skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Accelerated Application Development: The ORNL Titan Experience

Abstract

The use of computational accelerators such as NVIDIA GPUs and Intel Xeon Phi processors is now widespread in the high performance computing community, with many applications delivering impressive performance gains. However, programming these systems for high performance, performance portability and software maintainability has been a challenge. In this paper we discuss experiences porting applications to the Titan system. Titan, which began planning in 2009 and was deployed for general use in 2013, was the first multi-petaflop system based on accelerator hardware. To ready applications for accelerated computing, a preparedness effort was undertaken prior to delivery of Titan. In this paper we report experiences and lessons learned from this process and describe how users are currently making use of computational accelerators on Titan.

Authors:
 [1];  [1];  [1];  [1];  [1];  [2];  [3];  [4];  [1];  [1];  [1];  [1];  [1];  [1]
  1. Oak Ridge National Lab. (ORNL), Oak Ridge, TN (United States)
  2. National Renewable Energy Laboratory (NREL)
  3. NVIDIA, Santa Clara, CA (United States)
  4. Cray, Inc., Knoxville, TN (United States)
Publication Date:
Research Org.:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States). Oak Ridge Leadership Computing Facility (OLCF)
Sponsoring Org.:
USDOE Office of Science (SC)
OSTI Identifier:
1240519
Alternate Identifier(s):
OSTI ID: 1245279
Grant/Contract Number:  
AC05-00OR22725
Resource Type:
Journal Article: Accepted Manuscript
Journal Name:
Computers and Electrical Engineering
Additional Journal Information:
Journal Volume: 46; Journal ID: ISSN 0045-7906
Publisher:
Elsevier
Country of Publication:
United States
Language:
English
Subject:
97 MATHEMATICS AND COMPUTING; High performance computing; computational accelerators; GPU computing; software development; code refactoring; software optimization

Citation Formats

Joubert, Wayne, Archibald, Richard K., Berrill, Mark A., Brown, W. Michael, Eisenbach, Markus, Grout, Ray, Larkin, Jeff, Levesque, John, Messer, Bronson, Norman, Matthew R., Philip, Bobby, Sankaran, Ramanan, Tharrington, Arnold N., and Turner, John A. Accelerated Application Development: The ORNL Titan Experience. United States: N. p., 2015. Web. doi:10.1016/j.compeleceng.2015.04.008.
Joubert, Wayne, Archibald, Richard K., Berrill, Mark A., Brown, W. Michael, Eisenbach, Markus, Grout, Ray, Larkin, Jeff, Levesque, John, Messer, Bronson, Norman, Matthew R., Philip, Bobby, Sankaran, Ramanan, Tharrington, Arnold N., & Turner, John A. Accelerated Application Development: The ORNL Titan Experience. United States. https://doi.org/10.1016/j.compeleceng.2015.04.008
Joubert, Wayne, Archibald, Richard K., Berrill, Mark A., Brown, W. Michael, Eisenbach, Markus, Grout, Ray, Larkin, Jeff, Levesque, John, Messer, Bronson, Norman, Matthew R., Philip, Bobby, Sankaran, Ramanan, Tharrington, Arnold N., and Turner, John A. 2015. "Accelerated Application Development: The ORNL Titan Experience". United States. https://doi.org/10.1016/j.compeleceng.2015.04.008. https://www.osti.gov/servlets/purl/1240519.
@article{osti_1240519,
title = {Accelerated Application Development: The ORNL Titan Experience},
author = {Joubert, Wayne and Archibald, Richard K. and Berrill, Mark A. and Brown, W. Michael and Eisenbach, Markus and Grout, Ray and Larkin, Jeff and Levesque, John and Messer, Bronson and Norman, Matthew R. and Philip, Bobby and Sankaran, Ramanan and Tharrington, Arnold N. and Turner, John A.},
abstractNote = {The use of computational accelerators such as NVIDIA GPUs and Intel Xeon Phi processors is now widespread in the high performance computing community, with many applications delivering impressive performance gains. However, programming these systems for high performance, performance portability and software maintainability has been a challenge. In this paper we discuss experiences porting applications to the Titan system. Titan, which began planning in 2009 and was deployed for general use in 2013, was the first multi-petaflop system based on accelerator hardware. To ready applications for accelerated computing, a preparedness effort was undertaken prior to delivery of Titan. In this paper we report experiences and lessons learned from this process and describe how users are currently making use of computational accelerators on Titan.},
doi = {10.1016/j.compeleceng.2015.04.008},
url = {https://www.osti.gov/biblio/1240519}, journal = {Computers and Electrical Engineering},
issn = {0045-7906},
number = ,
volume = 46,
place = {United States},
year = {Sat May 09 00:00:00 EDT 2015},
month = {Sat May 09 00:00:00 EDT 2015}
}

Journal Article:

Citation Metrics:
Cited by: 19 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Parallelizing Code for Real Applications on the T3D
journal, January 1995


High performance radiation transport simulations: Preparing for TITAN
conference, November 2012

  • Baker, C.; Davidson, G.; Evans, T. M.
  • 2012 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, 2012 International Conference for High Performance Computing, Networking, Storage and Analysis
  • https://doi.org/10.1109/SC.2012.64

An Evaluation of Molecular Dynamics Performance on the Hybrid Cray XK6 Supercomputer
journal, January 2012


Hybridizing S3D into an Exascale application using OpenACC: An approach for moving to multi-petaflops and beyond
conference, November 2012

  • Levesque, John M.; Sankaran, Ramanan; Grout, Ray
  • 2012 SC - International Conference for High Performance Computing, Networking, Storage and Analysis, 2012 International Conference for High Performance Computing, Networking, Storage and Analysis
  • https://doi.org/10.1109/SC.2012.69

Dynamic implicit 3D adaptive mesh refinement for non-equilibrium radiation diffusion
journal, April 2014


Adapting a message-driven parallel application to GPU-accelerated clusters
conference, November 2008


Modeling the propagation of elastic waves using spectral elements on a cluster of 192 GPUs
journal, April 2010


Works referencing / citing this record:

Cost minimization of scheduling scientific workflow applications on clouds
journal, September 2019


LASSIE: simulating large-scale models of biochemical systems on GPUs
journal, May 2017


LASSIE: simulating large-scale models of biochemical systems on GPUs
journal, May 2017