Diogenes
Authors:
Open Access
- 17 November 2019
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-20
Abstract
GPU accelerators have become common on today's leadership-class computing platforms. Exploiting the additional parallelism offered by GPUs is fraught with challenges. A key performance challenge faced by developers is how to limit the time consumed by synchronization and memory transfers between the CPU and GPU. We introduce the feed-forward measurement (FFM) performance tool model that automates the identification of unnecessary or inefficient synchronization and memory transfer, providing an estimate of potential benefit if the problem were fixed. FFM uses a new multi-stage/multi-run instrumentation model that adjusts instrumentation based application behavior on prior runs, guiding FFM to problematic GPU operations that were previously unknown. The collected data feeds a new analysis model that gives an accurate estimate of potential benefit of fixing the problem. We created an implementation of FFM called Diogenes that we have used to identify problems in four real-world scientific applications.Keywords
Funding Information
- Oak Ridge National Laboratory (4000164398,4000151982)
- Lawrence Livermore National Laboratory (B617863,B634650)
- National Science Foundation (ACI-1449918)
This publication has 16 references indexed in Scilit:
- The MovieLens DatasetsACM Transactions on Interactive Intelligent Systems, 2015
- Score-P: A Joint Performance Measurement Run-Time Infrastructure for Periscope, Scalasca, TAU, and Vampir2012
- OpenACC — First Experiences with Real-World Applications2012
- Parallel Performance Measurement of Heterogeneous Parallel Systems with GPUs2011
- OpenMP for Accelerators2011
- The Tau Parallel Performance SystemThe International Journal of High Performance Computing Applications, 2006
- Tools for application-oriented performance tuning2001
- A Tool Framework for Static and Dynamic Analysis of Object-Oriented Software with Templates2000
- Dynamic control of performance monitoring on large scale parallel systems1993
- Quartz: a tool for tuning parallel program performance1990