Rendering differential performance preference through intelligent network edge in cloud data centers

摘要

Sharing the network infrastructure, the performance of emerging distributed applications andservices in data centers is directly impacted by the network. As these applications are becomingmore and more demanding, it is challenging to satisfy their requirements of low latency, highthroughput, and low packet loss rate simultaneously. Prior approaches typically resort to flowcontrol or scheduling mechanisms, prioritizing flows according to their demands. However, noneof the methods can solely satisfy the various demands of data center applications. Addressingthis challenge, we propose tasch, a preference aware flow scheduling mechanism equippedin the software network edge (ie, end-host networking). This mechanism utilizes multipleseparate queues for flows with different preferences, which guarantees low packet delay forlatency-sensitive flows and provides bandwidth guarantees for throughput-sensitive flows. Acoordinating algorithm is presented to share the network resource among multiple queueswith pareto-optimality. tasch is implemented as a thin and plugable kernel module in Linuxbased hypervisors, which lies between the complicated physical network and tenants VMs.Subsequently, based on the flow traces of real-world applications, extensive experiments wereconducted to verify the effectiveness of network management mechanism.