A Novel Approach to Deploying High Performance Computing Applications on Cloud Platform

Now people can run HPC applications on cloud platform and they do not need to own or maintain physical infrastructure. However, they have the burden of managing large pools of virtual machines (VMs), because cloud users today must manually instantiate, configure and maintain the virtual hosts. In this paper, a novel approach is proposed to deploy HPC applications on cloud platform easily which integrates a message passing interface (MPI) based simple programming model and single system image (SSI) technology. By hiding the details of MPI applications and VMs cluster, the approach allows cloud users to execute MPI applications on VMs cluster as normal multi-process applications on a single machine. So cloud users are promised to create and maintain a single VM only to deploy and execute HPC applications on hundreds of VMs. The experimental results show that the approach can reduce the burden of deploying HPC applications on cloud platform greatly.

[1]  George Bosilca,et al.  Open MPI: Goals, Concept, and Design of a Next Generation MPI Implementation , 2004, PVM/MPI.

[2]  Sanjay Ghemawat,et al.  MapReduce: Simplified Data Processing on Large Clusters , 2004, OSDI.

[3]  Chandra Krintz,et al.  Evaluating the Performance Impact of Xen on MPI and Process Execution For HPC Systems , 2006, First International Workshop on Virtualization Technology in Distributed Computing (VTDC 2006).

[4]  Marianne Shaw,et al.  Scale and performance in the Denali isolation kernel , 2002, OSDI '02.

[5]  Andrew Warfield,et al.  Live migration of virtual machines , 2005, NSDI.

[6]  Michael Vrable,et al.  Scalability, fidelity, and containment in the potemkin virtual honeyfarm , 2005, SOSP '05.

[7]  Randy H. Katz,et al.  Above the Clouds: A Berkeley View of Cloud Computing , 2009 .

[8]  Eyal de Lara,et al.  SnowFlock: rapid virtual machine cloning for cloud computing , 2009, EuroSys '09.

[9]  Alexandru Iosup,et al.  Performance Analysis of Cloud Computing Services for Many-Tasks Scientific Computing , 2011, IEEE Transactions on Parallel and Distributed Systems.

[10]  David E. Irwin,et al.  Dynamic virtual clusters in a grid site manager , 2003, High Performance Distributed Computing, 2003. Proceedings. 12th IEEE International Symposium on.