Revisiting Size-Based Scheduling with Estimated Job Sizes

TL;DR

This paper investigates the robustness of size-based scheduling algorithms under inaccurate job size estimates, revealing workload-dependent performance and proposing a mitigation technique that enhances practicality across diverse scenarios.

Contribution

It provides a comprehensive analysis of how estimation errors affect size-based schedulers and introduces a simple, effective method to mitigate these issues without complex modifications.

Findings

Size-based schedulers outperform size-oblivious ones when workload is not heavily skewed.

Under-estimation of job sizes causes more severe scheduling issues than over-estimation.

The proposed mitigation approach significantly improves scheduling performance in diverse workloads.

Abstract

We study size-based schedulers, and focus on the impact of inaccurate job size information on response time and fairness. Our intent is to revisit previous results, which allude to performance degradation for even small errors on job size estimates, thus limiting the applicability of size-based schedulers. We show that scheduling performance is tightly connected to workload characteristics: in the absence of large skew in the job size distribution, even extremely imprecise estimates suffice to outperform size-oblivious disciplines. Instead, when job sizes are heavily skewed, known size-based disciplines suffer. In this context, we show -- for the first time -- the dichotomy of over-estimation versus under-estimation. The former is, in general, less problematic than the latter, as its effects are localized to individual jobs. Instead, under-estimation leads to severe problems that may affect a large number of jobs. We present an approach to mitigate these problems: our technique requires no complex modifications to original scheduling policies and performs very well. To support our claim, we proceed with a simulation-based evaluation that covers an unprecedented large parameter space, which takes into account a variety of synthetic and real workloads. As a consequence, we show that size-based scheduling is practical and outperforms alternatives in a wide array of use-cases, even in presence of inaccurate size information.