COMPASS: A Framework for Automated Performance Modeling and Prediction
- ORNL
Flexible, accurate performance predictions offer numerous benefits such as gaining insight into and optimizing applications and architectures. However, the development and evaluation of such performance predictions has been a major research challenge, due to the architectural complexities. To address this challenge, we have designed and implemented a prototype system, named COMPASS, for automated performance model generation and prediction. COMPASS generates a structured performance model from the target application's source code using automated static analysis, and then, it evaluates this model using various performance prediction techniques. As we demonstrate on several applications, the results of these predictions can be used for a variety of purposes, such as design space exploration, identifying performance tradeoffs for applications, and understanding sensitivities of important parameters. COMPASS can generate these predictions across several types of applications from traditional, sequential CPU applications to GPU-based, heterogeneous, parallel applications. Our empirical evaluation demonstrates a maximum overhead of 4%, flexibility to generate models for 9 applications, speed, ease of creation, and very low relative errors across a diverse set of architectures.
- Research Organization:
- Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
- Sponsoring Organization:
- USDOE
- DOE Contract Number:
- AC05-00OR22725
- OSTI ID:
- 1265688
- Resource Relation:
- Conference: the 29th ACM on International Conference on Supercomputing - Newport Beach, California, United States of America - 6/8/2015 12:00:00 AM-6/11/2015 12:00:00 AM
- Country of Publication:
- United States
- Language:
- English
LogP: towards a realistic model of parallel computation
|
conference | January 1993 |
Multifacet's general execution-driven multiprocessor simulator (GEMS) toolset
|
journal | November 2005 |
Keeneland: Bringing Heterogeneous GPU Computing to the Computational Science Community
|
journal | September 2011 |
Inferred Models for Dynamic and Sparse Hardware-Software Spaces
|
conference | December 2012 |
Cetus: A Source-to-Source Compiler Infrastructure for Multicores
|
journal | December 2009 |
PEMOGEN: automatic adaptive performance modeling during program runtime
|
conference | January 2014 |
OpenARC: open accelerator research compiler for directive-based, efficient heterogeneous computing
|
conference | January 2014 |
Rose: Compiler Support for Object-Oriented Frameworks
|
journal | June 2000 |
LogGP: incorporating long messages into the LogP model---one step closer towards a realistic model for parallel computation
|
conference | January 1995 |
Pin: building customized program analysis tools with dynamic instrumentation
|
conference | January 2005 |
The structural simulation toolkit
|
journal | March 2011 |
Aspen: A domain specific language for performance modeling
|
conference | November 2012 |
Using simulation to design extremescale applications and architectures: programming model exploration
|
journal | March 2011 |
A bridging model for parallel computation
|
journal | August 1990 |
Accurate and efficient regression modeling for microarchitectural performance and power prediction
|
conference | January 2006 |
Similar Records
Parallel beamlet dose calculation via beamlet contexts in a distributed multi‐GPU framework
HPC-Colony: Services and Interfaces to Aupport Systems With Very Large Numbers of Processors