1 / 21

HYDRA Using Windows Desktop Systems in Distributed Parallel Computing

NSF Site Visit. 2-23-2006. HYDRA Using Windows Desktop Systems in Distributed Parallel Computing . NSF Site Visit. 2-23-2006. Introduction…. Windows desktop systems at IUB student labs 2300 systems, 3 year replacement cycle

Antony
Download Presentation

HYDRA Using Windows Desktop Systems in Distributed Parallel Computing

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. NSF Site Visit 2-23-2006 HYDRAUsing Windows Desktop Systems in Distributed Parallel Computing

  2. NSF Site Visit 2-23-2006 Introduction… • Windows desktop systems at IUB student labs • 2300 systems, 3 year replacement cycle • Pentium IV (>=1.6 GHz), 256/512/1024 MB memory, 10/100 Mbps/GigE, Windows XP • More than 1.5 TF

  3. NSF Site Visit 2-23-2006 Possibly Utilize Idle Cycles? Red: total ownerBlue: total idleGreen: total Condor

  4. NSF Site Visit 2-23-2006 Problem Description • Once again... Windows desktop systems at IUB student labs: • As a scientific resource • Harvest idle cycles

  5. NSF Site Visit 2-23-2006 Constraints • Systems dedicated to students using desktop office applications — not parallel scientific computing – making their availability unpredictable and sporadic • Microsoft Windows environment • Daily software rebuild (updates)

  6. NSF Site Visit 2-23-2006 What could these systems be used for? • Many small computations and a few small messages • Foreman-worker • Parameter studies • Monte Carlo • Goal: High Throughput Computing (not HPC) • Parallel runs of the aforementioned small computations to make better use of resource • Parallel libraries – MPI, PVM, etc. – have constraints if availability of resources is ephemeral i.e. not predictable

  7. NSF Site Visit 2-23-2006 Solution • Simple Message Brokering Library (SMBL) • Limited replacement for MPI • Both server and client library based on TCP socket abstraction • Porting from MPI is fairly straight forward • Process and Port Manager (PPM) • Plus … • Condor for job management, file transfer, no checkpointing or parallelism • Web portal for job submission

  8. NSF Site Visit 2-23-2006 The Big PictureWe’ll discuss each part in more detail next… The shaded box indicates components hosted on multiple desktop computers

  9. NSF Site Visit 2-23-2006 SMBL (Server) SMBL Server Process Table for 4 CPU parallel session • SMBL server maintains a dynamic pool of client process connections • Worker job manager hides details of ephemeral workers at the application level

  10. NSF Site Visit 2-23-2006 SMBL (Server) SMBL Server Process Table for 4 CPU parallel session • SMBL server maintains a dynamic pool of client process connections • Worker job manager hides details of ephemeral workers at the application level

  11. NSF Site Visit 2-23-2006 SMBL (Client) • Client library implements selected MPI-like calls • MPI_Send ()  SMBL_Send () • MPI_Recv ()  SMBL_Recv () • In charge of message delivery for each parallel process

  12. NSF Site Visit 2-23-2006 Process and Port Manager (PPM) • Starts the SMBL server and application processes on demand • Assigns port/host to each parallel session • Directs workers to their servers

  13. NSF Site Visit 2-23-2006 PPM (cont’d ...) PPM with two SMBL servers (two parallel sessions) Parallel Session 1 Parallel Session 2

  14. NSF Site Visit 2-23-2006 Once again … the big picture The shaded box indicates components hosted on multiple desktop computers

  15. NSF Site Visit 2-23-2006 Recent Development • Hydra cluster Teragrid enabled! (Nov 2005) • Allow TG users to use resource • Virtual Host based solution – two different URLs for IU and Teragrid users • Teragrid users authenticate against PSC’s Kerberos server

  16. NSF Site Visit 2-23-2006 System Layout • PPM, SMBL server, Condor and web portal running on Linux server • Dual Intel Xeon 3.0 GHz, 4 GB memory, GigE • Second Linux server running Samba to serve BLAST database

  17. NSF Site Visit 2-23-2006 Portal • Creates and submits Condor files, handles data files • Apache/PHP based • Kerberos authentication • URLs: • http://hydra.indiana.edu (IU users) • http://hydra.iu.teragrid.org (Teragrid users)

  18. NSF Site Visit 2-23-2006 Utilization of Idle Cycles Red: total ownerBlue: total idleGreen: total Condor

  19. NSF Site Visit 2-23-2006 Summary • Large parallel computing facility created at a low cost • SMBL parallel message passing library that can deal with ephemeral resources • PPM port broker that can handle multiple parallel sessions • SMBL Homepage • http://smbl.sourceforge.net (Open Source)

  20. NSF Site Visit 2-23-2006 Links and References • Hydra Portal • http://hydra.indiana.edu (IU users) • http://hydra.iu.teragrid.org (Teragrid users) • SMBL home page: http://smbl.sourceforge.net • Condor home page: http://www.cs.wisc.edu/condor/ • IU Teragrid home page – http://iu.teragrid.org

  21. NSF Site Visit 2-23-2006 Links and References (cont’d..) • Parallel FastDNAml: http://www.indiana.edu/~rac/hpc/fastDNAml • Blast: http://www.ncbi.nlm.nih.gov/BLAST • Meme: http://meme.sdsc.edu/meme/intro.html

More Related