XenLoop

Publisher Website
Google Scholar
Abstract
Advances in virtualization technology have focused mainly on strengthening the isolation barrier between virtual ma- chines (VMs) that are co-resident within a single physical machine. At the same time, a large category of communica- tion intensive distributed applications and software compo- nents exist, such as web services, high performance grid ap- plications, transaction processing, and graphics rendering, that often wish to communicate across this isolation bar- rier with other endpoints on co-resident VMs. State of the art inter-VM communication mechanisms do not adequately address the requirements of such applications. TCP/UDP based network communication tends to perform poorly when used between co-resident VMs, but has the advantage of be- ing transparent to user applications. Other solutions exploit inter-domain shared memory mechanisms to improve com- munication latency and bandwidth, but require applications or user libraries to be rewritten against customized APIs - something not practical for a large majority of distributed applications. In this paper, we present the design and im- plementation of a fully transparent and high performance inter-VM network loopback channel, called XenLoop, in the Xen virtual machine environment. XenLoop does not sac- rifice user-level transparency and yet achieves high commu- nication performance between co-resident guest VMs. Xen- Loop intercepts outgoing network packets beneath the net- work layer and shepherds the packets destined to co-resident VMs through a high-speed inter-VM shared memory channel that bypasses the virtualized network interface. Guest VMs using XenLoop can migrate transparently across machines without disrupting ongoing network communications, and seamlessly switch between the standard network path and the XenLoop channel. XenLoop can operate in both bridged mode and routed mode configurations of Xen. Evaluations using a number of unmodified benchmarks show that Xen- Loop reduces the inter-VM round trip latency by up to a factor of 5 and increases bandwidth by up to a factor of 6.

This publication has 6 references indexed in Scilit: