Time based agent garbage collection algorithm for multicore architectures

Abstract

Multicore architecture consists of large number of processors that are arranged together on the same chip uses hyper threading technology (using virtualization). The parallel processing exhibited by the large number of processor increases the challenges and design issues for the real time process (critical task) execution on the processor cores. This is a vital part from the operating system scheduling and storage compaction point of view. In this paper we combine the AMAS theory of multiagent system with the affinity based processor scheduling and this will be best suited for critical task execution on multicore platforms. This hard-soft processor affinity scheduling algorithm promises in minimizing the average waiting time of the non critical tasks in the centralized queue and avoids the context switching of critical tasks. That is we assign hard affinity for critical tasks and soft affinity for non critical tasks. The algorithm is applicable for the system that consists of both critical and non critical tasks in the ready queue. Since we use the actual round robin scheduling for non critical tasks and due to soft affinity the load balancing is done automatically for non critical tasks. The novel storage compaction (garbage collection) algorithm consists of local and global garbage collectors which efficiently identifies and removes the unwanted files or information in the distributed or multicore environment making a room for efficient scheduling. We actually modified and simulated the linux 2.6.11 kernel process scheduler to incorporate this scheduling concept. Our result shows the maximum cpu utilization for the non critical tasks and high throughput for the critical tasks.