wright

Information about wright

Published on October 3, 2007

Author: Laurie

Source: authorstream.com

Content

MPI Scheduling in Condor: An Update Paradyn/Condor Week Madison, WI 2002:  MPI Scheduling in Condor: An Update Paradyn/Condor Week Madison, WI 2002 Outline:  Outline Review of Dedicated/MPI Scheduling in Condor Dedicated vs. Opportunistic Backfill Supported MPI Implementations Supported Platforms Future Work What is MPI?:  What is MPI? MPI is the “Message Passing Interface” A library for writing parallel applications Fixed number of nodes Cannot be preempted Lots of scientists use it for large problems MPI is a standard with many different implementations Dedicated Scheduling in Condor:  Dedicated Scheduling in Condor To schedule MPI jobs, Condor must have access to dedicated resources More and more Condor pools are being formed from dedicated resources Few schedulers handle both dedicated and non-dedicated resources at the same time Problems with Dedicated Compute Clusters:  Problems with Dedicated Compute Clusters Dedicated resources are not really dedicated Most software for controlling clusters relies on dedicated scheduling algorithms Assume constant availability of resources to compute fixed schedules Due to hardware and software failure, dedicated resources are not always available over the long-term Look Familiar?:  Look Familiar? Two common views of a Cluster::  Two common views of a Cluster: The Condor Solution:  The Condor Solution Condor overcomes these difficulties by combining aspects of dedicated and opportunistic scheduling into a single system Opportunistic scheduling involves placing jobs on non-dedicated resources under the assumption that the resources might not be available for the entire duration of the jobs This is what Condor has been doing for years The Condor Solution (cont’d):  The Condor Solution (cont’d) Condor manages all resources and jobs within a single system Administrators only have to maintain one system, saving time and money Users can submit a wide variety of jobs: Serial or parallel (including PVM + MPI) Spend less time learning different scheduling tools, more time doing science Claiming Resources for Dedicated Jobs:  Claiming Resources for Dedicated Jobs When the dedicated scheduler (DS) has idle jobs, it queries the collector to find all dedicated resources DS does match-making to decide which resources it wants DS sends requests to the opportunistic scheduler to claim those resources DS claims resources and has exclusive control (until it releases them) Backfilling: The Problem:  Backfilling: The Problem All dedicated schedulers leave “holes” Traditional solution is to use backfilling Use lower priority parallel jobs Use serial jobs However, if you can’t checkpoint the serial jobs, and/or you don’t have any parallel jobs of the right size and duration, you’ve still got holes Backfilling: The Condor Solution:  Backfilling: The Condor Solution In Condor, we already have an infrastructure for managing non-dedicated nodes with opportunistic scheduling, so we use that to fill the holes in the dedicated schedule Our opportunistic jobs can be checkpointed and migrated when the dedicated scheduler needs the resources again Allows dedicated resources to be used for opportunistic jobs as needed Specific MPI Implementations :  Specific MPI Implementations Supported: MPICH Planned: MPIPro LAM Others? Condor’s MPICH Support:  Condor’s MPICH Support MPICH uses rsh to spawn jobs Condor provides our own rsh tool Older versions of MPICH need to be built without a hard-coded path to rsh Newer versions of MPICH (1.2.2.3 and later) support an environment variable, P4_RSHCOMMAND, which specifies what program should be used Condor and MPIPro:  Condor and MPIPro We’ve investigated supporting MPIPro jobs with Condor MPIPro has some issues with selecting a port for the head node in your computation, and we’re looking for a good solution Condor + LAM = "LAMdor”:  Condor + LAM = "LAMdor” LAM's API is better suited for a dynamic environment, where hosts can come and go from your MPI universe Has a different mechanism for spawning jobs than MPICH Condor working to support their methods for spawning LAMdor (Cont’d):  LAMdor (Cont’d) LAM working to understand, expand, and fully implement the dynamic scheduling calls in their API LAM also considering using Condor’s libraries to support checkpointing of MPI computations Other MPI implementations:  Other MPI implementations What are people using? Do you want to see Condor support any other MPI implementations? If so, let us know by sending email to: [email protected] Supported Platforms:  Supported Platforms Condor’s MPI support is now available on all Condor platforms: Unix Linux, Solaris, Digital Unix, IRIX, HPUX Windows (new since last year) NT, 2000 Future work (short-term):  Future work (short-term) Implementing more advanced dedicated scheduling algorithms Integrating Condor’s user priority system with its dedicated scheduling Adding support for user-specified job priorities (among their own jobs) Condor-MPI support for the Tool Daemon Protocol Future work (longer term):  Future work (longer term) Solving problems w/ MPI on the Grid "Flocking" MPI jobs to remote pools, or even spanning pools with a single computation Solving issues of resource ownership on the Grid (i.e. how do you handle multiple dedicated schedulers on the grid wanting to control a given resource?) More Future work:  More Future work Support for other kinds of dedicated jobs: Generic dedicated jobs We gather and schedule the resources, then call your program, give it the list of machines, and let the program spawn itself Linda (parallel programming interface) Gaussian (computational chemistry) More Future work:  More Future work Better support for preempting opportunistic jobs to facilitate running high-priority dedicated ones “Checkpointing” vanilla jobs to swap space Checkpointing entire MPI computations MW using Condor-MPI How do I start using MPI with Condor?:  How do I start using MPI with Condor? MPI support added and tested in the current development series (6.3.X) MPI support is a built-in feature of the next stable series of Condor (6.4.X) 6.4.0 will be released Any Day Now™ Thanks for Listening!:  Thanks for Listening! Questions? Come to the MPI “BoF”, Wednesday, 3/6/02, 11am-noon, 3385 CS For more information: www.cs.wisc.edu/condor [email protected]

