Site Index | Contact


BG/L Home

Using BG/L

Photo Gallery





BG/L in the News



Three-dimensional (3D) torus network

BlueGene/L uses a three-dimensional (3D) torus network in which the nodes (red balls) are connected to their six nearest-neighbor nodes in a 3D mesh. In the torus configuration, the ends of the mesh loop back, thereby eliminating the problem of programming for a mesh with edges. Without these loops, the end nodes would not have six near neighbors.

Current configuration of BlueGene/L


Original Configuration

BlueGene/L boasts a peak speed of over 596 teraFLOPS and a total memory of 69 tebibytes. The full system has 106,496 dual-processor compute nodes. Multiple communications networks enable extreme application scaling:

  • Nodes are configured as a 32 x 32 x 64 3D torus; each node is connected in six different directions for nearest-neighbor communications
  • A global reduction tree supports fast global operations such as global max/sum in a few microseconds over 106,496 nodes
  • Multiple global barrier and interrupt networks allow fast synchronization of tasks across the entire machine within a few microseconds
  • 1,024 gigabit-per-second links to a global parallel file system to support fast input/output to disk

More information is available on BlueGene/L architecture:

BGL Configuration
BlueGene/L technology builds a supercomputer one dual-processor chip at a time. Chips are
aggregated into compute cards, which are then assembled into node cards. Each rack holds
32 node cards, and the full machine now comprises 104 racks.


  UCRL-WEB-148749 | Privacy & Legal Notice   November 20, 2008  
Lawrence Livermore National Laboratory
7000 East Avenue • Livermore, CA 94550
Operated by Lawrence Livermore National Security,
, for the Department of Energy's
National Nuclear Security Administration
NNSA LLNL SNL LANL LLNL LLNL Department of Energy logo NNSA