home exascale fpc ctr tfsa soe me
"Computers are incredibly fast, accurate, and stupid; humans are incredibly slow, inaccurate and brilliant; together they are powerful beyond imagination."
Albert Einstein




Stanford Researchers Break Million-core Supercomputer Barrier

Presented by: Joseph Nichols, Stanford University
Thursday February 7th - 8th

Location: Paul Brest Hall Bldg. 4555 Salvatierrra Walk

Researchers at the Center for Turbulence Research set a new record in supercomputing, harnessing a million computing cores to model supersonic jet noise. Work was performed on the newly installed Sequoia IBM Bluegene/Q system at Lawrence Livermore National Laboratories.


HPCToolkit Workshop: Introduction to Application Performance Analysis on Linux Systems

Presented by: Johan Larsson, Stanford University
Thursday July 31st - 10:00 am

Location: Building 260, Room 007 (basement level)

HPCToolkit consists of components for measuring the performance of fully-optimized executables generated by vendor compilers, analyzing application binaries to correlate measurements with program structure, and novel analysis techniques for pinpointing performance bottlenecks in parallel programs. More information can be found here:

TotalView Workshop - Use of TotalView Debugger
Presented by: Ed Hinkle, TotalView and Paul Gray, University of Northern Iowa, Chair of Education, Supercomputing 2008
Date: April 14th 1:30 PM - 3:30 PM

Location: Wallenberg Hall, Peter Wallenberg Learning Theater

This workshop includes an overview of debugging C/C++ and Fortran programs using the TotalView debugger. It will cover setup, navigation, process control, actionpoints, and data examination, as well as memory debugging. This class will also demonstrate how to use the TotalView Debugger to debug MPI and multi-threaded programs.
TotalView includes features such as process groups and asynchronous program control to help to debug difficult parallel program problems.

Stanford University builds 212-node Intel® Cluster Ready (ICR) Cluster in 11 days
Dell, Clustercorp, Panasas, and Intel discuss the different approach used to enable fast cluster installation and use.
Free Webinar gives you the inside look at how it was done: April 8, 2008.

Moderated by Michael Feldman, HPCwire Managing Editor
, and featuring a presentation by Steve Jones, director of Stanford University's High-Performance Computing Center, along with a panel discussion by the solution providers who collaborated to make it happen.



Rocks Clusters BOF

Wednesday, Nov. 14 Time: 12:15pm - 1:15pm

The Rocks Cluster Toolkit is quickly growing into the leading  Linux-based cluster distribution. The second annual Supercomputing  Birds of a Feather meeting for Rocks Developers and users serves as a  forum to discuss new technologies and trends in designing, deploying  and managing large-scale Rocks Clusters. Meet and discuss developer  and end-user issues with members of the core Rocks development team  from the San Diego Supercomputer Center, leading Rocks system  administrators, and key leaders developing Rolls and providing support  in the commercial space. Discussions will include new elements in  recently released Rocks versions, Rocks Rolls, and feedback on the  roadmap for the future of Rocks ? focusing on an interactive session  between users and developers to help bring community needs into focus.

Steve Jones
Stanford University
Greg Bruno
University of California, San Diego
Mason Katz
University of California, San Dieg

Scaling I/O Capability in Parallel File Systems

Tuesday, Nov. 13 Time: 5:30pm - 7:00pm

HPC clusters have been growing in size and capacity. To scale  application performance to meet these advances, data storage and  server I/O capability needs to scale proportionately. Innovations in  data storage technology are enabling applications to achieve  performance levels demanded by HPC users. Additionally, new  multi-fabric switches are enabling the use of high performance  interconnects such as InfiniBand and 10GigE to provide new levels of  bandwidth between compute clusters and storage systems.

