© 2006 ibm corporation computación de alto rendimiento en ibm research septiembre 2006 dr. josé...

© 2006 IBM Corporation

Computación de Alto Rendimientoen IBM ResearchSeptiembre 2006

Dr. José G. Castañ[email protected] T.J. Watson Research Center

IBM Research


El Projecto “Blue Gene”

En Diciembre 1999, IBM Research anuncia el Blue Gene Producir nuevos adelantos en simulaciones biomoleculares Investigar nueva tecnologias en hardware y software para producir computadoras

de alto rendimiento

Blue Gene sigue un enfoque modular, donde el bloque basico (o célula) puede copiarse ad infinitum

Procesadores de bajo consumo – permite rendimientos combinados mas altos

– PowerPC 440 System-on-a-chip ofrece ventajas en costo/rendimiento

– Menor complejidad– Alta densidad (2048 procesadores por rack, enfriado por aire)– Redes integradas para gran escala

Ambiente de software familiar, simplificado para HPC Mucha atención a RAS (“reliability, availability, and serviceability”) en todo el

sistema

IBM Research


El Chip Blue Gene/L (ASIC)

PLB (4:1)

“Double FPU”

Ethernet Gbit

JTAGAccess

144 bit wide DDR256/512MB

JTAG

Gbit Ethernet

440 CPU

440 CPUI/O proc

L2

L2

MultiportedSharedSRAM Buffer

Torus

DDR Control with ECC

SharedL3 directoryfor EDRAM

Includes ECC

4MB EDRAM

L3 CacheorMemory

6 out and6 in, each at 1.4 Gbit/s link

256

256

1024+144 ECC256

128

128

32k/32k L1

32k/32k L1

“Double FPU”

256

snoop

Tree

3 out and3 in, each at 2.8 Gbit/s link

GlobalInterrupt

4 global barriers orinterrupts

128

• IBM CU-11, 0.13 µm• 11 x 11 mm die size• 25 x 32 mm CBGA• 474 pins, 328 signal• 1.5/2.5 Volt

IBM Research


Arquitectura de Blue Gene/L

IBM Research


Blue Gene/L en Lawrence Livermore National Laboratory

BG/L

Number of Racks 64

Number of Nodes 65536

Processor Frequency 700 Mhz

Peak Performance

Rack 5.7 TF

Machine 360 TF

Linpack 280 TF

App 101 TF

Memory Rack 256 GB

Machine 16 TB

Power ~2 MW

Size 250 sq.m.

Bisection Bandwidth 700 GB/s

Cables Number 5,000

Length 25 km

Storage Disk 200 TB

IBM Research


Source: www.top500.org

# Ven-dor

Rmax TFlops

Installation

1 IBM 280.6DOE/NSSA/LLNL

(64 racks BlueGene/L)

2 IBM 91.2BlueGene at Watson


3 IBM 75.8ASC Purple LLNL

(1280 nodes p5 575)

4 SGI 51.9 NASA/Columbia (Itanium2)

5 Bull 42.90CEA/DAM Tera10

(Itanium2)

6 Dell 38.27Sandia -Thunderbird(EM64T/Infiniband)

7 Sun 38.18Tsubame Galaxy TiTech

(Opteron/Infiniband)

8 IBM 37.33FRZ – Juelich


9 Cray 36.19Sandia – Red Storm

(XT3 Opteron)

10 NEC 35.86Japan Earth Simulator

(NEC)

# Ven-dor

Rmax TFlops

Instalation

11 IBM 27.91MareNostrum Barcelona

Supercomputer (JS20)

12 IBM 27.45ASTRON Netherlands (6 racks BlueGene/L)

13 Cray 20.52ORNL – Jaguar (XT3 Opteron)

14 Calif Dig 19.94LLNL

(Itanium2)

15 IBM 18.20AIST - Japan

(4 rack BlueGene/L)

16 IBM 18.20EPFL - Switzerland(4 rack BlueGene/L)

17 IBM 18.20KEK – Japan

(4 rack BlueGene/L)

18 IBM 18.20KEK – Japan

(4 rack BlueGene/L)

19 IBM 18.20IBM – On Demand Ctr(4 rack BlueGene/L)

20 Cray 16.97ERDC MSRC

(Cray XT3 Opteron)

Blue Gene en los Top500

IBM Research


Motivacion del Software de Sistema

Nodos de Computación dedicados a ejecutar una sola aplicacion, and casi nada más

Compute node kernel (CNK) Simplicidad!

Nodos de I/O corren Linux and proveen servicios de OS – files, sockets, comenzar programas, señales, debugging, and fianalización de tareas

Solution estandar: Linux

Nodos de Servicio ejecutan todos los servicios de administración (e.g., latidos, checkean errores)

transparente para el programa de los usuarios

IBM Research


Blue Gene/L: Architectura del Software de Sistema

Functional Gigabit Ethernet

Functional Gigabit Ethernet

I/O Node 0

Linux

ciod

C-Node 0

CNK

I/O Node 1023

Linux

ciod

C-Node 0

CNK

C-Node 63

CNK

C-Node 63

CNK

Control Gigabit

Ethernet

Control Gigabit

Ethernet

IDo chip

LoadLeveler

SystemConsole

CMCS

JTAG

torus

tree

DB2

Front-endNodes

Pset 1023

Pset 0

I2C

FileServers

fs client

fs client

Service Node

app app

appapp

IBM Research


Classical MD – ddcMD2005 Gordon Bell Prize Winner!!

Scalable, general purpose code for performing classical molecular dynamics (MD) simulations using highly accurate MGPT potentials

MGPT semi-empirical potentials, based on a rigorous expansion of many body terms in the total energy, are needed in to quantitatively investigate dynamic behavior of d-shell and f-shell metals.

524 million atom simulations on 64K nodes achieved 101.5 TF/s sustained. Superb strong and weak scaling for full machine - (“very impressive machine” says PI)

Visualization of important scientific findings already achieved on BG/L: Molten Ta at 5000K demonstrates solidification during isothermal compression to 250 GPa

2,048,000 Tantalum atoms

IBM Research


Resolidificación Rapida del Tantalum (ddcMD)

Nucleation of solid is initiated at multiple independent sites throughout each sample cellGrowth of solid grains initiates independently, but soon leads to grain boundaries which span the simulation cell: size of cell is now influencing continued growth2,048,000 simulation recently performed indicates formation of many more grains

© 2006 ibm corporation computación de alto rendimiento en ibm research septiembre 2006 dr. josé...

Documents