Skip to main content

Kombyne: Accelerated HPC Workflows via Production "In Transit"

Speaker: Steve M. Legensky, Intelligent Light

Presentation

Seminar Slide Deck (PDF-2.2MB)

Abstract

Over the past decade, the convergence of accessible HPC and high-fidelity simulation has created a bottleneck in analysis workflows: writing, managing and reading very large files. Intelligent Light has provided in situ solutions via VisIt/Libsim used to access information directly from the memory of the running solver code. Based on our experience and customer input, we built Kombyne from the ground up to simplify the integration into solver codes and minimize the runtime impact on memory and performance via true "in transit" operation. An in transit workflow uses a separate process that quickly receives data from the solver code and performs visualization and analysis without interfering with the running solver. This process, called an Endpoint, can directly output extracts such as iso surfaces, cutting planes or point samples for data science and can render images as well. The Endpoint can also act as a bridge to popular visualization codes such as FieldView, VisIt and ParaView, enabling interactive visualization without stopping the solver code. Kombyne is the first commercially supported product that provides full in situ and in transit post-processing, rendering and data science sampling.

Biography

Photo of Steve M. Legensky
Steve M. Legensky is the President, CTO and founder of Intelligent Light, Transformative Tools for Engineering. He has worked in visualization and data management since the mid-70s and is responsible for the product launches of Intelligent Light’s original 3DV products, P3/Animation, licensed to PDA Engineering and FieldView, which was spun off in October of 2019. Steve has led the company’s efforts in Data Science applications in engineering. He is a global speaker on data science for CFD [Japan’s NAGARE (Dec. 2017), Parallel CFD (2018) and SIAM (Feb. 2019)], visualization, data management, and post-processing, and is an Associate Fellow of the AIAA.