Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
Last revisionBoth sides next revision
start [2016/02/24 21:48] neyronstart [2016/02/25 13:42] – [Main features] neyron
Line 14: Line 14:
 {{ ::oar-logo_100px.png?nolink|}} {{ ::oar-logo_100px.png?nolink|}}
   * Batch and Interactive jobs   * Batch and Interactive jobs
 +  * Advance Reservation
   * Admission rules   * Admission rules
   * Walltime   * Walltime
Line 23: Line 24:
   * Check compute nodes before launching   * Check compute nodes before launching
   * Epilogue/Prologue scripts   * Epilogue/Prologue scripts
-  * Activity visualization tools (Monika)+  * Jobs and resources visualization tools (Monika, Drawgantt)
   * No Daemon on compute nodes   * No Daemon on compute nodes
-  * rsh and ssh as remote execution protocols (managed by Taktuk)+  * SSH based remote execution protocols (managed by [[http://taktuk.gforge.inria.fr|TakTuk]])
   * Dynamic insertion/deletion of compute node   * Dynamic insertion/deletion of compute node
   * Logging   * Logging
   * Backfilling   * Backfilling
   * First-Fit Scheduler with matching resource   * First-Fit Scheduler with matching resource
-  * Advance Reservation +  * On demand OS deployment support with [[http://kadeploy3.gforge.inria.fr|Kadeploy3]] coupling 
-  * Environnement of Demand support (Ka-tools integration) +  * Grid computing support with [[http://cigri.imag.fr|Cigri]]
-  * Grid integration with [[http://cigri.imag.fr|Cigri system]] +
-  * Simple Desktop Computing Mode+
  
 ===== Why using OAR ===== ===== Why using OAR =====
-==== OAR advantages ==== 
- 
-We present below some points that explain benefits of the new version of OAR. 
  
 === A better resource management === === A better resource management ===
Line 93: Line 89:
   * scalable exploiting tool Taktuk.   * scalable exploiting tool Taktuk.
  
-It is flexible enough to be suitable to manage production HPC clusters as well as other computer infrastructures such as research testbeds like [[http://www.grid5000.fr|Grid'5000]]. +It is flexible enough to be suitable to manage large HPC clusters as well as other computer infrastructures such as research testbeds like [[http://www.grid5000.fr|Grid'5000]]. 
  
-OAR advantages:+Features:
  
-  * No specific daemon on nodes. +  * Only need a SSH daemon on nodes. 
-  * No dependence on specific computing libraries like MPI. We support all sort of parallel user applications. +  * No tied to any specific libraries such as MPI. OAR supports any sort of parallel user applications. 
-  * Upgrades are made on the servers, nothing to do on computing nodes. +  * Cpuset/cgroup integration to restricts jobs to assigned resources (corememory bench
-  * CPUSET (2.6 linux kernel) integration which restricts the jobs on assigned resources (also useful to clean completely a jobeven parallel jobs). +  * Remote procedure calls using the [[http://taktuk.gforge.inria.fr/|TakTuk]] software: a large scale remote execution deployment tool
-  * All administration tasks are performed with the taktuk command (a large scale remote execution deployment): http://taktuk.gforge.inria.fr/+  * Hierarchical resource requests (multiple/heterogeneous clusters support). 
-  * Hierarchical resource requests (handle heterogeneous clusters). +  * Gantt scheduling (can visualize the internal scheduler decisions).
-  * Gantt scheduling (so you can visualize the internal scheduler decisions).+
   * Full or partial time-sharing.   * Full or partial time-sharing.
   * Checkpoint/resubmit.   * Checkpoint/resubmit.
-  * Licences servers management support+  * Support for software licenses servers (license tokens)
-  * Best effort jobs : if another job wants the same resources then it is deleted automatically (useful to execute programs like SETI@home)+  * Best effort jobs: such a job is stopped automatically as soon as another job requires the resources. 
-  * Environment deployment support (Kadeploy): http://kadeploy3.gforge.inria.fr/. +  * Special job types: 
- +    * deploy: support for in job OS deployment with software such as [[http://kadeploy3.gforge.inria.fr|Kadeploy]] 
-Other more common features+    * cosystem: support for a delegation to another job and resource management system 
- +    * noopreservation only jobs (no execution) 
-  * Batch and Interactive jobs. +    * placeholder, container: meta jobs   
-  * Admission rules. +  * Batch and Interactive jobs 
-  * Walltime.+  * Advance reservations 
 +  * Admission rules 
 +  * Job walltime
   * Multi-schedulers support.   * Multi-schedulers support.
-  * Multi-queues with priority+  * Multi-queues with priority 
-  * Backfilling. +  * First-Fit Scheduler with conservative backfilling 
-  * First-Fit Scheduler. +  * Moldable tasks.
-  * Reservation. +
-  * Support of moldable tasks+
-  * Check compute nodes.+
   * Epilogue/Prologue scripts.   * Epilogue/Prologue scripts.
-  * Support of dynamic nodes.+  * Dynamic resources definition
   * Logging/Accounting.   * Logging/Accounting.
   * Suspend/resume jobs.   * Suspend/resume jobs.
  
start.txt · Last modified: 2018/05/27 11:59 by neyron
Recent changes RSS feed GNU Free Documentation License 1.3 Donate Powered by PHP Valid XHTML 1.0 Valid CSS Driven by DokuWiki