CIMdata PLM Industry Summary Online Archive

6 December 2005

Product News

Noesis Solutions Unleashes OPTIMUS Power at AUDI

Noesis Solutions announced the availability of the Parallel Modules of OPTIMUS 5.1 tightly integrated with Platform Computing's LSF grid management solution. The combination of OPTIMUS and Platform LSF allows organizations to utilize existing software and hardware resources much more effectively in developing better-performing and more reliable products.

Several major automotive OEMs located both in Europe and in the US, have successfully implemented OPTIMUS Parallel on top of their existing Platform LSF networks. Mr. Michael Kaufmann of Audi AG comments, "We were able to demonstrate that it becomes feasible and practical to perform large-scale multi-disciplinary optimization problems - including Crash, NVH, Passenger and Pedestrian Safety applications - with OPTIMUS on Platform LSF-enabled networks. OPTIMUS is a powerful environment to perform MDO analyses on multiple clusters. Even with individual simulation times ranging from several minutes to days idle times on the cluster nodes can be reduced significantly. It thus became possible to compress our turn-around times for Multi-Disciplinary Optimization."

"Performing Optimization, Design of Experiment or Robust Design studies requires the submission of multiple "virtual experiment" runs. The OPTIMUS Parallel Modules allow for a transparent submission of these multiple runs on a computer network on which Platform LSF has been installed", says Mr. Nick Tzannetakis, CTO of Noesis Solutions. "The Deep-Level Parallelization capability of OPTIMUS truly exploits the power of parallelization in a Multi-Disciplinary Optimization (MDO) environment."

Typically, an MDO process contains a sequence of multiple analysis steps. The classical approach will sequentially submit the analysis jobs one after the other. Recent approaches are parallelizing and running the complete MDO sequence on multiple CPUs. This approach often leads to significant idle times on several nodes of the network and under-utilization of the software resources. In contrast OPTIMUS' Deep-Level Parallelization assigns individual Platform LSF queues for each analysis code that is part of the total workflow. OPTIMUS takes care of the dependencies and checks whether results of earlier analyses are available before launching the subsequent analyses. All this results in an optimal usage of the available software licenses on the network and tremendously reduces the total MDO turnaround times.

The OPTIMUS Parallel Modules integrated with Platform LSF 6.1 are available on Windows XP, Linux and UNIX platforms.

Become a member of the CIMdata PLM Community to receive your daily PLM news and much more.

Tell us what you think of the CIMdata Newsletter. Send your feedback.

CIMdata is committed to your privacy. Your personal information will never be sold or shared outside of CIMdata without your express permission.

Subscribe