A heuristic-based real-time disk scheduling algorithm for mixed-media workload

Abstract

For mixed-media workload, disk scheduling strategies should meet the deadlines of requests with timing constraints while optimizing the disk utilization. To design a scheduling algorithm in mixed-media workload environment is very intricate because finding an optimal schedule is NP-hard as well as different QoS (Quality of Service) of each request should be guaranteed. In this paper, we present a novel real-time disk scheduling algorithm which meets the QoS of each requests. Our algorithm employs several heuristics using the requirements of each request and thus could reduce the huge solution space to a feasible extent. Through trace-driven simulation, we show that our algorithm outperforms other algorithms in terms of response time, throughput, and QoS guarantee for real-time requests.