1 / 27

The Glidein Service

The Glidein Service. Gideon Juve juve@usc.edu. What are glideins?. A technique for creating temporary, user-controlled Condor pools using resources from remote Grid sites Grid jobs (called “glideins”) are submitted to grid site using normal mechanisms (Globus, etc.)

hastin
Télécharger la présentation

The Glidein Service

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. The Glidein Service Gideon Juve juve@usc.edu

  2. What are glideins? • A technique for creating temporary, user-controlled Condor pools using resources from remote Grid sites • Grid jobs (called “glideins”) are submitted to grid site using normal mechanisms (Globus, etc.) • Glideins start Condor worker daemons on remote resources • Glidein workers join user’s Condor pool and are used to run application jobs

  3. How glideins work LOCAL SITE GRID SITE Globus User LRM Head Node Condor Worker Nodes To use glideins, the user runs a Condor central manager on a local machine that they control. This Condor pool will manage glidein resources allocated from a remote grid site. Condor Central Manager

  4. How glideins work LOCAL SITE GRID SITE Submit Glidein Job Globus User LRM Head Node Condor Worker Nodes The user submits a job to the grid site. This job (called a “glidein”) will start Condor on the remote worker nodes. It is assumed that the Condor worker executables are pre-installed at the site. Condor Central Manager

  5. How glideins work LOCAL SITE GRID SITE Globus User LRM Head Node master Start Glidein startd Condor Worker Nodes The glidein job configures and starts the Condor worker daemons on the grid site’s worker nodes. Condor Central Manager

  6. How glideins work LOCAL SITE GRID SITE Globus User LRM Head Node master Contact Central Manager startd Condor Worker Nodes The Condor daemons contact the user’s central manager and become part of the user’s Condor pool. Condor Central Manager

  7. How glideins work LOCAL SITE GRID SITE Globus User LRM Submit Application Job Head Node master startd Condor Worker Nodes The user submits application jobs to their Condor pool. The jobs are matched with glidein resources. Condor Central Manager

  8. How glideins work LOCAL SITE GRID SITE Globus User LRM Head Node master Run Application Job startd job Condor Worker Nodes The application jobs are dispatched to the worker nodes for execution. Multiple application jobs can be executed on a single glidein resource. Condor Central Manager

  9. What are glideins good for? • Running short jobs on the grid • Condor can dispatch jobs faster than Globus • Bypassing site scheduling policies • Max submitted/running jobs • Priority for large jobs • Avoiding competition for resources • Glideins reserve resources for multiple jobs • Reducing load on head node/LRM • Fewer Globus jobmanagers polling for status

  10. Other Approaches • Advance Reservations • Ask the scheduler for exclusive access to resources • Not supported at many sites • Typically managed by site administrator (not users) • Users are charged a premium for resources • Unused reservations cannot be returned • Task Clustering • Group multiple, independent jobs together • Can delay the release of some jobs • May reduce parallelism • Not mutually exclusive • Can use a combination of techniques

  11. Glidein Service • GT4 grid service for running glideins • Automates the installation and configuration of Condor on grid site • Simplifies the complex setup and configuration required to run glideins • Separate setup and provisioning steps • Create “sites” for remote installation and setup of Condor executables • Create “glideins” for resource allocation

  12. How the Glidein Service works LOCAL SITE GRID SITE Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes The setup is similar to regular glideins. In addition to the Condor central manager, the user runs a Globus container that hosts the Glidein Service. Condor Central Manager

  13. How the Glidein Service works LOCAL SITE GRID SITE Create Site Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes The user creates a new site by sending a request to the Glidein Service. The request specifies details about the grid site including: job submission information, file system paths, etc. Condor Central Manager

  14. How the Glidein Service works LOCAL SITE GRID SITE C Install & Configure Condor Globus User Glidein Service LRM Globus Container Head Node Condor The Glidein Service installs Condor executables on a shared file system at the grid site. The appropriate executables are automatically selected and downloaded from a central repository based on architecture, and operating system. Worker Nodes Condor Central Manager

  15. How the Glidein Service works LOCAL SITE GRID SITE Create Glidein C Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes The user allocates resources by submitting glidein requests to the Glidein Service. The request specifies the number of hosts/CPUs to acquire and the duration of the reservation. Condor Central Manager

