Introduction to Cluster Computing Seminar Topic:
Cluster computing is the integration of standalone computers and resources connected with the help of hardware, networks, and software to behave like a single computer. The technologies available today, have redefined cluster computing. It incorporates load balancing clusters and high availability clusters. Clusters are made up of standalone computers connected to a single or a group of interconnected switches operating at layer 2 and within a virtual LAN. Each computer may have different properties like number of processors, access to various storage devices. The network used for this, contains high speed, low latency switches.
The factors like price, performance, and compatibility with other cluster hardware and system software as well as communication characteristics of the applications that will use the cluster decide the type of network technology to be used. Decisions like type of hardware the nodes will run on, which interconnect to be used, which switches architecture to build on, are to taken before assembling of a cluster. Each decision affects the others.
The benefits of the cluster computing are scalability, availability, and performance. In scalability, the compute nodes process together to run cluster-enabled applications. In availability, if one compute node fails, then other nodes provide backup to the running applications. For performance, all the nodes all the nodes run same application, and hence provide better performance.
There are different clusters to accomplish different tasks. High Availability or Failover Cluster are those which provide high availability to the applications or services. In this type of cluster, the compute nodes are connected in such a way that if a node fails then the applications running over it are transferred onto the other node. So there is uninterrupted availability of services to the user, though it reduces some performance, but in the mean time the failover node can be repaired.
This type of cluster is best for mails, file, database, web, or application servers. Load Balancing Cluster is another cluster type which is used to provide high availability. In this cluster, the nodes contain same contents and are able to handle same request. The incoming request is distributed among the different nodes. If some node fails, then the request is distributed again to the remaining nodes. This cluster along with fail over cluster gives high reliability and availability and is used in mail, web, news, or FTP services.
Download Seminar Report on Cluster Computing .