Blue Gene ranks on the TOP500 list1 along with three other entries in the top 10.
The IBM® System Blue Gene® Solution is the result of an IBM supercomputing project dedicated to building a new family of supercomputers optimized for bandwidth, scalability and the ability to handle large amounts of data while consuming a fraction of the power and floor space required by today’s high performance systems.
Blue Gene is also extremely efficient. Because of unique design points that allow dense packaging of processors, memory and interconnect, power and floor space consumption.
NARSS Blue Gene/L contains one rack:
· Compute nodes (Dual processor; 1024 per rack)
· Processor (PowerPC 440 700MHz; two per node)
· Memory (512 MB SDRAM-DDR per node)
· Operating Systems (Compute Node – Lightweight proprietary kernel, I/O Node – Embedded Linux, Front End and Service Nodes – SuSE SLES 9 Linux)
· Peak performance per rack – 5.73 TFlops
To make an application fit on Blue Gene/L it should be
· Coded in C, C++, or FORTRAN with MPI for communication.
· Single threaded applications
· 32-bit Application addressing
· MPI version is MPICH v1.2
· SIMD (Single Instruction Multiple Data)
· MPI task requiring memory less than 500 MB
· Floating point-intensive code
Current applications running on NARSS Blue Gene/L
· Climate Modeling (WRF Model)
· Affects of Mobile on Human Brain