Home / Editor's Pick of the Week / Editor’s Pick: New ANSYS HPC to Boost Productivity for Simulation-driven Product Development

Editor’s Pick: New ANSYS HPC to Boost Productivity for Simulation-driven Product Development

By Anthony J. Lockwood

Dear Desktop Engineering Reader:

I’ve made no bones about it in the past. I think HPC (high-performance computing) and simulation are a match made in heaven. Bob Eubanks should get the two of them on the “Newlywed Game” so that everyone can see it as far as I am concerned. And ANSYS just one upped the relationship with what  they call ANSYS HPC software.

Basically, ANSYS HPC is a scalable HPC solution that supports the full ANSYS portfolio and enables you to do multiple physics simulations on your workstation or your cluster. And that means that ANSYS HPC makes accessible larger-scale parallel processing for simulation for companies large and small. It does so in a couple of ways: scalability and multiphysics.

Now, multiphysics means that rather having single applications, ANSYS HPC provides the parallel computing capability that brings together multiple physics applications—structural, fluids, thermal, and electromagnetics simulation—in a single solution. By combining complex physics phenomena, you can address highly complex analyses in your product development that have exceeded your grasp or required multiple runs of different solvers. ANSYS HPC is about efficiency, because you do not have to get the data from one solver to another. ANSYS HPC does that for you.

Scalability takes a couple of forms. First though, let me talk about accessibility. ANSYS HPC is designed so that in a few clicks you can set up and launch jobs in parallel on your multicore workstation or on a remote compute cluster. It supports load managers and job schedulers like Platform LSF and it offers capabilities so that you can integrate it into your workload management environment without a whole lot of fuss.

Now, scalability. ANSYS HPC will provide parallel processing on your multicore workstation and it scales for clusters running Linux or Microsoft Windows HPC Server 2008. ANSYS has some benchmarks, which you’ll find linked off today’s Pick of the Week, showing fluids jobs nearly linear scaling out to 2048 processors.

And that’s the deal in a nutshell. If you’re a small outfit and you have workstations with, say, 16 or 32 cores, ANSYS HPC gets you high-end multiple physics simulations. If you’re larger and have 128 cores or a huge cluster, ANSYS HPC scales up readily so that you can leverage that power for simulations. Either way, whether large or small, parallel processing enables you to run detailed, multiple physics analyses that leverage your resources efficiently so that you can get a better idea of what it is that you’re designing and how well it will behave in the real world of multiple physics phenomena.

You can learn more about ANSYS HPC from today’s Pick of the Week write-up. Hit the link and sign up for the on-demand webinar or, better yet, sign-up for the live webinar next week. HPC and simulation are really meant for each other. I recommend that you learn how ANSYS HPC helps you match them up.

Thanks, pal. — Lockwood

Anthony J. Lockwood
Editor at Large, Desktop Engineering

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

About Anthony J. Lockwood

Anthony J. Lockwood is Desktop Engineering's Editor-at-Large. Contact him via de-editors@deskeng.com.