Flex: Closing the Gaps between Usage and Allocation

TL;DR

This paper introduces Flex, an online resource manager for data centers that significantly improves resource utilization and request admission rates while maintaining QoS, addressing underutilization issues caused by demand estimation gaps.

Contribution

Flex is a novel online resource management system that closes the gap between usage and allocation, leading to higher utilization and request acceptance in data centers.

Findings

Flex admits up to 1.74x more requests

Flex achieves 1.6x higher utilization

Flex maintains QoS in large-scale evaluations

Abstract

Data centers are giant factories of Internet data and services. Worldwide data centers consume energy and emit emissions more than airline industry. Unfortunately, most of data centers are significantly underutilized. One of the major reasons is the big gaps between the real usage and the provisioned resources because users tend to over-estimate their demand and data center operators often rely on users' requests for resource allocation. In this paper, we first conduct an in-depth analysis of a Google cluster trace to unveil the root causes for low utilization and highlight the great potential to improve it. We then developed an online resource manager Flex to maximize the cluster utilization while satisfying the Quality of Service (QoS). Large-scale evaluations based on real-world traces show that Flex admits up to 1.74x more requests and 1.6x higher utilization compared to tradition schedulers while maintaining the QoS.