Load Balancing plays a significant role in improving system performance. As cloud is composed of huge servers, load balancing techniques are used to balance load over multiple servers. Workload aware partitioning helps to balance the load in cloud efficiently as it distributes the incoming load as per the type of work. This reduces the response time and thus helps to improve the performance of the system. In this paper, we proposed an algorithm for logical partitioning of resources in cloud and distributing the workload accordingly to avoid overloading of servers. The goal of this approach is to design and develop the concept of load balancing using cloud partitioning by classification of jobs. Thus proposed partitioning algorithm should help in placement of workloads in correct region and balance the overall load effectively.