A Codesign Framework for Online Data Analysis and Reduction

Citation

K. Mehta et al., "A Codesign Framework for Online Data Analysis and Reduction," 2019 IEEE/ACM Workflows in Support of Large-Scale Science (WORKS), Denver, CO, USA, 2019, pp. 11-20, doi: 10.1109/WORKS49585.2019.00007.

Abstract

In this paper we discuss our design of a toolset for automating performance studies of composed HPC applications that perform online data reduction and analysis. We describe Cheetah, a new framework for performing parametric studies on coupled applications. Cheetah facilitates understanding the impact of various factors such as process placement, synchronicity of algorithms, and storage vs. compute requirements for online analysis of large data. Ultimately, we aim to create a catalog of performance results that can help scientists understand tradeoffs when designing next-generation simulations that make use of online processing techniques. We illustrate the design choices of Cheetah by using a reaction-diffusion simulation (Gray-Scott) paired with an analysis application to demonstrate initial results of fine-grained process placement on Summit, a pre-exascale supercomputer at Oak Ridge National Laboratory. keywords: {data analysis;data reduction;data visualisation;hardware-software codesign;parallel machines;parallel processing;codesign framework;online data analysis;toolset;HPC applications;online data reduction;coupled applications;Cheetah facilitates;compute requirements;online analysis;designing next-generation simulations;online processing techniques;reaction-diffusion simulation;analysis application;fine-grained process placement;performance studies automation;exascale; cheetah; savanna; codar; workflows; in situ; online; reduction; codesign; summit},URL: https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8943548&isnumber=8943488

Read Publication Related Projects CODAR: Center for Online Data Analysis and Reduction