Skip to content Skip to navigation
University of Warwick
  • Study
  • |
  • Research
  • |
  • Business
  • |
  • Alumni
  • |
  • News
  • |
  • About
  • Text only
  • |
  • Sign in
  • Search Computer Science
  • Search University of Warwick
  • Search for people at Warwick
  • Search Warwick Blogs
  • Search past exam papers
  • Search video
  • More…

    Department of Computer Science

    • Research
    • Teaching
    • Admissions
    • People
    • Schools
    • Events
    • News
    • Performance Computing and Visualisation »
    • News
    University of Warwick

    Performance Computing and Visualisation News

    HPC Guest Lecture by Eric Barton (Whamcloud)

    Eric Barton, CTO of Whamcloud will be giving a guest lecture as part of the 4th year HPC (CS402) module. The hour-long lecture will take place Friday 2nd March at 12pm in CS1.01.

    The title of the talk is "HPC Filesystems". During his lecture Eric will address:

    • I/O in HPC systems
    • HPC Filesystems: the current state of the art
    • Future challenges and research directions in this space

    For more information regarding the event, please contact Andrew Mallinson acm@dcs.warwick.ac.uk or Oliver Perks o.perks@warwick.ac.uk 

    A copy of the slides will be made available here.


    Thu 01 March 2012, 10:04 | Tags: Guest Lecture High Performance and Scientific Computing

    image

    Guest Lecture by David Lecomber (Allinea)

    Dr. David Lecomber of Allinea will be giving a guest lecture as part of the 4th year HPC (CS402) module. The hour long lecture will take place on Tuesday the 21st of February at 1pm in CS001.

    The focus of the lecture will be parallel debugging, and will feature a hands on demo of their debugging software DDT, in addition to a presentation.

    For more information please see the website Allinea - DDT.

    For more information regarding the event, please contact Andrew Mallinson acm@dcs.warwick.ac.uk or Oliver Perks o.perks@warwick.ac.uk


    Mon 20 February 2012, 10:47 | Tags: Guest Lecture High Performance and Scientific Computing

    image

    Guest Lecture by Jim Cownie (Intel)

    Jim Cownie of Intel will be giving a guest lecture as part of the 4th year HPC (CS402) module. The hour-long lecture will take place Friday 17th February at 12pm in CS1.01.

    The title of the talk is "Beyond threads: scalable composable parallelism with Cilk and TBB" where he will be discussing alternative programming models for new and future hardware. In particular Jim's talk will address:

    • What’s happening in micro-electronics and what does it mean for architectures and HPC?
    • What do we mean by “scalable” and “composable” parallelism?
    • What’s wrong with threads and OpenMP?
    • What isIntel® Cilk™ Plus?
    • What is TBB?
    • Why do they matter to HPC? (Where are we going withIntelMIC?)

    For more information on these topic areas you might check out Intel Cilk or Intel TBB.

    For more information regarding the event, please contact Andrew Mallinson acm at dcs dot warwick dot ac dot uk or Oliver Perks o dot perks at warwick dot ac dot uk

    A copy of the slides has been made available Here.


    Tue 14 February 2012, 11:00 | Tags: Guest Lecture High Performance and Scientific Computing

    PCAV Research in Many-Core Architectures

    Performance Computing and Visualisation (PCAV) at The University of Warwick welcomes back John Pennycook, who has recently returned from a 6 month placement at Intel Labs, Santa Clara, California. John brings back with him an in-depth knowledge of Intel's vector intrinsics (SSE/AVX) and the forthcoming Many-Integrated Core (MIC) architecture, adding to the group's existing expertise in GPUs and distributed many-core architectures. For more information on PCAV's research in many-core and GPU see one of our recently published research papers:
    • On the Acceleration of Wavefront Applications using Distributed Many-Core Architectures, The Computer Journal (In Press)
    • Performance Analysis of a Hybrid MPI/CUDA Implementation of the NAS-LU Benchmark, ACM SIGMETRICS Performance Evaluation Review, 38 (4)

    Tue 13 December 2011, 14:02 | Tags: High Performance and Scientific Computing

    image

    From the Data Centre to Capability Class Supercomputing - PCAV Researchers Demonstrate Tools for Improving Workloads, Reducing Memory and Profiling Parallel I/O

    Researchers from the Performance Computing and Visualisation group at the University of Warwick have used the UK Performance Engineering Workshop (UKPEW) 2011 to demonstrate four novel techniques for benchmarking, profiling and optimising the performance of parallel workloads.

    The research covers four key areas: business computing transaction processing, high-performance computing workloads, parallel application memory requirements and the profiling of input/output activities such as parallel file writes/reads.

    UKPEW is the leading UK forum for the presentation of all aspects of performance modelling and analysis of computer and telecommunication systems. This year's workshop will run from July 7-8th at the University of Bradford, UK.

    For more details on the publications presented see the following papers:

    • A Modular Failure-Aware Resource Allocation Architecture for Cloud Computing. (Business Transaction Processing and Data Centre Workloads)
    • RIOT - A Parallel Input/Output Tracer. (Parallel Input/Output Profiling)
    • WMTrace - A Lightweight Memory Allocation Tracker and Analysis Framework (Memory Usage/Profiling)
    • The Effect of an Application Performance Modelling Tool. (Supercomputing Machine Workloads)

    Thu 07 July 2011, 15:55 | Tags: High Performance and Scientific Computing Business and Internet Computing

    EPSRC funds new research in radiation hydrodynamics code for laser fusion energy

    The EPSRC, as part of their recent HPC Software Development call, have announced that they are to fund a joint Warwick/Imperial project on the development of a new radiation hydrodynamics ALE code for laser fusion energy. The radiation hydrodynamics (rad-hydro) code will include the laser energy deposition, multiple material interface tracking, arbitrary equations of state (EoS) and energy transport. Stage-one, which will be completed within two years, will deliver the core rad-hydro, laser energy deposition and diffusive transport. More sophisticated models of radiation and energy transport and fusion physics will be added in stage-two. The code developed in stage-one will be of use for initial target design studies and for modeling experiments at the Central Laser Facility (CLF) at the UK Rutherford Appleton Laboratory (RAL).

     


    Tue 15 March 2011, 10:17

    Show selected
    Select all / none
    • Business and Internet Computing
    • Digital Forensics
    • Fault Tolerant and Reliable Systems
    • Guest Lecture
    • High Performance and Scientific Computing
    • Intelligent and Distributed Systems
    • Multimedia Processing and Computer Vision
    • Untagged

    rss
    Add to Start.Warwick
    facebook twitter linkedin
    Intranet

    Department of Computer Science, University of Warwick, Coventry CV4 7AL

    Directions to the University
    Jobs in Computer Science
    Contact details

    Close this email form
    Page contact: Stephen Jarvis Last revised: Thu 1 Mar 2012
    • Sign in
    • |
    • Powered by Sitebuilder
    • |
    • © MMXII
    • |
    • Privacy
    • |
    • Accessibility