Managing Bufferbloat in Cloud Storage Systems

TL;DR

This paper investigates how bufferbloat affects schedulability and performance in cloud storage systems, proposing an adaptive admission control mechanism called SlowFast CoDel to mitigate bufferbloat effects.

Contribution

It identifies the impact of bufferbloat on scheduling in cloud storage and introduces an adaptive admission control solution, SlowFast CoDel, to address this issue.

Findings

Bufferbloat impacts system schedulability and performance isolation.

Traditional static admission controls are ineffective for dynamic workloads.

SlowFast CoDel effectively mitigates bufferbloat in cloud storage environments.

Abstract

Today, companies and data centers are moving towards cloud and serverless storage systems instead of traditional file systems. As a result of such a transition, allocating sufficient resources to users and parties to satisfy their service level demands has become crucial in cloud storage. In cloud storage, the schedulability of system components and requests is of great importance to achieving QoS goals. However, the bufferbloat phenomenon in storage backends impacts the schedulability of the system. In a storage server, bufferbloat happens when the server submits all requests immediately to the storage backend due to a large buffer in the backend. In recent decades, many studies have focused on the bufferbloat as a latency problem. Nevertheless, none of these works investigate the impact of bufferbloat on the schedulability of the system. In this paper, we demonstrate that the bufferbloat impacts scheduling and performance isolation and identify utilizing admission control in the storage backend as an easy-to-adopt solution to mitigate bufferbloat. Moreover, we show that traditional static admission controls are inadequate in the face of dynamic workloads in cloud environments. Finally, we propose SlowFast CoDel, an adaptive admission control, as a starting point for developing adaptive admission control mechanisms to mitigate bufferbloat in cloud storage.