High-Performance Processing of Large-Scale Parallel Applications in Heterogeneous Cloud Computing Data Centers

Abstract

Efficient application processing is critical for achieving high performance in heterogeneous computing systems, i.e., Optimal System configuration and load distribution of some given types of applications, such that the average response time of tasks is minimized. Such performance optimization is important for a cloud computing provider to efficiently utilize all the available resources and to deliver the highest quality of service. To achieve this, first we use clustering algorithm to group the tasks into distinct classes with similar characteristics in terms of resource and performance requirements. Second, a multicore server processor is treated as a group of queuing systems with multiple servers, i.e., M/M/m queuing systems. Third, we formulate and solve the optimal load distribution of tasks and the problem of optimal multicore server processing partitioning for multiple heterogeneous multicore servers. We show that although the problem is sophisticated, it can be solved by a numerical algorithm. We obtain not only detailed assessment of cloud center performance, but also insights into equilibrium arrangement, and power consumption to be kept under control.