In this session we will discuss the following:

  1. Innovative cluster architectures for scaling I/O over high-speed interconnects such as InfiniBand or 10 Gigabit Ethernet to Parallel File Systems
  2. Use of InfiniBand to Ethernet & Storage Gateway Devices and benchmarks in real world environment with applications
  3. How Parallel File System innovations achieve maximum I/O application performance at scale
Nikhil Kelshikar
Cisco Systems, Inc.
Brent Welch
Panasas, Inc.
Steve Jones
Stanford University

Enormous Computational Demands in Flow and Quantum Physics
- Steve Jones and Masoud Aryanpour
Monday September 17 @ Moscone Center

Infiniband Trade Alliance


Stanford High Performance Computing Conference

Tuesday August 21st 8:00am - 5:00pm more..

NASA Administrators Visit Stanford University

Michael Griffin, Administrator, and Lisa Porter, Associate  Administrator of Aeronautics Research, meet with students from  Mechanical Engineering & Aeronautics and Astronautics. The slides are  available here:


Supercomputing '06 - Rocks Clusters BoF
November 15, 2006

Session Chair: Steve Jones, Stanford
Meet to discuss issues with members of the core Rocks development team from the San Diego Supercomputer Center, leading Rocks system administrators, and key leaders in bringing Rocks into the commercial space. More...

Stanford High Performance Computing Conference II

August 29, 2006 at Wallenberg Hall

The event will be held August 29th at Wallenberg Hall
http://wallenberg.stanford.edu More...


CTR Summer Program 2006
July 9 - August 4, 2006


Stanford HPC Center Rocks with Intel Cluster Ready Solution
by Steve Jones
Stanford University

In just eleven days during 2007, the Stanford University High-Performance Computing Center nearly doubled the performance of its existing compute system. Steve Jones, the founder and manager of the Stanford HPC Center, writes about the design and deployment of this system.


A Dell/Stanford Case Study Utilizing Rocks+

Challenge - Build a highly available, standards-based high-performance computing (HPC) cluster that supports researchers conducting groundbreaking work in mathematical simulation for engineering design and flow control analysis.

Solution - The Dell Advanced System Group designed and installed a 212-node, 1,696-processor cluster based on Dell PowerEdge servers, knitted together with Cisco Infiniband, Dell PowerConnect switches and running the Clustercorp Rocks+ operating system.

Engineering by the Numbers - Dell Case Study
Stanford Wins 'Best Practices in Storage' Award
Spring 2007 Best Practices Awards Program
Building a Visualization Cluster with Rocks by Steve Jones, Lee Shunn, and Tim McIntire
Stanford University Advances High-Performance Computing
Stanford Engineering School Speeds Communications For Research Efforts In Large-Scale Simulations And Visualization
Cisco Demos HPC Technology
Building a Visualization Cluster with Rocks by Steve Jones, Lee Shunn, and Tim McIntire
Stanford University Advances High-Performance Computing
Stanford Engineering School Speeds Communications for Research Efforts in Large-Scale Simulations and Visualization
Cisco Demos Its HPC Technology at SC06
Stanford Selects Cisco InfiniBand Switches for HPC Center
Panasas CTO Keynotes HPC Conference at Stanford
Industry Visionary Garth Gibson Keynotes High Performance Computing Conference
Industry Visionary Garth Gibson Keynotes High Performance Computing Conference at Stanford University
Industry Visionary Garth Gibson Keynotes High Performance Computing Conference at Stanford University
Stanford High Performance Computing Conference II - Tuesday August 29th 8:00 am - 5:00 pm
Linux World [Cover Story] Rapid Cluster Deployment: From Delivery to Production in Hours
Panasas Powers Stanford
Panasas Rocks Stanford
Panasas Assists Stanford Researchers
Panasas Cluster Shakes Research at Stanford
Panasas ActiveScale Storage Cluster Powers Turbulence Simulation Research at Stanford University
Panasas ActiveScale Storage Cluster Powers Turbulence Simulation Research at Stanford Univerity
Linux Journal - My Other Computer is a Supercomputer
BioIT World - Take Two for Stanford's Bio-X Project
eWeek - Supercomputing for the Masses?
enVision Magazine Science Quarterly - Hot Commodities






© Stanford University. All Rights Reserved. Stanford, CA 94305. (650) 723-2300. Terms of Use | Copyright Complaints