Editor’s Pick: Altair Unveils PBS Professional 11.0 HPC Workload Manager

Includes optimized PetaFLOPS scalability, GPU and topology scheduling.

Includes optimized PetaFLOPS scalability, GPU and topology scheduling.

By Anthony J. Lockwood

Dear Desktop Engineering Reader:

 

To borrow a convention, this is the best of times and the worst of times to be an engineer. The best means software: CAE applications are becoming stunningly robust and widely accessible through SaaS (software as a service). The best also means hardware: HPC (high-performance computing) and cloud computing are becoming affordable and widely accessible. The worst of the times means both software and hardware: With every technological breakthrough in your CAE application suite and HPC or cloud investments, your workloads become more complex and ever more difficult to manage. It’s enough to scare the dickens out of any ‘umble IT administrator.

This is why I found the recent announcement of PBS Professional 11.0 from Altair Engineering interesting. The skinny on PBS Professional is that it is a workload manager intended to increase your software utilization and help improve overall productivity (as well as your ROI) by giving you the tools to tame all that goes into managing your distributed assets – hardware and software.

PBS Professional is the keystone of Altair’s PBS Works suite of on-demand cloud computing technologies. In addition to PBS Professional, the suite includes PBS Analytics, a data analysis and visualization solution to support data-driven planning and decision making, and PBS Catalyst, which helps to make managing distributed server and application configurations easier. The key takeaway on PBS Works is that, the company says, it is the most widely implemented software environment for grid-, cloud-, cluster-, and on-demand computing. It’s deployed at more than 1,400 sites worldwide, and NEC and SGI use PBS Professional as their workload management tool. In short, PBS Works,  and by extension PBS Professional, are becoming something of the standard for HPC workload and resource management.

So, what’s up with version 11.0? Faster than earlier versions: Faster scheduling, faster job submission, and faster cold starts. There’s new support for Cray systems and optimization of task placement for major HPC network topologies like GigE, IBM, and Infiniband. Perhaps most importantly, PBS Professional supports GPU (graphic processing units)  configuration and scheduling. By enabling your disparate CPUs and GPUs to function as a unified computing platform, PBS Professional 11.0 can increase your application speed and lower power requirements dramatically.

The raw power advantages of HPC and cloud computing are legion, especially in the engineering environment. Channeling that power to maximize productivity and your investment is no easy thing. PBS Professional could save you a heap of hassles by providing you the tools to develop the intelligent policies to manage your mixed-vendor hardware and software assets efficiently. You can read more about PBS Professional version 11.0 from today’s Pick of the Week write-up. Well worth your time.

Thanks, pal.—Lockwood

Anthony J. Lockwood
Editor at Large, Desktop Engineering

Read today’s Pick of the Week write-up.

This is sponsored content. See how it works.

Share This Article

Subscribe to our FREE magazine, FREE email newsletters or both!

Join over 90,000 engineering professionals who get fresh engineering news as soon as it is published.


About the Author

Anthony J. Lockwood's avatar
Anthony J. Lockwood

Anthony J. Lockwood is Digital Engineering’s founding editor. He is now retired. Contact him via [email protected].

Follow DE
#5886