/IP stack inside the

Linux kernel to support IP load balancing

technologies


14/33

Three ways to balance load

IPVS supports following three ways to

balance loads:

Virtual Server via NAT(VS/NAT)

Virtual Server via Tunneling(VS/TUN)

Virtual Server via Direct Routing(VS/DR)


15/33

Virtual Server via NAT (VS/NAT)


16/33

VS/NAT Workflow1. When a user accesses a virtual service provided by the server

cluster, a request packet destined for the virtual IP address (the IPaddress to accept requests for virtual service) arrives at the load

balancer.

2. The load balancer examines the packet's destination address and

port number. If they match a virtual service in the virtual server ruletable, a real server is selected from the cluster by a schedulingalgorithm and the connection is added to hash table that recordsconnections. Then, the destination address and the port of the

packet are rewritten to those of the selected server, and the packet isforwarded to the server. When an incoming packet belongs to anestablished connection, the connection can be found in the hash

table and the packet is rewritten and forwarded to the right server.3. The request is processed by one of the physical servers.

4. When response packets come back, the load balancer rewrites thesource address and port of the packets to those of the virtualservice. When a connection terminates or timeouts, the connectionrecord is removed from the hash table.

5. A reply is sent back to the user.


17/33

An example of Virtual Server via Nat


18/33

Packet rewriting flow

The incoming packet for web service:

The load balancer will choose a real server andrewritten forwards the packet to it:

Replies get back to the load balancer:

The packet is rewritten and forwarded back to theclient


19/33

VS-NAT advantages and

disadvantages Advantages:

Real servers can run any OS that supports TCP/IP

Only an IP address is needed for the load balancer, realservers can use private IP address

Disadvantages The maximum number of server nodes is limited,

because both request and response packers are rewritten

by the load balancer. When the number of server nodesincrease up to 20, the load balancer will probably

become a new bottleneck


20/33

Virtual Server via IP Tunneling

(VS/TUN) IP tunneling(also calledIP encapsulation) is a

technique to encapsulate IP datagrams within IPdatagrams, which allows datagrams destined for

one IP address to be wrapped and redirected toanother IP address.

This technique can also be used to build a virtualserver: the load balancer tunnels the request packetsto the different servers, the servers process therequests, and return the results to the clientsdirectly. Thus, the service appears as a virtualservice on a single IP address.


21/33

VS/TUN architecture


22/33

VS-TUN workflow


23/33

VS-TUN advantages and

disadvantages Advantages:

Real servers send response packets to client directly,

which can follow different network routes Real servers can be in different networks, LAN/WAN

Greatly increasing the scalability of Virtual Server

Disadvantages:

Real server must support IP tunneling protocol


24/33

Virtual Server via Direct Routing

(VS/DR) The load balancer and the real servers must have one of their

interfaces physically linked by an uninterrupted segment of LAN suchas an Ethernet switch.

The virtual IP address is shared by real servers and the load balancer.

Each real server has a non-ARPing, loopback alias interfaceconfigured with the virtual IP address, and the load balancer has aninterface configured with the virtual IP address to accept incoming

packets. The workflow of VS/DR is similar to that of VS/NAT or VS/TUN. In

VS/DR, the load balancer directly routes a packet to the selected

server (the load balancer simply changes the MAC address of the dataframe to that of the server and retransmits it on the LAN). When the server receives the forwarded packet, the server determines

that the packet is for the address on its loopback alias interface,processes the request, and finally returns the result directly to the user.


25/33

VS/DR architecture


26/33

VS-DR workflow


27/33

VS-DR advantages and disadvantages

Advantages:Real servers send response packets to clients

directly, which can follow different networkroutes

No tunneling overhead

Disadvantages:

Servers must have non-arp alias interface

The load balancer and server must have one oftheir interfaces in the same LAN segment


28/33

Comparison

VS/NAT VS/TUN VS/DR

Server any Tunneling Non-arp device

server network private LAN/WAN LANserver number low (10~20) High (100) High (100)

server gateway load balancer own router Own router

Note: those numbers are estimated based on the

assumption that load balancer and backend servers

have the same hardware configuration.


29/33

Scheduling algorithms

Round-Robin

Weighted Round-Robin

Least-Connection

Weighted Least-Connection


30/33

The LocalNode feature

In a virtual server of only a few nodes(2,3

or more), it is a resource waste if the load

balancer is only used to direct packets. The LocalNode feature enable that the load

balancer not only can redirect packets, but

also can process some packets locally


31/33

LVS cluster management software

RedHat Cluster Server / Piranha

LVS+Piranha Cluster Management tools.

UltraMoney: Open-Source Server Farm

LVS+lvs-gui+heartbeat+ldirectord

heartbeat+ldirectord

heartbeat+mon

...


32/33

Some sites using LVS

UK National JANET Cache

(wwwcache.ja.net)

www.linux.com

sourceforge.net

One of largest PC manufacturers

www.netwalk.com


33/33

References

Wikipedia

http://www.linux-vs.org

13- clustering and load balancing

Documents