Load Balancing in Cloud Computing
Workload distribution maintains both IT resource overutilization and underutilization up to a certain extent. At the time of dynamic workload distribution, it is important to know the type and geographical location of the IT resource. Monitoring of runtime workload tracking and data processing is also necessary for proper load balancing.
Load balancing is the technique which is responsible to distribute the load, that is number of requests, number of users, etc. across one or additional servers, network interfaces, hard drives or other computing resources. Many reasons to use load balancing include improved performance, reliability, elasticity, scalability and availability. In cloud computing, load balancing is different from the conventional architecture of load balancing.
For proper load balancing across various IT resources and to increase performance, there is a load balancer mechanism which handles runtime distribution. The distribution function includes:
1. A high-power resource for handling the workload that requires more processing power.
2. Providing services based on priority.
3. Directing traffic to a particular IT resource as per the requirement of cloud users.
The load balancer can perform functions such as optimizing resource utilization, better throughput, maintaining the quality of services, etc. It can be hardware or software. It can be a dedicated hardware device, dedicated software for handling traffic or software that can be handled by cloud management software. It is not visible to cloud users, as it acts between users’ requests and IT resource allocation. Distribution of workload is based on user demand.