1 / 7

CIS669 Distributed and Parallel Processing

CIS669 Distributed and Parallel Processing. Spring 2002 Dr. Yuan Shi Lecture 4. Synergy Architecture. Static Daemons: pmd + cid Dynamic daemons: tsh + fah Language Injection Library: libsng.a Runtime components: prun + prun x -debug + pcheck CSL compiler (Configuration Spec. Lang.)

saxon
Télécharger la présentation

CIS669 Distributed and Parallel Processing

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. CIS669 Distributed and Parallel Processing Spring 2002 Dr. Yuan Shi Lecture 4

  2. Synergy Architecture • Static Daemons: pmd + cid • Dynamic daemons: tsh + fah • Language Injection Library: libsng.a • Runtime components: prun + prun x -debug + pcheck • CSL compiler (Configuration Spec. Lang.) • DAC(Distributed Application Controller) • Cds –f (check daemons) • Other Tools • Host management: shosts, chosts, addhost, delhost, dhosts • Failure detection: fdd • Daemon check: cds

  3. Parallel Processing Environment • Cluster(s) of networked computers • ~/.sng_hosts holds all known and/or useable machines. • SNG_PATH points to synergy root on the compiling host. • /synergy/bin in local command search path. • A real ~/bin directory on each host. • Pmd/cid started on each host.

  4. Application Creation Process Step 1 Step 2 Step 3 ap1.c apps.csl %prun/pcheck ap2.c tsh conf cc/gcc cid cid cid fah cid dac libsng.a ~/bin/ap1…

  5. Debugging %prun matrix -debug == Checking Processor Pool: ++ Benchmark (218) ++ (ackbar.yoda.temple.edu) ready. ++ Benchmark (218) ++ (tauntaun.yoda.temple.edu) ready. == Done. Conf Default node (rancor-hps.tc.cornell.edu) == CONFiguring: (matrix.csl) ++ Application Console: (rancor.yoda.temple.edu) ++ Default directory (/usr/users/shi/synergy/apps/ssc/matrix) ++ Debugging mode ON. ++ Automatic program assignment: (worker)->(ackbar.yoda.temple.edu) ++ Automatic slave generation: (worker1)->(tauntaun.yoda.temple.edu) ++ Automatic program assignment: (master)->(ackbar.yoda.temple.edu) ++ Automatic object assignment: (problem)->(ackbar.yoda.temple.edu) pred(1) succ(6) ++ Automatic object assignment: (result)->(ackbar.yoda.temple.edu) pred(6) succ(1) == Done. == Staring distributed objects. sng_prun. file_name(matrix.objs.prcd) ConnectCid: cid_host(246544283) port(26012) ConnectCid: cid_host(246544283) port(26012) ****** (matrix.objs.prcd) verified, all components executable. ConnectCid: cid_host(246544283) port(26012) ****** (matrix.objs.prcd) started. Fault Tolerance activated ::: ret(0) FDD failed to start. == Done. ++ Entering Synergy Debugging Session ... Use .dbx to start debugging each program. Please keep .ini file along with each program. ++ Type matrix.end to terminate the debugging session.

  6. Applications • Computing intense applications • IO intense applications

  7. Future Research • Stateless processing • Distributed stateless processing (transaction) • Fault tolerant cluster research

More Related