Abstract.

The paper presents the problem of high-performance computer clustering to operate the current CAE systems. The main issues being the matter of IT services when solving this problem as well as the methods of their solution are given. The principles of HPC clusters designing in view of the specific character of the problems to be solved are considered. The main subsystems of the compute cluster are shown. The overview of the tools for complex server infrastructure operation computerization, support, and maintenance is given.

Keywords:

CAE, HPC-cluster, system administration, monitoring, distributed resource management.

PP. 10-16.

REFERENCES

1. Computer-aided engineering [Online resource]. [https://ru.wikipedia.org/wiki/Computer-aided_engineering]. Link checked 03.03.3017.
2. Infiniband [Online resource] [https://ru.wikipedia.org/wiki/InfiniBand]. Link checked 03.03.3017.
3. Oracle EDR Infiniband Fabric [Online resource] [https://www.oracle.com/ru/corporate/pressrelease/edr-infiniband-fabric-20160222.html]. Link checked 03.03.3017.
4. XCAT [Online resource] [https://xcat.org/]. Link checked 03.03.3017.
5. OpenLDAP [Online resource] [http://www.openldap.org/]. Link checked 03.03.3017.
6. Nagios [Online resource] [https://www.nagios.org/]. Link checked 03.03.3017.
7. Ganglia [Online resource] [http://ganglia.info/]. Link checked 03.03.3017.
8. Matt Butcher. Mastering OpenLDAP. 2007 Packt Publishing – 482 s.
9. Matt Massie, Bernard Li, Brad Nicholes, and Vladimir Vuksan. Monitoring with Ganglia. O’Reilly Media, 2012 – 254 s.
10. Evy Nemeth, Garth Snyder, Trent R. Hein, Ben Whaley. Unix and Linux system administration handbook. “I.D. Williams”, 2012 – 1312 s.