Cutting-edge HPC, Visualization, Data Analysis, and Data-Intensive Computing

After a remarkably successful run, the Stampede supercomputer at TACC at UT Austin completed its full four-year period on January 7, 2017. During its four years of operations, Stampede ran more than eight million successful jobs, delivered over three billion core hours of computation, and ran jobs for more than 11,000 users on over 3,000 projects in the open science community. In January 2017, TACC began the deployment of the Stampede2 system, which entered full production in November 2017.

Stampede was one of the most powerful and significant supercomputers in the U.S. for open science research. Able to perform nearly 10 quadrillion operations per second, Stampede offered opportunities for computational science and technology, ranging from highly parallel algorithms, high-throughput computing, scalable visualization, and next generation programming languages. This Dell PowerEdge cluster equipped with Intel Xeon Phi coprocessors pushed the envelope of computational capabilities, enabling breakthroughs never before imagined. Stampede was funded by the National Science Foundation (NSF) through award ACI-1134872

Thank you to everyone who has helped make Stampede possible: National Science Foundation, The University of Texas at Austin, Dell Inc., Intel Corporation, Mellanox Technologies, Clemson University, Cornell University, The Ohio State University, The University of Texas at El Paso, Indiana University, The University of Colorado at Boulder, and the Institute for Computational Engineering and Sciences (also at The University of Texas at Austin).

 Stampede Retrospective Webpage

 Stampede User Guide

 Stampede Minisite (archived)

Host Name:
Operating System: Linux (CentOS distribution)
Number of Nodes: 6,400
Number of Processing Cores: 522,080
Total Memory: 260TB
Peak Performance: 2+ PF (compute cluster), 7+ PF (coprocessors)
Total Disk: 14PB (shared)
1.6PB (local)

2016 Stampede Upgrade

Stampede was upgraded in 2016 with additional compute nodes built around the second generation of the Intel Xeon Phi many-core, x86 architecture, known as Knights Landing. The new Xeon Phi's function as the primary processors in the new system.

The upgrade ranked #116 on the June 2016 Top 500, and was the only KNL system on the list.

Tech Specifications

  • Over 500 additional compute nodes with second generation Intel Xeon Phi processors
  • About 1.5 petaflops of additional performance
  • New Intel OmniPath network fabric between the new nodes
  • High speed MCDRAM memory on each node