Scheduling in the Presence of Data Intensive Compute Jobs

TL;DR

This paper analyzes how non-adaptive scheduling policies affect job wait times in systems with both regular and sporadically data-intensive jobs, proposing indicators to predict policy performance.

Contribution

It introduces simplified performance indicators to predict scheduling policy effectiveness in systems with mixed job types, supported by simulation validation.

Findings

Performance indicators correlate with actual system performance.

Scheduling policies impact average job wait times.

Indicators effectively predict policy performance in mixed workloads.

Abstract

We study the performance of non-adaptive scheduling policies in computing systems with multiple servers. Compute jobs are mostly regular, with modest service requirements. However, there are sporadic data intensive jobs, whose expected service time is much higher than that of the regular jobs. Forthis model, we are interested in the effect of scheduling policieson the average time a job spends in the system. To this end, we introduce two performance indicators in a simplified, only-arrival system. We believe that these performance indicators are good predictors of the relative performance of the policies in the queuing system, which is supported by simulations results.