In a recent ANSYS survey, when almost 3,000 respondents were asked to name the biggest pressures on their design activities, 52% cited “reducing the time required to complete design cycles.” At the same time, 28% of respondents named “producing more reliable products that result in lower warranty-related costs” as a chief concern.
Engineers want simulations to be faster. Fast simulations can impact product design, enables the possibility of simulating larger models and can open the door to doing parametric (or design variation) studies.
Engineers today also want to model “larger” models. This means doing entire assemblies instead of individual parts. It means meshing CAD models instead of spending days/weeks to remove small features and “fix” CAD models to they can be meshed in CFD software. And it means have more cells in the model to better capture flow phenomena. Similarly engineers also want to extend their simulations to handle complicated physics (e.g. more complicated turbulence solutions, multiple physics and particle tracking simulations).
Last but not least, engineers want to evaluate many more design ideas, look for the “best” feasible design and ensure product integrity over a life span.
All of these involve more computations which drives the “need for speed”. The answer is high performance computing (HPC). HPC is about removing the obstacle of limited compute resources from engineers who want to use the full simulation capabilities.
It is possible to significantly reduce the limitations of computer hardware by properly configuring an HPC system. Equally important to achieving this goal is software development both of new algorithms and new implementations that take advantage of parallel processing and other aspects of HPC.
The computer industry continues to deliver enormous increases in computing speed and power at consistently lower costs. The average workstation that you may use today is equivalent in power to the entry-level HPC cluster of just a few years ago. While compute power has increased a thousand-fold over the last decade, large-scale computing is now within the reach of more and more engineering teams, with the promise of new trends, like cloud computing, to make access even more widespread.
Today, parallel simulations using 128-256 compute cores are increasingly common and there is an increasing demand for scaling to thousands of cores per simulation. Organizations like NASA and Airbus are expressing their ‘extreme scalability’ demands for running huge simulation models either in the form of complete aircraft engines, or real-time CFD-based in-flight simulations. The packaging company Tetra Pak wants to be able to tackle package sterilization problems with advanced Large Eddy Simulation (LES) models using ten to one hundred thousand of cores. Note that other perspectives from the industry on extreme scalability can be watched in this webinar.
From software side, ANSYS has responded with consistent, dramatic solution improvements, developed specifically to sustain speed and scaling on the latest HPC platforms. This focus ensures that ANSYS customers benefit from leading performance both today and into the future, as HPC technology continues to evolve.
ANSYS and the National Center for Supercomputing Applications (NCSA) achieved a supercomputing milestone on 17 November 2014, ANSYS® Fluent® scaled to 36,000 CPU cores – an industry first that could lead to greater efficiencies and increased innovation throughout manufacturers’ product development processes.
With this example, ANSYS shows its commitment to HPC by taking its customers to new heights of simulation fidelity and engineering insight. It is clear that HPC is a hardware and software initiative!
“With HPC in mind, hardware is brain and software is soul.”
Author Bio: Vaibhav’s responsibility is leading ANSYS HPC and system support activities in India. He helps ANSYS users to improve their productivity by running the simulation faster, running bigger and bigger simulations and improving their IT infrastructure. He helps them in understanding HPC technology, its setup at their end including after sales support. He works closely with Hardware vendors to provide one stop solution for our customers. He is Microsoft, RedHat and Cisco certified engineer. Vaibhav has a Bachelor’s degree in Electronics & Telecommunications