TY - GEN
T1 - An evaluation of multiple communication interfaces for virtualized SMP clusters
AU - Atif, Muhammad
AU - Strazdins, Peter
PY - 2009
Y1 - 2009
N2 - Clusters with multiple CPU nodes are becoming increasingly popular due to their cost/performance ratio. Due to its many potential advantages, interest in using virtualization on these systems has also increased. Although several studies on the applicability of Xen for high performance computing have been made, most overlook the issue of multiple network interfaces. In this paper, we present an update to the state of art of Xen and give a comprehensive performance evaluation of the various network configurations that can be implemented using multiple gigabit ethernet (GigE) interfaces. We introduce new Xen network configurations, which enable the Xen guests to efficiently utilize the available network infrastructure compared to the default Xen network configurations. The evaluation of these configurations show 10-50% improvement in the NAS Parallel Benchmark suite compared to the default configurations. For these new configuration on multiple SMP nodes, the results also indicate that the need for fast intra-domain communication mechanisms is not compelling. We also detail the MPI implementations in the case of multiple GigE interfaces and their impact on a virtualized environment.
AB - Clusters with multiple CPU nodes are becoming increasingly popular due to their cost/performance ratio. Due to its many potential advantages, interest in using virtualization on these systems has also increased. Although several studies on the applicability of Xen for high performance computing have been made, most overlook the issue of multiple network interfaces. In this paper, we present an update to the state of art of Xen and give a comprehensive performance evaluation of the various network configurations that can be implemented using multiple gigabit ethernet (GigE) interfaces. We introduce new Xen network configurations, which enable the Xen guests to efficiently utilize the available network infrastructure compared to the default Xen network configurations. The evaluation of these configurations show 10-50% improvement in the NAS Parallel Benchmark suite compared to the default configurations. For these new configuration on multiple SMP nodes, the results also indicate that the need for fast intra-domain communication mechanisms is not compelling. We also detail the MPI implementations in the case of multiple GigE interfaces and their impact on a virtualized environment.
KW - Communication interfaces
KW - MPI
KW - SMP clusters
KW - Virtualization
KW - Xen
UR - http://www.scopus.com/inward/record.url?scp=70349161494&partnerID=8YFLogxK
U2 - 10.1145/1519138.1519140
DO - 10.1145/1519138.1519140
M3 - Conference contribution
SN - 9781605584652
T3 - Proceedings of the 3rd ACM Workshop on System-level Virtualization for High Performance Computing, HPCVirt'09
SP - 9
EP - 16
BT - Proceedings of the 3rd ACM Workshop on System-level Virtualization for High Performance Computing, HPCVirt'09
T2 - 3rd ACM Workshop on System-level Virtualization for High Performance Computing, HPCVirt'09
Y2 - 31 March 2009 through 31 March 2009
ER -