A Podcast for HPC Folk

A Podcast for HPC Folk

Podcast Feed

Sponsor

(Opens New Window)

www.mlds-networks.com

  • Websites
  • E-Mail
  • Streaming Audio
  • Podcast
  • Virtual Servers

Podcast

RCE 79: Gluster FS

 

Brock Palen and Jeff Squyres speak with Jeff Darcy of Red Hat about Gluster FS, a distributed scalable open source filesystem.

MP3 (Right Click Save As)

Jeff has been working on distributed storage since NFS version 2 at Encore in 1990. Most recently he worked on Lustre at SiCortex, and then started his own project HekaFS at Red Hat. Since Red Hat acquired Gluster, he has been an architect and ambassador for that project.

RCE 78: Netflix Cache Appliance

Brock Palen and Jeff Squyres speak with Scott Long of Netflix about the new Netflix Cache Appliance. A FreeBSD box with a pile of disk and 2x10gig-e connections. Capable of massive reductions in bandwidth use on ISP networks. We will also cover just what happens when you press play on your Netflix streaming device.

MP3 (Right Click Save As)

Netflix Tech Blog

Scott began his computing career when he discovered Unix in 1992 while an undergrad at the University of Michigan. In 2000 he became a project committer in the FreeBSD project, and from 2002-2006 he was the project's Release Engineer. After 13 years of writing and maintaining device drivers and core operating system infrastructure at Adaptec, Yahoo!, and other technology companies, Scott joined Netflix in May 2012 to work on their OpenConnect streaming server platform. In 2011 he received a Bachelor's degree in Aviation Science, and shortly afterwards did a brief stint as a commercial airline pilot. Now back in the real world of computer engineering, he enjoys working on device drivers and system scalability, advocating FreeBSD, and building mad-scientist high voltage electronics with his family in Colorado.

RCE 77: PerfExpert

Brock Palen and Jeff Squyres speak with James Browne, Leonardo Fialho and Ashay Rane about PerfExpert, An Easy-to-Use Performance Diagnosis Tool for HPC Applications with Suggestions for Bottleneck Remediation.

 

MP3 (Right Click Save As)

James C. Browne

Browne is Professor Emeritus of Computer Science and Physics at The University of Texas at Austin. Browne earned his Ph.D. in Chemical Physics at The University of Texas in 1960 Browne's current research interests span parallel programming and computation, performance optimization, software engineering, formal methods and fault/failure management for complex systems.

Browne has attained Fellow status in five professional societies and areas: The Association for Computing Machinery, the British Computer Society, the American Physical Society, the American Association for the Advancement of Science and the Institute for Constructive Capitalism. Browne has published approximately 100 papers in computational physics and 250 papers in Computer Science.

Browne received the 2004 University of Texas at Austin Career Research Excellence Award for maintaining a superior research program in such fields as parallel computation, performance measurement and analysis, operating systems, and software engineering for a long time span. Browne has also been active in technology transfer and the technology industry. He founded Hyperformix, an Austin Texas based software company which sells products and services for performance management and engineering for enterprise level computer systems. Hyperformix was acquired by Computer Associates in October, 2010.

Leonardo Fialho

Leonardo Fialho was doing research on fault tolerance for MPI applications at the University Autonoma of Barcelona, Spain, where he obtained an MSc and PhD degrees. Hist first posdoctoral position was at Dublin City University (DCU), Ireland, doing research on power efficiency for telecommunication networks. Currently, Leonardo is an associate researcher at the Texas Advanced Computing Center at the University of Texas at Austin.

Ashay Rane

Ashay Rane is a PhD student in the Computer Science department at the University of Texas at Austin. Prior to joining the PhD program in the Fall of 2012, he researched, developed and maintained PerfExpert from December 2010 until August 2012. His experience with performance optimization dates back his Master's degree in Computer Science at Arizona State University, which resulted in a thesis titled, "A Study of the Hybrid MPI/OpenMP Programming Paradigm on Multicore Architectures". His areas of interest include compilers, computer architecture and performance tuning.

RCE 76: Distributed MultiThreaded CheckPointing

Brock Palen and Jeff Squyres speak with Gene Cooperman and Kapil Arya of Distributed MultiThreaded CheckPointing (DMTCP) a tool for kernel-modification-free, transparent user-space checkpoint and restart.

MP3 (Right Click Save As)

Gene Cooperman received his Ph.D. from Brown University in 1978 in Applied Mathematics. After some time in both industry and academia, he became an associate professor at Northeastern University in 1986, and a full professor in 1992. He has worked both in symbolic algebra and cluster computing and has over 80 refereed publications. His early work in cluster computing from the 1990s emphasized TOP-C (Task Oriented Parallel Computing).

RCE 75: Globus

Globus Toolkit

Brock Palen and Jeff Squyres speak with Stuart Martin about Globus, grid software for distributed resource sharing and data transfer.

MP3 (Right Click Save As)

Stuart Martin is a Software Development Manager with the Computation Institute at the University of Chicago and Argonne National Laboratory. Stuart manages software development for the Globus Toolkit and Globus Online. Stuart has worked on the Globus Toolkit for 15 years and Globus Online for 3 years.

