Don’t Hurry Be Happy: A Deadline-Based Backfilling Approach

Abstract

Abstract. Computing resources in data centers are usually managed by a Resource and Job Management System whose main objective is to complete submitted jobs as soon as possible while maximizing resource usage and ensuring fairness among users. However, some users might not be as hurried as the job scheduler but only interested in their jobs to complete before a given deadline. In this paper, we derive from this initial hypothesis a low-complexity scheduling algorithm, called Deadline-Based Backlling (DBF), that distinguishes regular jobs that have to complete as early as possible from deadline-driven jobs that come with a deadline before when they have to nish. We also investigate a scenario in which deadline-driven jobs are submitted and evaluate the impact of the proposed algorithm on classical performance metrics with regard to state-of-the-art scheduling algorithms. Experiments conducted on four dierent workloads show that the proposed algorithm signicantly reduces the average wait time and average stretch when compared to Conservative Backlling.

On the use of HCM to develop a resource allocation algorithm for heterogeneous clusters

“…Other works have objectives that differ somewhat from the goal of RAA. Some of them focus on a scheduler designed for parallel supercomputers, using a backfill‐based scheduling scheme . The backfill scheduling scans the queue and selects smaller jobs that may utilize the available resources as long as these jobs do not delay the start of another job.…”

“…Some of them focus on a scheduler designed for parallel supercomputers, using a backfill-based scheduling scheme. [18][19][20] The backfill scheduling scans the queue and selects smaller jobs that may utilize the available resources as long as these jobs do not delay the start of another job. Backfilling requires the execution time of each job to be known a priori.…”