  16. How the Glidein Service works LOCAL SITE GRID SITE C Submit Glidein Job Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes The Glidein Service translates the user’s request into a glidein job, which is submitted to the grid site. Condor Central Manager

  17. How the Glidein Service works LOCAL SITE GRID SITE C Globus User Glidein Service LRM Globus Container Head Node master Start Glidein startd Condor Worker Nodes The glidein job starts Condor daemons on the worker nodes of the grid site. Condor Central Manager

  18. How the Glidein Service works LOCAL SITE GRID SITE C Globus User Glidein Service LRM Globus Container Head Node master startd Condor Contact Central Manager Worker Nodes The Condor daemons contact the user’s Condor central manager and become part of the user’s resource pool. Condor Central Manager

  19. How the Glidein Service works LOCAL SITE GRID SITE C Globus User Glidein Service LRM Submit Application Job Globus Container Head Node master startd Condor Worker Nodes The user submits application jobs to their Condor pool. The jobs are matched with available glidein resources. Condor Central Manager

  20. How the Glidein Service works LOCAL SITE GRID SITE C Globus User Glidein Service LRM Globus Container Head Node master startd job Condor Run Application Job Worker Nodes Condor Central Manager The application jobs are dispatched to the remote workers for execution.

  21. How the Glidein Service works LOCAL SITE GRID SITE Cancel Glidein C Globus User Glidein Service LRM Globus Container Head Node master startd Condor Worker Nodes When the user is finished with their application they cancel their glidein request (or let it expire automatically). Condor Central Manager

  22. How the Glidein Service works LOCAL SITE GRID SITE C Cancel Glidein Job Globus User Glidein Service LRM Globus Container Head Node Un-register Condor Worker Nodes When the glidein is cancelled the Glidein Service cancels the glidein job, which causes the Condor daemons on the worker nodes to un-register themselves with the user’s Condor pool and exit. Condor Central Manager

  23. How the Glidein Service works LOCAL SITE GRID SITE Remove Site C Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes The user can submit multiple glidein requests for a single site. When the user is done with the site they ask the Glidein Service to remove the site. Condor Central Manager

  24. How the Glidein Service works LOCAL SITE GRID SITE Uninstall Condor Globus User Glidein Service LRM Globus Container Head Node Condor Worker Nodes Finally, the Glidein Service removes Condor from the shared file system at the grid site. This removes all executables, config files, and logs. Condor Central Manager

  25. Features • Auto-configuration • Detect architecture, OS, glibc -> Condor package • Determine public IP • Generate Condor config file • Multiple interfaces • Command-line • SOAP • Java API • Automatic resubmission of glideins • Indefinitely, N times, until date/time • Notifications • History tracking

  26. Challenges • Firewalls / Private IPs • Hinder communication between glideins and user’s pool • Potential solutions • GCB • VPN? • Run central manager on head node (restrictions) • Shared File System • Currently required for Glidein Service to stage Condor executables

  27. Command Line Example $ glidein create-site --site-name mercury --condor-version 7.0.0 \ --environment GLOBUS_LOCATION=/usr/local/globus-4.0.1-r3 \ --install-path /home/ac/juve/glidein \ --local-path /gpfs_scratch1/juve/glidein \ --staging-service 'gt2 grid-hg.ncsa.teragrid.org/jobmanager-fork’ \ --glidein-service 'gt2 grid-hg.ncsa.teragrid.org/jobmanager-pbs’ $ glidein list-site ID NAME CREATED LAST UPDATE STATE MESSAGE 3 mercury 08-27 09:16 08-27 10:17 READY Installed $ glidein create-glidein --site 3 --count 1 --host-count 1 --num-cpus 1 \ --wall-time 30 --condor-host juve.isi.edu $ glidein list-glidein ID SITE CONDOR HOST SLOTS WTIME CREATED LAST UPDATE STATE MESSAGE 1 mercury (3) juve.isi.edu 1 30 08-27 09:23 08-27 10:24 RUNNING Running $ condor_status Name OpSys Arch State Activity LoadAv Mem ActvtyTime tg-c408.ncsa.terag LINUX IA64 Unclaimed Idle 0.100 1024 0+00:00:05 Total Owner Claimed Unclaimed Matched Preempting Backfill IA64/LINUX 1 0 0 1 0 0 0 Total 1 0 0 1 0 0 0

More Related