International Journal of Computer Applications (0975 – 8887) Volume 95– No.20, June 2014 21 The Expedient Approach for High Availability in Web Server Services for HPC Attained by Clustering using Virtualization Mamta Gadoria M.Tech CS (Student) J.V.W.U Jaipur Sarvesh Singh Assistant Professor J.V.W.U Jaipur Pratibha Gangwar M.Tech CS (Student) J.V.W.U Jaipur ABSTRACT In this paper we have presented the model of clustering using Heartbeat and implemented with Virtualized environment. . For making a machine to be virtualized, the hardware must endow with virtualization capabilities we believe that we understand what makes the websites unreliable; the key to build a more reliable website is to make a cluster. Clustering is typically transparent from the clients as well as developers, perk up the system availability to the online users , as the failed server will automatically pack up and the users will automatically switched at that particular instant to the another server. Heartbeat is an open source portable tool for clustering. General Terms Heartbeat Apache, clustering, High availability Keywords HA, 2-node cluster 1. INTRODUCTION Apache Heartbeat, a High availability implementation for web services, is a broadly acceptable technology. While configuring Apache Web Server, schemes have been already addressed in previous literature. Nowadays providing reliable online services is a trend for almost every organizations and enterprises. Reliability here means, reliability of the server, hosting websites. Unreliable websites are the only major source of user frustration. In 2-nodes clustering, you have one front end node and a back end node. A 2-node cluster schema is used here to make a transparent layer i.e. requirement of both application layer and transport layer mechanisms, to the clients. Although the website reliability is a key concern from the very first day it has been accessed because resubmitting the requests to the server throws bad impression to the clients. So, this reason made competition a bit higher in the field of servers. Clustering offers the finest scalability i.e. can be increased dynamically to meet the demands and high availability i.e. can persist when a server instance fails. In cluster structures, the server applications can reliably make use of the resources in the least down time. In this paper, we have spotlight on the HA clusters with heartbeat tool for hiding the server failures from the clients using clustering, an accepted technology for high availability, parallel processing applications and shared computing resources. The principle of this record is to give explanation of how to put into practice, a 2-node HA cluster. Devoid of clustering if a server running application becomes unavailable at a particular instant, then the services will not be accessible until the application server is fixed. Server cluster labels how the nodes are tied to store the cluster configurations. The impression of this approach is to supply the tightly coupled services to the clients. The model proposed here is an Active-Active Model of nodes, where both the nodes make use of homogenous software configurations. In other words we can also say that, in the event of failure on the node 1, node 2 will automatically serve the services, i.e. restarting departed machine on the secondary node. Heartbeat program sprints on every Linux platform. High Availability scheme implemented here has a design of three Linux systems (Centos), i.e. client, primary node and secondary node. The Heartbeat tool is high performance computing tool which serves with easy observing between two systems. The resources are directed between the two nodes in the cluster, if one fails, another will serve by managing resources. By configuring the heartbeat tool for clustering user will without doubt experience a bare minimum of tumult in services that are offered by the servers. To create a cluster with 2-nodes, you need to configure the heartbeat properly with accurate hardware with network infrastructure. 2. LITERATURE SURVEY Reliable online services are the most recent movement to avoid the user frustration. In this paper, we centered on the heartbeat tool for making a 2-node cluster, hiding the server failure from the clients. High availability demotes to consistency, sustainable and integrity of data access. High availability is typically capable to assemble the development in the reliability of servers and applications. Some models necessitate massive data processing in huge scale computing environments i.e. clouds. Managing a server with loosely coupled architecture is an intricate task. By creating a cluster, you can make tightly coupled environment. High availability clustering is defined by asymmetric active/active model. Both hardware and software fails, unavoidably. There can be a number of nodes which can break down the whole thing; it only necessitates the hardware to reinstate or Operating system to restart. 3. HIGH AVAILABILITY CLUSTERING 3.1 Node Cluster The 2- node clustering through Heartbeat proffers much uncomplicated takeover semantics with dependencies and policies. Almost every protocol centered on increasing the reliability of the messages i.e. fully attentive towards the ensuring of high availability in the event. The 2-nodes in the cluster are also known as neighbors of the private network. This approach is to provide high availability to the web