QP: A Heterogeneous Multi-Accelerator Cluster

We present a heterogeneous multi-accelerator cluster developed and deployed at NCSA. The cluster consists of 16 AMD dual-core CPU compute nodes each with four NVIDIA GPUs and one Xilinx FPGA. Cluster nodes are interconnected with both InfiniBand and Ethernet networks. The software stack consists of standard cluster tools with the addition of accelerator-specific software packages and enhancements to the resource allocation and batch sub-systems. We highlight several HPC applications that have been developed and deployed on the cluster. We also present our Phoenix application development framework that is meant to help with developing new applications and migrating existing legacy codes to heterogeneous systems.

[1]  Volodymyr V. Kindratenko,et al.  Phoenix: A Runtime Environment for High Performance Computing on Chip Multiprocessors , 2009, 2009 17th Euromicro International Conference on Parallel, Distributed and Network-based Processing.

[2]  Manish Vachharajani,et al.  GPU acceleration of numerical weather prediction , 2008, 2008 IEEE International Symposium on Parallel and Distributed Processing.

[3]  Volodymyr Kindratenko,et al.  Investigating Application Analysis and Design Methodologies for Computational Accelerators , 2009 .

[4]  Klaus Schulten,et al.  Adapting a message-driven parallel application to GPU-accelerated clusters , 2008, 2008 SC - International Conference for High Performance Computing, Networking, Storage and Analysis.