Analyzing dynamic task-based applications on hybrid platforms: an agile scripting approach
Abstract
In this paper, we present visual analysis techniques to evaluate the performance of HPC task-based applications on hybrid architectures. Our approach is based on composing modern data analysis tools (pjdump, R, ggplot2, plotly), enabling an agile and flexible scripting framework with minor development cost. We validate our proposal by analyzing traces from the full-fledged implementation of the Cholesky decomposition available in the MORSE library running on a hybrid (CPU/GPU) platform. The analysis compares two different workloads and three different task schedulers from the StarPU runtime system. Our analysis based on composite views allows to identify allocation mistakes, priority problems in scheduling decisions, GPU tasks anomalies causing bad performance, and critical path issues.
References
[1]
H. Topcuoglu, S. Hariri, and M.-Y. Wu, "Performance-effective and low complexity task scheduling for heterogeneous computing," IEEE Trans. Par. Distr. Syst., vol. 13, no. 3, pp. 260--274, 2002.
[2]
E. Schulte, D. Davison, T. Dye, and C. Dominik, "A multi-language computing environment for literate programming and reproducible research," J. of Stat. Soft., vol. 46, no. 3, 2012.
[3]
E. Agullo, G. Bosilca, B. Bramas, C. Castagnede, O. Coulaud, E. Darve, J. Dongarra, M. Faverge, N. Furmento, L. Giraud, X. Lacoste, J. Langou, H. Ltaief, M. Messner, R. Namyst, P. Ramet, T. Takahashi, S. Thibault, S. Tomov, and I. Yamazaki, "Poster: Matrices over runtime systems at exascale," in High Performance Computing, Networking, Storage and Analysis (SCC), 2012 SC Companion:, Nov 2012, pp. 1332--1332.
[4]
C. Augonnet, S. Thibault, R. Namyst, and P.-A. Wacrenier, "StarPU: a unified platform for task scheduling on heterogeneous multicore architectures," Conc. and Comp.: Pract. and Exp., vol. 23, no. 2, 2011.
[5]
E. Agullo, J. Demmel, J. Dongarra, B. Hadri, J. Kurzak, J. Langou, H. Ltaief, P. Luszczek, and S. Tomov, "Numerical linear algebra on emerging architectures: The plasma and magma projects," Journal of Physics: Conference Series, vol. 180, no. 1, 2009.
[6]
A. Duran, E. Ayguadé, R. M. Badia, J. Labarta, L. Martinell, X. Martorell, and J. Planas, "OmpSs: a proposal for programming heterogeneous multi-core architectures," Par. Proc. Letters, vol. 21, no. 02, 2011.
[7]
G. Bosilca, A. Bouteiller, A. Danalis, T. Herault, P. Lemarinier, and J. Dongarra, "DAGuE: A generic distributed DAG engine for high performance computing," Parallel Computing, vol. 38, no. 1--2, 2012.
[8]
C. Augonnet, O. Aumage, N. Furmento, R. Namyst, and S. Thibault, "StarPU-MPI: Task programming over clusters of machines enhanced with accelerators," in Proc. European Conf. Recent Advances in the Message Passing Interface (EuroMPI). Springer, 2012, pp. 298--299.
[9]
S. Ohshima, S. Katagiri, K. Nakajima, S. Thibault, and R. Namyst, "Implementation of FEM Application on GPU with StarPU," in SIAM Conference on Computational Science and Engineering, 2013.
[10]
V. Martínez, D. Michéa, F. Dupros, O. Aumage, S. Thibault, H. Aochi, and P. O. A. Navaux, "Towards seismic wave modeling on heterogeneous many-core architectures using task-based runtime system," in Intl. Symp. on Comp. Arch. and High Perf. Comp. (SBAC-PAD). IEEE, Oct. 2015.
[11]
E. Agullo, L. Giraud, A. Guermouche, S. Nakov, and J. Roman, "Task-based Conjugate Gradient: from multi-GPU towards heterogeneous architectures," Inria Bordeaux, Research Report 8912, May 2016.
[12]
X. Lacoste, M. Faverge, P. Ramet, S. Thibault, and G. Bosilca, "Taking advantage of hybrid systems for sparse direct solvers via task-based runtimes," in HCW'2014 workshop of IPDPS. IEEE, 2014, pp. 29--38.
[13]
R. L. Graham, "Bounds for certain multiprocessing anomalies," Bell System Technical Journal, vol. 45, no. 9, pp. 1563--1581, 1966.
[14]
K. Coulomb, M. Faverge, J. Jazeix, O. Lagrasse, J. Marcoueille, P. Noisette, A. Redondy, and C. Vuchener, "Visual trace explorer ViTE."