RCE 74: Modern Fortran

Brock Palen and Jeff Squyres speak with Steve Lionel of Intel about the modern updates to Fortran.

MP3 (Right Click Save As)

Steve Lionel is a Staff Consultant for the Developer Products Division of Intel Corporation. He has been working in and around Fortran products since 1978, including several years as project lead for DEC's VAX Fortran compiler, plus stints working on Ada, Pascal and C++ compilers. Steve joined Intel in 2001 and is now part of Intel's Compiler Technical Support team. Steve is a member of the ISO Fortran Standards Committee, posts frequently in the Intel Developer Zone forums (http://software.intel.com) and he blogs and tweets as "DoctorFortran" at http://intel.com/software/drfortran and @DoctorFortran.

RCE 73: FFTW - Fastest Fourier Transform in the West

Brock Palen and Jeff Squyres speak with Steven Johnson and Matteo Frigo about FFTW - Fastest Fourier Transform in the West.

MP3 (Right Click Save As)

Matteo Frigo is Consulting Member of Technical Staff at Quanta Research Cambridge, Inc. Previously he was Chief Scientist at Axis Semiconductor, and Chief Scientist and Founder at Cilk Arts, Inc. In the past he has worked on parallel programming systems, compilers for exotic machines, signal processing, medical devices, and cache oblivious algorithms.

Steven G. Johnson is Associate Professor in Applied Mathematics at MIT, where he joined the faculty in 2004. He previously received his PhD in physics from MIT, along with bachelor's degrees in physics, mathematics, and computer science at MIT. His work on FFTW began in 1997 with Matteo Frigo when the two were graduate students, and led to their receiving the 1999 Wilkinson Prize in Numerical Software. Most of his research centers around the theoretical understanding, computational modeling, and engineering design of nanoscale physics, especially electromagnetism in media structured at the wavelength scale or below. This has included contributions to the theory of nanostructured dielectric waveguides, the design of solar cells and optical fibers, the understanding of van der Waals forces and thermal radiation, and the design of nonlinear optical devices. He is co-author of the 2008 textbook, Photonic Crystals: Molding the Flow of Light, has published over 150 articles in scientific journals, along with several popular free-software packages for scientific computation and optimization.

RCE 72 HUBzero nanoHUB

Brock Palen and Jeff Squyres speak with Gerhard Klimeck about HUBzero and an implementation of HUBzero MP3 (Right Click Save As)

Gerhard Klimeck is the Director of the Network for Computational Nanotechnology and Professor of Electrical and Computer Engineering at Purdue University. He was previously with NASA/JPL and Texas Instruments leading the Nanoelectronic Modeling Tool development (NEMO). He leads the development of nanoHUB.org and research in his Nanoelectronic Modeling group. His work is documented in over 340 peer-reviewed journal and proceedings articles and over 160 invited and 330 contributed conference presentations. He is a Fellow of the American Physical Society and the Institute of Physics and a senior member of the IEEE.

UPDATE: Dr. Klimeck has a video on Purdue TEDx.

RCE 71 HPC Allocation Models

Brock Palen and Jeff Squyres speak with Preston Smith from Purdue, Brian Guilfoos from The Ohio Supercomputer Center, and Andy Caird from The University of Michigan about their CPU cycle allocation policies on HPC systems.

MP3 (Right Click Save As)

Guilfoos leads the Client and Technology Support Group, which provides training and user support to the OSC user community. Guilfoos also works directly with OSC clients to help convert computer codes, develop batch scripting, compiling and code development so that these researchers can efficiently use the center's supercomputers and licensed software. Guilfoos developed and delivered training in MATLAB as a part of the U.S. Department of Defense High Performance Computing Modernization Program support. Prior to joining OSC, he was contracted by the Air Force Research Laboratory to focus on software development in support of unmanned aerial vehicle (UAV) interface research. He was a key technical member of a team that was awarded the 2004 Scientific and Technological Achievement Award by the AFRL Human Effectiveness Directorate.

Guilfoos earned a bachelor's degree in electrical engineering in 2000 from The Ohio State University.

Preston Smith is the Manager of Research Support for Purdue University's Rosen Center for Advanced Computing. Preston's background is from systems engineering, and over the years, he has worked on grid and distributed computing, led development and expansion of Purdue's large scale Condor pool, and was formerly the site lead for Purdue's CMS TIer-2 facility. Today, his team primarily supports users of our community cluster program.

Andy Caird is the Director of HPC at the University of Michigan's College of Engineering Computer Aided Engineering Network (CAEN). He started in HPC in 1992 as the student employee responsible for rebooting U-M's KSR-1 and is now responsible for the staff and operation of Michigan's largest public HPC platform, Flux, in joint effort with Office of Research Cyberinfastructure (ORCI). Andy received his BSE and MSE in Nuclear Engineering from the University of Michigan.

 
Joomla 1.5 Templates by Joomlashack