Kitware Develops a Customizable Simulation Framework to Provide HPC for Small to Mid-Sized Manufacturers

May 16, 2013

Kitware today announces new Phase II SBIR funding from the U.S. Department of Energy for the continued development of an open-source, high performance computing (HPC) simulation and computation framework. The developed model-centric integration framework enables researchers to select only the toolkits and applications needed to perform a set of simulation, computation, or analysis tasks.

Manufacturers increasingly rely on simulation tools for rapid prototyping as they have proven to reduce cost and time to market and have resulted in safer, more effective products. However, due to the size and complexity of simulation data, HPC resources are required to truly benefit from these advanced capabilities – making it challenging for small and medium-sized manufacturers to be competitive.

Phase I of this project led to the creation of the Simulation Model Based Architecture (SiMBA) framework for supporting the entire simulation lifecycle, the Simulation Modeling Toolkit (SMTK) which provides an abstract layer for the model geometry, and an attribute data manager for storing the necessary simulation information.

Phase II will focus on enhancing the current system, in addition to enabling the easy creation and providing support for the execution of applications in an HPC environment. The result will be a model-centric simulation framework for building simplified applications that are customized to a specific workflow. This approach has numerous benefits including the use of HPC computational tools for a wide range of users and skillsets, simplified use of the HPC computing environment, and the ability to quickly build customized applications by integrating only the modules that are needed.

“We’re thrilled with the enthusiasm we’ve seen in regards to our model-centric framework,” said Robert O’Bara, Technical Leader at Kitware and Principal Investigator on this project. “The interchangeability of applications in our approach can result in any number of different simulation workflows to meet end-user requirements.”

To learn more about Kitware’s Scientific Computing expertise and how it can be leveraged to meet the needs of your organization through custom development and consulting support, please visit our website or contact kitware@kitware.com.

This work is supported by the U.S. Department of Energy under Contract No. DOE-SC0007615. The views, opinions and/or findings contained in this report are those of the author(s) and should not be construed as an official DOE position, policy, or decision unless so designated by other documentation.

Leave a Reply