SLO-aware Colocation of Data Center Tasks Based on Instantaneous Processor Requirements

TL;DR

This paper introduces a probabilistic method called Gaussian Percentile Approximation (GPA) for efficient task colocation in data centers, accounting for heterogeneous CPU demands and ensuring SLO compliance.

Contribution

It proposes a novel GPA method that uses Gaussian approximations of high percentiles for better task placement decisions in heterogeneous environments.

Findings

GPA accurately predicts task fit with desired SLO compliance.

Empirical CPU usage distributions do not follow a single distribution.

GPA maintains capacity violations within acceptable SLO levels.

Abstract

In a cloud data center, a single physical machine simultaneously executes dozens of highly heterogeneous tasks. Such colocation results in more efficient utilization of machines, but, when tasks' requirements exceed available resources, some of the tasks might be throttled down or preempted. We analyze version 2.1 of the Google cluster trace that shows short-term (1 second) task CPU usage. Contrary to the assumptions taken by many theoretical studies, we demonstrate that the empirical distributions do not follow any single distribution. However, high percentiles of the total processor usage (summed over at least 10 tasks) can be reasonably estimated by the Gaussian distribution. We use this result for a probabilistic fit test, called the Gaussian Percentile Approximation (GPA), for standard bin-packing algorithms. To check whether a new task will fit into a machine, GPA checks whether the resulting distribution's percentile corresponding to the requested service level objective, SLO is still below the machine's capacity. In our simulation experiments, GPA resulted in colocations exceeding the machines' capacity with a frequency similar to the requested SLO.