DYNAMIC MAPPING AND DEPLOYMENT OF INDEPENDENT TASKS ON DISTRIBUTED SYSTEMS USING RESIDUAL EXECUTION TIME

摘要

Mapping and scheduling are complex computational problems. They are known to be NP-Complete except under a few special situations. There is a wide variety of approaches to the problem of mapping and scheduling in HC systems that are either static or dynamic. In this paper, we propose a preemptive (migratory) dynamic mapping technique that is based on the famous Min-min heuristic. In our approach, we added a deployment step to allow dynamic reallocation of running tasks to improve the overall performance of the HC system. The reallocation (migration) decisions are taken based on the expected residual execution times for the suspended tasks. The goal is to minimize the total execution time (makespan) of the metatask.