Related presentations


Other presentations created by Laurie

Biofilm Dynamics
29. 10. 2007
0 views

Biofilm Dynamics

Lazarus Suellen
13. 04. 2008
0 views

Lazarus Suellen

SCOR OverviewEC
30. 03. 2008
0 views

SCOR OverviewEC

FAOCS new program
27. 03. 2008
0 views

FAOCS new program

skawinski
18. 03. 2008
0 views

skawinski

Photonics1
14. 03. 2008
0 views

Photonics1

KoreaVisualsIR English
12. 03. 2008
0 views

KoreaVisualsIR English

memberbenefits
11. 03. 2008
0 views

memberbenefits

Basics of Molecular Imaging
04. 03. 2008
0 views

Basics of Molecular Imaging

CI and Dev DEF 2
26. 02. 2008
0 views

CI and Dev DEF 2

chhimala
03. 10. 2007
0 views

chhimala

Ch 7 MM
01. 12. 2007
0 views

Ch 7 MM

stress reduction
05. 12. 2007
0 views

stress reduction

ML inventory
10. 12. 2007
0 views

ML inventory

Aircraft fuel systems
07. 11. 2007
0 views

Aircraft fuel systems

AFG 04 MODELO PRACTICO SEPT 2006
12. 11. 2007
0 views

AFG 04 MODELO PRACTICO SEPT 2006

Neurosurgery
14. 11. 2007
0 views

Neurosurgery

Derek Matthews
15. 11. 2007
0 views

Derek Matthews

lrn11retrieval
15. 11. 2007
0 views

lrn11retrieval

Osteoarthritis Atlanta
13. 12. 2007
0 views

Osteoarthritis Atlanta

Till
25. 12. 2007
0 views

Till

POL103 6th presentation Iran
29. 12. 2007
0 views

POL103 6th presentation Iran

Parasitology CNS Cestoda
07. 01. 2008
0 views

Parasitology CNS Cestoda

Wongf food overview
07. 01. 2008
0 views

Wongf food overview

liste prix coffre
05. 01. 2008
0 views

liste prix coffre

3 Social Movements
27. 12. 2007
0 views

3 Social Movements

DanceW
23. 11. 2007
0 views

DanceW

8 Predator Prey
01. 01. 2008
0 views

8 Predator Prey

teliasonera finland
19. 02. 2008
0 views

teliasonera finland

JeepProfilemktg374
16. 11. 2007
0 views

JeepProfilemktg374

QLogic2005
01. 01. 2008
0 views

QLogic2005

Turkey Starting Conference
23. 11. 2007
0 views

Turkey Starting Conference

Fujii
30. 12. 2007
0 views

Fujii

APres ANS propostas
29. 12. 2007
0 views

APres ANS propostas

Pulerwitz
04. 01. 2008
0 views

Pulerwitz

Matthew 2
01. 10. 2007
0 views

Matthew 2

CLADE07 barla
03. 01. 2008
0 views

CLADE07 barla

Summary
04. 12. 2007
0 views

Summary

Se parlassimo tutti Celtico4
05. 11. 2007
0 views

Se parlassimo tutti Celtico4

geese
16. 11. 2007
0 views

geese

635 Techno010514
20. 11. 2007
0 views

635 Techno010514

CH05 65 CharitableFunds slides
02. 11. 2007
0 views

CH05 65 CharitableFunds slides

ERL07 May17 07
21. 11. 2007
0 views

ERL07 May17 07

gca powerpoint
04. 10. 2007
0 views

gca powerpoint

donacijahupt
28. 11. 2007
0 views

donacijahupt

BIPN100 Lecture4
02. 11. 2007
0 views

BIPN100 Lecture4