Throughput-Optimal Multiresource-Job Scheduling with Continuous Requirement Distribution

TL;DR

This paper introduces the first throughput-optimal scheduling policies for multiresource jobs with continuous requirement distributions, improving efficiency and validating performance on real-world datacenter data.

Contribution

It develops a family of throughput-optimal scheduling policies for continuous resource requirement models, including efficient variants and validation on Google Borg data.

Findings

Policies achieve throughput optimality for continuous MRJ models.

Discretization approach improves computational efficiency.

Validated state-of-the-art performance on Google Borg trace data.

Abstract

Modern computing systems process jobs with resource requirements such as CPU and memory, which are described by multiresource jobs (MRJ) queueing models. In practice, job resource requirements are spread out over so many values, that it is rare to see the same value twice. This pattern is best modeled by a continuous distribution of requirement values. However, the existing theoretical work on stability or throughput-optimality focuses on queueing models with class-based resource requirements. In class-based models, the number of distinct resource requirements must be small to demonstrate strong empirical performance, making them a poor match for these practical systems. We introduce the first throughput-optimal family of scheduling policies for the continuous MRJ model, with both preemptive and nonpreemptive variants. We further introduce several efficient policy families, which remain throughput-optimal while considerably improving computational efficiency, under some distributional assumptions. We use a discretization approach, where we choose the discretization granularity based on the system load and the distribution of resource requirements. We validate the real-world applicability of our policies by comparing them against existing index-based policies on parametrized distributions and on datacenter trace data from the Google Borg scheduler, demonstrating state-of-the-art performance.