Unified Code Generation for the Parallel Computation of Pairwise Interactions using Partial Evaluation

Conference contribution
(Abstract of a poster)


Publication Details

Author(s): Schmitt J, Köstler H, Eitzinger J, Membarth R, Pérard-Gayot A
Publication year: 2018


Abstract

The evaluation of pairwise interactions is fundamental for the simulation of most molecular processes. Their efficient computation is therefore crucial for the overall
performance of these simulations on modern computer architectures. We show how code for the computation of pairwise interactions on
parallel and heterogeneous platforms can be generated from a unified
base through partial evaluation of higher-order functions using the
AnyDSL framework. We assess the performance of our implementation in a number of test cases on modern CPU and GPU hardware.


FAU Authors / FAU Editors

Eitzinger, Jan Dr.
Regionales Rechenzentrum Erlangen (RRZE)
Köstler, Harald Prof. Dr.
Lehrstuhl für Informatik 10 (Systemsimulation)
Schmitt, Jonas
Lehrstuhl für Informatik 10 (Systemsimulation)


External institutions
Universität des Saarlandes
Universität des Saarlandes (UdS)


How to cite

APA:
Schmitt, J., Köstler, H., Eitzinger, J., Membarth, R., & Pérard-Gayot, A. (2018). Unified Code Generation for the Parallel Computation of Pairwise Interactions using Partial Evaluation. Poster presentation at International Symposium on Computational Science at Scale (CoSaS), Erlangen.

MLA:
Schmitt, Jonas, et al. "Unified Code Generation for the Parallel Computation of Pairwise Interactions using Partial Evaluation." Presented at International Symposium on Computational Science at Scale (CoSaS), Erlangen 2018.

BibTeX: 

Last updated on 2019-24-04 at 05:21