High Performance Computing | News
Platform Computing HPC Management Software Enhances App Support and Job Scheduling
- By Dian Schaffhauser
A company that specializes in applications for managing high performance computing (HPC) clusters has upgraded its HPC management application. Platform Computing's latest version of Platform HPC sports enhanced support for applications and job scheduling, as well as a more user-friendly Web interface. At the same time the company has also improved its workload scheduling products to better manage graphics processing units (GPUs) where available in clusters.
The new release of Platform HPC includes better support for applications, including job submission templates for commonly used commercial applications and scripting guidelines for simplified job submission; job scheduling for graphics processing units and expanded monitoring capabilities for those jobs; and an interface intended to simplify submission, monitoring, and management of jobs.
According to the company, the new version provides easier cluster provisioning, workload management, reporting and monitoring, and message passing interface libraries, all of which are accessible via the Web interface.
"GreenPlanet HPC [Cluster] leverages Platform Computing's software to assist interdisciplinary scientific collaborators across process intensive computing projects, such as climate modeling simulations done by our Earth Systems Science researchers, to predict worldwide weather patterns," said Ronald Hubbard, executive director of GreenPlanet HPCC at the University of California, Irvine School of Physical Sciences. "We place tremendous added value in having a robust and complete cluster solution that ensures our scientists can administer the system independently while accelerating their calculation times so that their research can be performed in a more productive,
efficient, cost-effective manner."
Platform's workload scheduler programs, Platform HPC, Platform LSF, and Platform Symphony, now have improvements for managing computing jobs optimized for graphics processing units (GPUs). For example, now Platform's users can use the newly updated software to allocate jobs only to resources that have GPUs.
Platform products currently support Nvidia's Tesla GPU computing processors, with plans to support future GPU accelerators when they become available.
Dian Schaffhauser is a senior contributing editor for 1105 Media's education publications THE Journal and Campus Technology. She can be reached at firstname.lastname@example.org or on Twitter @schaffhauser.