Blue Gene Technology Full Seminar Report Download

[seminar class]

Presented by;
Sunitha M. Jenarius

---

[attachment=10852]
BLUE GENE
What is Blue Gene
 A massively parallel supercomputer using tens of thousands of embedded PowerPC processors supporting a large memory space
 With standard compilers and message passing environment
 Why the name “Blue Gene”?
 “Blue”: The corporate color of IBM
 “Gene”: The intended use of the Blue Gene clusters – Computational biology, specifically, protein folding
History
 Dec’99, IBM Research announced $100M US effort to build a Petaflop scale supercomputer.
 Two goals of The Blue Gene project :
– Massively parallel machine architecture and software
– Bio-Molecular Simulation – advance orders of magnitude
 November 2001, Partnership with Lawrence Livermore National Laboratory (LLNL)
and this resulted in …
 Results
 Linpack Top 500 Supercomputers
Blue Gene Projects
 Four Blue Gene projects :
– BlueGene/L
– BlueGene/C
– BlueGene/P
– BlueGene/Q
Blue Gene/L
 The first computer in the Blue Gene series
 IBM first announced the Blue Gene/L project, Sept. 29, 2004
 Final configuration was launched in October 2005
– Blue Gene/L - Unsurpassed Performance
 Designed to deliver the most performance per kilowatt of power consumed
 Theoretical peak performance of 360 TFLOPS
 Final Configuration (Oct. ‘05) scores over 280 TFLOPS sustained on the Linpack benchmark.
 Nov 14, ‘06, at Supercomputing 2006, Blue Gene/L was awarded the winning prize in all HPC Challenge Classes of awards.
Blue Gene/L Architecture
 Can be scaled up to 65,536 compute or I/O nodes, with 131,072 processors
 Each node is a single ASIC with associated DRAM memory chips
 Each ASIC has 2 700 MHz IBM PowerPC processors
 PowerPC processors
– Low-frequency, low-power embedded processors, superior to today's high-frequency, high-power microprocessors by a factor of 2 or more
– Double-pipeline-double-precision Floating Point Unit
– A cache sub-system with built-in DRAM controller
 Node CPUs are not cache coherent with one another
 FPUs and CPUs are designed for low power consumption
– Using transistors with low leakage current
– Local clock gating
– Putting the FPU or CPU/FPU pair to sleep
 1 rack holds 1024 nodes or 2048 processors
 Nodes optimized for low power consumption
 ASIC based on System-on-a-chip technology
– Large numbers of low-power system-on-a-chip technology allows it to outperform commodity clusters while saving on power
– Aggressive packaging of processors, memory and interconnect
– Power Efficient & Space Efficient
– Allows for latencies and bandwidths that are significantly better than those for nodes typically used in ASC scale supercomputers
Blue Gene/L Networks
 Each node is attached to 3 main parallel communication networks
– 3D Torus network - peer-2-peer between compute nodes
– Collective network – collective & global communication
– Ethernet network - I/O and management (such as access to any node for configuration, booting and diagnostics )
Blue Gene/L System Software
 System software supports efficient execution of parallel applications
 Compiler support for DFPU (C, C++, Fortran)
 Compute nodes use a minimal operating system called “BlueGene/L compute node kernel”
– A lightweight, single-user operating system
– Supports execution of a single dual-threaded application compute process
– Kernel provides a single and static virtual address space to one running compute process
– Because of single-process nature, no context switching required
– Blue Gene/L System Software contd…
 To allow multiple programs to run concurrently
– Blue Gene/L system can be partitioned into electronically isolated sets of nodes
– The number of nodes in a partition must be a positive integer power of 2
– To run program – reserve this partition
– No other program can use till partition is done with current program
– With so many nodes, component failures are inevitable. The system is able to electrically isolate faulty hardware to allow the machine to continue to run
 Parallel Programming model
– Message Passing – supported through an implementation of MPI
– Only a subset of POSIX calls are supported
– Green threads are also used to simulate local concurrency
Blue Gene/C
 Sister-project to BlueGene/L
 Renamed to Cyclops64
 Massively parallel, supercomputer-on-a-chip cellular architecture
 Cellular architecture gives the programmer the ability to run large numbers of concurrent threads within a single processor.
Blue Gene/P
 Architecturally similar to BlueGene/L
 Expected to operate around one petaflop
 Expected around 2008
Blue Gene/Q
 Last known supercomputer in the Blue Gene series
 Expected to reach 3-10 petaflops