Home
 
Getting Started
 
Access Information
 
Code Development
 
Computing Resources
 
Running Jobs
 
Documentation and Training
 
Events
 
Careers at A*CRC
 
   
Computing Resources  
 
Computing Resources
ACRC operates, manages and provisions computing resources comprising several machines of various capacity, software, storage and networking.  The A*CRC team has collectively a vast pool of the expert HPC knowledge and is ready to assist and support the users in attaining maximum performance and productivity in their computational research. Additionally, superb visualization hardware, equipment and specialized software and support is available at Advanced Computing Program, Institute of High Performance Computing.
 
Hardware
The hardware currently available to A*CRC users fall into three main categories:
  • Cluster of large number of small, dual processor nodes for heavy loads of many serial jobs as well as highly parallel codes that run well with MPI (Fuji);
  • Cluster of smaller number of larger nodes, more suitable for large memory-footprint applications with higher degree of CPU and thread communication (Cirrus);
  • SMP machine, aimed at very large tera-scale tasks in terms of both processor and memory load (Aurora).

The detailed, comparative list of all A*CRC machines is at Hardware page.
 

System Name CPU Platform Clock Speed [GHz] OS # Nodes # of Cores/Nodes Total CPU
Cumulus IBM PowerPC A2
1.6
RHEL 6.3 512 (compute card)
16
8192
Cirrus IBM Power7
3.3
AIX 6.1 30
32
960
Aurora Xeon X7542 2.67 SuSE Linux Enterprise 11 2 2048
576
2624
Fuji Xeon X5570 2.97 Redhat Enterprise Linux 5.3 378
72 
8
12 
3888

  
Software
ACRC provides system software, compilers, middleware, open-source tools, and mathematical libraries. A*CRC staff helps to install and maintain application software. It is a sole responsibility of a user or a research program to purchase and maintain a license of any commercial application software that they require.

For complete information refer to Software page.

  
Storage
All A*CRC cluster systems have substantial amounts of adequate storage provided, both for local per-node boot and scratch requirements, as well as for the permanent shared store. A variety of file systems  are used on both the cluster and SMP systems and are described in File Systems page.

Storage page contains the comprehensive list of the available storage resources on each of our systems, as well as the dedicated PetaByte scale storage facilities that are becoming increasingly critical in research areas such as genomics and bio-imaging.
 
 
Visualisation
Realistic real-time 3-D presentation is indispensable in pre- and post- processing of huge datasets resulting from the large-scale computations and simulations. A*CRC users may gain access to dedicated visualization resources, including multimega-pixel, six-Tile Display wall, VR development, visualization software and supporting expertise from Advanced Computing Program at the Institute of High Performance Computing. Further information is contained in Visualisation page.

  
Expert Knowledge
Installation and porting of codes, selecting the best hardware for the combination of the type of computational problem and application software, parallelization of codes, tuning, optimizing, performance profiling and monitoring, or even simply running a complex HPC applications are tasks requiring a very wide range of specialized expertise. It is almost impossible to expect a computational researcher, a user of A*CRC systems, to have mastery of all intermediate software, tools, application programming interfaces and techniques.

ACRC team collectively possesses extensive HPC expertise to assist users in the above issues. The new software group is being formed to further strengthen this support capability.

Page Expert Knowledge contains more information.
Last Updated - 29th Nov 2016
 
     
Privacy Policy