Home
 
Getting Started
 
Access Information
 
Code Development
 
Computing Resources
 
Running Jobs
 
Documentation and Training
 
Events
 
Careers at A*CRC
 
   
Hardware  
 
Cluster systems are the main hardware resource in A*CRC. Ranging in peak performance from 1 to 100 TFLOPs, these systems are meant to host a variety of tasks from large number of small (serial) jobs to large, multi-thread, multi-task or/and large memory-footprint programs that require very large resources.
 
Here is the detailed list of the computing resources of A*CRC:

SYSTEM NAME
Cumulus Cirrus Aurora Fuji
Axle
HOST NAME
cumulus.acrc.a-star.edu.sg cirrus.acrc.a-star.edu.sg
aurora.acrc.a-star.edu.sg
fuji.acrc.a-star.edu.sg
axle.acrc.a-star.edu.sg
Brand / Model IBM Bluegene/Q IBM P755
SGI Altix UV 1000
Fujitsu BX900
HP ProLiant DL785
CPU PLATFORM & GHZ
IBM PowerPC A2
1.6Ghz 
IBM Power7 3.3GHz
Intel Xeon X7542 2.67GHz
Intel Xeon 5570 2.93 Ghz
AMD Opteron 8389 2.9 GHz
SYSTEM ARCHITECTURE
custom cluster
SMP
cluster
cluster
OPERATING SYSTEM
Linux AIX
Linux
Linux
Linux
OS VERSION
RHEL 6.3 AIX 6.1
SUSE Linux Enterprise Server 11
RHEL 5.3
CentOS 5.6
NUMBER OF NODES
512 (compute card)

30

2

378
72 

32

CORES PER NODE
16

32

2048
576

8
12 

32

TOTAL CPU NUMBER
8192

960

2624

3888

1024

MEMORY PER NODE (GB)
16

128

12288
3145

24
36 

128

TOTAL MEMORY (GB)
8192

3840

15433

11664

4096

TOTAL SHARED DISK (TB)
100 20 Scratch
20 Scratch
15 Scratch 
 12 Scratch (NFS)
INTERCONNECT
IB Switch with 5D Torus Dual Infiniband DDR
NUMALink 5
IB QDR
Infiniband DDR
COMPILERS
IBM XL Compiler IBM XL Compiler C, C++ Fortran, UPC
C/C++ Fortran
C, C++ Fortran
SCHEDULERS
SLURM Platform LSF
Platform LSF Platform LSF
Platform LSF
MATH LIBRARIES
ESSL ESSL
MKL, SGI Propack
ESL, MKL
MKL, ACML
Last Updated - 3rd Jul 2013