In 2008, the National Nuclear Security Administration's (NNSA) Lawrence Livermore National Laboratory teamed with 10 computing industry leaders to accelerate the development of powerful next-generation Linux clusters in a project dubbed Hyperion.

Hyperion brought together Dell, Intel, Supermicro, QLogic, Cisco, Mellanox, DDN, Sun, LSI and RedHat to create a large-scale testbed for high-performance computing technologies critical to NNSA's work to maintain the aging U.S. nuclear weapons stockpile without underground nuclear testing, and industry's ability to make petaflop/s (quadrillion floating operations per second) computing and storage more accessible for commerce, industry and research and development.

Hyperion Scalable Units

Phase 1-11.5 TF/s Scalable Unit

Deployment is a 4 SU, 46 TF/s Cluster with full IBA

• 576 Total Nodes and 4,608 cores, 12.1 TB/s memory bandwidth, 4.6 TB capacity
• IBA is expandable to 1,728 IB ports; single plane and can double to dual pane
• 80.0 GF/s dual-socket 2.5 GHz quad-core Intel LV Harpertown nodes
  • 8 GB from 4-channel FB-DIMM 667 RAM @ 21.6 GB/s
• Nodes utilize PCI-Express gen. 2 I/O which provides an upgrade path to IBA 4x QDR
• Storable Scalable Units (SSU) from DDN, LSI and Sun yielding >47 GB/s and 732 TB disk

Phase 2-13 TF/s Scalable Unit

Deployment is a 4 SU, 52 TF/s Cluster with full IBA

• 576 Total Nodes and 4,608 cores, 36.9 TB/s memory bandwitdh, 6.9 TB capacity
• IBA is expandable to 1,728 IB ports
• ~77 GF/s dual-socket 2.4 GHz quad-core Intel Nehalem nodes with 12GB of memory per node
  • ~8 GB from 6-channel DDR 1333 DRAM @ 64 GB/s
• Nodes utilize PCI-Express v2 I/O which provides an upgrade path to IBA 4x QDR
• Storable Scalable Units (SSU) from DDN, LSI and Sun yielding >47 GB/s and 732 TB disk
• Full system will require 400kW of power, 112 tons of cooling. Sited at LLNL.

Hyperion Storage