IBM Blue Gene/Q supercomputer can scale to 100 petaflops

EETimes – IBM’s officially unveiled its next generation Blue Gene/Q (BGQ) supercomputer, the third generation in its Blue Gene family, with 16 multi-processing core technology and a scalable peak performance of up to 100 petaflops. It can produce 2 gigaflops per watt. A 100 petaflop system would need 50 megawatts of power.

Based on IBM’s PowerPC A2 processing architecture, each of the BGQ’s 64-bit processors sports 16 compute cores, four times the number of cores used in the previous Blue Gene/P system, with each CPU able to handle four threads simultaneously.

IBM said the system also had an additional core to run the operating system administrative functions and a redundant spare core. Each processor sports 32 KB of L1 cache, divided equally between data and instruction, while the L2 consists of 8 MB of embedded DRAM.

A full BGQ rack would contain 1024 nodes, or 16K cores.

The memory and I/O controllers are integrated onto the chip itself and the I/O has been separated from the server nodes so that configurations can scale compute and I/O independently.

A BGQ rack can apparently accommodate between eight and 128 I/O nodes, which use the same Power A2 chip as the compute servers.

A new feature in this iteration of Blue Gene is the addition of a 5D Torus which uses fiber optics for server-to-server communication at up to 40 gigabits per second, four times faster than Blue Gene/P’s interconnect

Argonne National Laboratory (ANL) — one of the DOE’s oldest and largest labs for science and engineering research, located outside of Chicago– will also be using the Q to help solve global challenges including the design of electric car batteries, climate change and exploring the evolution of the universe.

Argonne’s 10-petaflop, 48 rack, two ton system, “Mira”, gives scientists and researchers the opportunity to work closely with both senior IBM and Argonne technical staff to adapt their codes in order to take full advantage of the system’s capabilities.

At LLNL –a multidisciplinary national security laboratory for DOE’s National Nuclear Security Administration (NNSA)– BGQ will be deployed in a system called “Sequoia”, expected to achieve 20 petaflops of peak performance, becoming one of the world’s fastest supercomputers. IBM will deploy 96 racks to LLNL beginning as early as December of this year.

LLNL said it will be using the system as part of its effort to maintain the U.S.’s aging nuclear capabilities without the need for physical testing, as well as tackling other problems such as grid and network management, energy research and climate change studies.

If you liked this article, please give it a quick review on ycombinator or StumbleUpon. Thanks