Enabling Work-conserving Bandwidth Guarantees for Multi-tenant Datacenters via Dynamic Tenant-Queue Binding

TL;DR

QShare is a practical in-network solution that achieves both bandwidth guarantees and work conservation in multi-tenant datacenter networks by using weighted fair queuing and dynamic tenant-queue binding, ensuring high utilization and predictable performance.

Contribution

This paper introduces QShare, a novel in-network approach that simultaneously guarantees bandwidth and maintains work conservation in multi-tenant datacenter networks, addressing a key challenge in cloud networking.

Findings

QShare ensures bandwidth guarantees for tenants.

QShare achieves over 91% network utilization.

QShare is implementable with existing switch hardware.

Abstract

Today's cloud networks are shared among many tenants. Bandwidth guarantees and work conservation are two key properties to ensure predictable performance for tenant applications and high network utilization for providers. Despite significant efforts, very little prior work can really achieve both properties simultaneously even some of them claimed so. In this paper, we present QShare, an in-network based solution to achieve bandwidth guarantees and work conservation simultaneously. QShare leverages weighted fair queuing on commodity switches to slice network bandwidth for tenants, and solves the challenge of queue scarcity through balanced tenant placement and dynamic tenant-queue binding. QShare is readily implementable with existing switching chips. We have implemented a QShare prototype and evaluated it via both testbed experiments and simulations. Our results show that QShare ensures bandwidth guarantees while driving network utilization to over 91% even under unpredictable traffic demands.