[15]
L. M. Schnorr, M. Faverge, F. Trahay, B. O. Stein, and J. C. de Kergommeaux, "The Paje trace file format," UFRGS, Tech. Rep., 2016.
[16]
V. Pillet, J. Labarta, T. Cortes, and S. Girona, "Paraver: A tool to visualize and analyze parallel code," in Proceedings of WoTUG-18: Transputer and occam Developments, vol. 44, 1995, pp. 17--31.
[17]
A. Knüpfer, H. Brunst, J. Doleschal, M. Jurenz, M. Lieber, H. Mickler, M. S. Müller, and W. E. Nagel, "The Vampir performance analysis toolset," in Tools for High Perf. Comp. Springer, 2008, pp. 139--155.
[18]
A. Huynh, D. Thain, M. Pericàs, and K. Taura, "DAGViz: A DAG Visualization Tool for Analyzing Task-parallel Program Traces," in Proc. 2nd Workshop Visual Perf. Analysis (VPA'15). ACM, 2015, pp. 3:1--3:8.
[19]
R. Keller, S. Brinkmann, J. Gracia, and C. Niethammer, Temanejo: Debugging of Thread-Based Task-Parallel Programs in StarSS. Springer, 2012, pp. 131--137.
[20]
B. Haugen, S. Richmond, J. Kurzak, C. A. Steed, and J. Dongarra, "Visualizing execution traces with task dependencies," in Proc. 2nd Workshop Visual Perf. Analysis (VPA'15). ACM, 2015, pp. 2:1--2:8.
[21]
L. M. Schnorr and A. Legrand, "Visualizing more performance data than what fits on your screen," in Tools for High Perf. Comp. Springer, 2013, pp. 149--162.
[22]
G. Pagano and V. Marangozova-Martin, "FrameSoC Workbench: Facilitating Trace Analysis through a Consistent User Interface," Inria, Technical Report RT-0447, Apr. 2014.
[23]
V. Danjean, R. Namyst, and P.-A. Wacrenier, "An efficient multi-level trace toolkit for multi-threaded applications," in European Conference on Parallel Processing. Springer, 2005, pp. 166--175.
[24]
E. Agullo, O. Beaumont, L. Eyraud-Dubois, J. Herrmann, S. Kumar, L. Marchal, and S. Thibault, "Bridging the Gap between Performance and Bounds of Cholesky Factorization on Heterogeneous Platforms," in Heterogeneity in Computing Workshop 2015, Hyderabad, India, 2015.
[25]
E. Agullo, A. Buttari, A. Guermouche, and F. Lopez, "Multifrontal QR factorization for multicore architectures over runtime systems," in European Conf. on Parallel Processing. Springer, 2013, pp. 521--532.
- Analyzing dynamic task-based applications on hybrid platforms: an agile scripting approach
Recommendations
Trigeneous Platforms for Energy Efficient Computing of HPC Applications
HIPC '15: Proceedings of the 2015 IEEE 22nd International Conference on High Performance Computing (HiPC)In this paper, we present two novel real-time heterogeneous platforms with three kinds of devices (CPU, GPU, FPGA), i.e. trigeneous platforms, for efficiently accelerating computation intensive applications in both the high-performance computing and the ...
Exploiting task and data parallelism for advanced video coding on hybrid CPU + GPU platforms
Considering the prevalent usage of multimedia applications on commodity computers equipped with both CPU and GPU devices, the possibility of simultaneously exploiting all parallelization capabilities of such hybrid platforms for high performance video ...
Performance and energy effects on task-based parallelized applications
Heterogeneity, parallelization and vectorization are key techniques to improve the performance and energy efficiency of modern computing systems. However, programming and maintaining code for these architectures poses a huge challenge due to the ever-...
Comments
Information & Contributors
Information
Published In
November 2016
34 pages
ISBN:9781509052264
- Conference Chairs:
- Peer-Timo Bremer,
- Judit Gimenez,
- Joshua A. Levine,
- Martin Schulz
Sponsors
- SIGHPC: ACM Special Interest Group on High Performance Computing, Special Interest Group on High Performance Computing
- IEEE-CS\DATC: IEEE Computer Society
In-Cooperation
Publisher
IEEE Press
Publication History
Published: 13 November 2016
Check for updates
Qualifiers
- Research-article
Conference
SC16
Sponsor:
- SIGHPC
- IEEE-CS\DATC
SC16: The International Conference for High Performance Computing, Networking, Storage and Analysis
November 13 - 18, 2016
Utah, Salt Lake City
Acceptance Rates
Overall Acceptance Rate 5 of 6 submissions, 83%
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 04 Oct 2024
Other Metrics
Citations
View Options
Get Access
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in