BIG DATA

Cat BIG DATA

AMD hardware added to IACAT accelerator cluster

The heterogenous computing cluster (called AC) used by the Institute for Advanced Computing Applications and Technologies (IACAT) to explore the potential of novel architectures to accelerate scientific computing soon will be enhanced with new hardware, thanks to a donation from AMD.

AMD is providing eight nodes, each with:

  • Two six-core Istanbul CPUs
  • Three ATI 5870 GPUs, each with 1GB of memory
  • 32 GB DDR2 main memory
  • QDR Infiniband (4GB/sec)

“The addition of AMD/ATI GPUs to AC enhances our ability to develop and validate new applications and novel programming tools for heterogeneous computing,” said Wen-mei Hwu, a professor and Sanders-AMD Chair in the Illinois Department of Electrical and Computer Engineering and leader of IACAT’s GPU research. “Generous donations like this keep Illinois at the forefront of computing applications and technology development from laptops to petaflops.”

This hardware is in addition to the cluster’s current 32 compute nodes, which feature:

  • Two dual-core 2.4 GHz AMD Opterons
  • 8 GB of memory
  • One NVIDIA Tesla S1070 containing 4 GT200 GPUs, each with 4 GB of memory
  • One Nallatech H101-PCIX FPGA accelerator, 16 MB SRAM, 512 MB SDRAM
  • QDR Infiniband (2GB/sec)

The cluster is used as a platform to support the development of GPU applications, tools, and environments to accelerate applications to the petascale and beyond, for electrical and computer engineering and computer science courses, for training workshops, and as a resource for science and engineering researchers.

For more information on AC, see http://www.iacat.uiuc.edu/resources/cluster/.

TRENDING