FinisTerrae

From Wikipedia, the free encyclopedia

Finisterrae is the 100th supercomputer in Top500 ranking. It is also the third most powerful supercomputer in Spain (after MareNostrum and Magerit, and it's placed in Galicia.

This project is stimulated by the Autonomous Government of Galicia and the Top Council of Scientific Researches(CSIC) in the year 2006 with scientific and research I+D+I purposes.

It is estimated that the base project will be completed in 2010 which is foreseen that reaches TOP10 of the list of the most powerful computers of the world when it reaches its full performance providing services of intensive calculation and advanced communications.

[edit] Architecture

FinisTerrae overview

Name	Finisterrae
Place	CESGA
Country	Spain
Year	2007
OS	Linux
Architecture	Itanium 64 (Intel)
Processors	2400
Memory	19584 GB
Power	12970 GFlops
Top500 Ranking	100
Web	www.cesga.es

FinisTerrae supercomputer, installed in CESGA is an integrated system by shared-memory nodes with and SMP NUMA architecture. FinisTerrae is composed of

144 computational nodes:

• Template:Convert/HP Integrity rx7640 nodes with 16 Itanium Montvale cores and 128 GB memory each one.
• Template:Convert/HP Integrity Superdome node, with 128 Itanium Montvale cores and 1024 GB memory.
• Template:Convert/HP Integrity Superdome node, with 128 Itanium2 cores and 384 GB memory.

An hierarchic storing system with:

• 22 nodes for storing management with 96 processing cores.
• 390.000 GB in disk.
• 2.200.000 GB in automated tape.

An interconnecting network between Infiniband 4x DDR at 20 Gbit/s nodes.

An external network at 10 Gbit/s.

System supports multiple operating systems like: Linux, Unix and Windows.

FinisTerrae includes open software and standard like: Linux, Lustre, Grid Engine and Globus.

System also has next compilators, libraries and development tools: Intel C/C y Fortran, Intel MKL, Vtune, HP-MPI y HP UPC.

[edit] External links & References

• CESGA

v • d • e Parallel computing topics General High-performance computing Parallelism Bit-level parallelism · Instruction level parallelism · Data parallelism · Task parallelism Theory Speedup · Amdahl's law · Flynn's taxonomy (SISD • SIMD • MISD • MIMD) · Cost efficiency · Gustafson's law · Karp-Flatt metric · Parallel slowdown Elements Process · Thread · Fiber · Parallel Random Access Machine Coordination Multiprocessing · Multithreading · Multitasking · Memory coherency · Cache coherency · Barrier · Synchronization · Distributed computing · Grid computing Programming Programming model · Implicit parallelism · Explicit parallelism Hardware Computer cluster · Beowulf · Symmetric multiprocessing · Non-Uniform Memory Access · Cache only memory architecture · Asymmetric multiprocessing · Simultaneous multithreading · Shared memory · Distributed memory · Massively parallel processing · Superscalar processing · Vector processing · Supercomputer · Stream processing · GPGPU Software Distributed shared memory  · Application checkpointing · Warewulf APIs POSIX Threads · OpenMP · Message Passing Interface (MPI) · Intel Threading Building Blocks Problems Embarrassingly parallel · Grand Challenge · Software lockout