Difference between revisions of "Grid5000:UsagePolicy"

From Grid5000
Jump to: navigation, search
(You are a Grid'5000 citizen ?)
(typos corrections / removal of "will be approved" intro statement because it IS approved now / linked to admins docs policies paragraph to replace old draft page...)
Line 3: Line 3:
  
 
= You are a Grid'5000 citizen =
 
= You are a Grid'5000 citizen =
''This charter, approved by the technical comitee, is scheduled for approval by the steering comitee at its next meeting''.
+
In a general manner, please keep in mind that you are not alone on Grid'5000 and that, due to communities overlaps and common research interests, a majority of users have similar scientific deadlines...
 +
 
 
== Principles ==
 
== Principles ==
 
The Grid'5000 platform is intended to support research in all areas of computer science related to grid computing.  
 
The Grid'5000 platform is intended to support research in all areas of computer science related to grid computing.  
 
You should use Grid'5000 in the perspective of '''large scale experiments''' (at least 3 sites and 1000 CPUs).  
 
You should use Grid'5000 in the perspective of '''large scale experiments''' (at least 3 sites and 1000 CPUs).  
  
You should '''NOT''' use Grid'5000 '''as a production platform'''. Any usage of Grid'5000 should make progress in Computer Science. You may generate useful results for other communities, as long as the community of computer science researchers learn something from your experiments.
+
You should '''NOT''' use Grid'5000 '''as a production platform'''. Any usage of Grid'5000 should lead to progress in Computer Science. You may generate useful results for other communities, as long as the community of computer science researchers learns something from your experiments.
  
 
It is a shared tool, used by many people with different and varying needs. The administrators pursue the following objectives, the main one being the first of this list:
 
It is a shared tool, used by many people with different and varying needs. The administrators pursue the following objectives, the main one being the first of this list:
 
+
# Make the tool available to experiments involving a significant number of nodes (in the 1000's). To make this possible, reservation fragmentation must be avoided as much as possible. Never using more than a single cluster for experiments without any justification will lead to investigations by the steering committee to check whether your experiment fits the objective of Grid'5000
# Make the tool available to experiments involving a significant number of nodes (in the 1000's). To make this possible, reservation fragmentation must be avoided as much as possible. Never using more than a single cluster for experiments without any justification will lead to investigations by the streering committee to check whether your experiment fits the objective of Grid'5000
 
 
# Keep the platform available for the development of experiments during the day. Therefore, reservations using all the nodes available on one site during work hours (in France) should be avoided.
 
# Keep the platform available for the development of experiments during the day. Therefore, reservations using all the nodes available on one site during work hours (in France) should be avoided.
# Allow for experiments to run while administrators are available because platform availability is still greatly dependant on the administrators solving day to day issues.  
+
# Prefer running experiments while administrators are available, because platform availability is still greatly dependant on the administrators solving day to day issues.  
 
# Increase the platform's machines usage as much as possible, as long as this usage doesn't interfere with the first three objectives.
 
# Increase the platform's machines usage as much as possible, as long as this usage doesn't interfere with the first three objectives.
  
Line 24: Line 24:
 
The different sites participating in Grid'5000 do not have clusters of the same size. Therefore, you should take into account job size when you plan an experiment.
 
The different sites participating in Grid'5000 do not have clusters of the same size. Therefore, you should take into account job size when you plan an experiment.
 
* '''Experiment''': An experiment is typically composed of one or more of jobs running on local clusters
 
* '''Experiment''': An experiment is typically composed of one or more of jobs running on local clusters
* '''Job size''': the size of a job is defined by the cpu time usable by the job: (Nb nodes) * (job walltime) * (number of procs). A 2 hour job using 32 bi-processor nodes has a size of 128h.
+
* '''Job size''': the size of a job is defined by the cpu time usable by the job: (nb nodes) * (nb procs) * (job walltime). A 2 hours job using 32 bi-processor nodes has a size of 128h.
  
 
== Good usage rules (All users) ==
 
== Good usage rules (All users) ==
# Please try to plan large scale experiments during nighttime or during week-ends
+
# Please try to plan large scale experiments during night-time or week-ends.
# Tuesday is the only day where you should feel authorized to use all the machines of a cluster during work hours. On the other days, please leave a few nodes for people developing or preparing an experiment. Between 9am and 6pm, you should not use more than the equivalent of 2 hours on all the processors of the cluster during a given day (e.g. on a 64 bi-processor cluster, you should not use more than 2*2*64 = 256h between 9am and 6pm).
+
# Tuesday is the only day where you should feel authorized to use all the machines of a cluster during work hours. On the other days, please leave a few nodes for people developing or preparing an experiment. Between 09:00 and 18:00 (CEST), you should not use more than the equivalent of 2 hours on all the processors of the cluster during a given day (e.g. on a 64 bi-processor cluster, you should not use more than 2*2*64 = 256h between 09:00 and 18:00 CEST).
# If you want to use machines on a Tuesday, please do not reserve them before 13:00pm on the preceding Monday if is not a large scale experiment.  
+
# If you plan to use machines on a Tuesday for anything else than a large scale experiment, please wait after 13:00 (CEST) on the preceding Monday to make your reservation.  
# You should not have more than 2 reservations in advance, because it kills good resource usage. Please submit jobs instead
+
# You should not have more than 2 reservations in advance, because it kills down resource usage. Please optimize and submit jobs instead.
# You should not be submitting jobs or making reservations if your '''experiment''' is not '''described in your [[Special:UserReports | User report]]'''
+
# You should manually release reserved resources (kill interactive session, delete job ID, etc.) when you completed your work before the planned end-time.
# You must '''mention Grid'5000 in all publications''' presenting results or content obtained or derived from the usage of Grid'5000.
+
# You should not be submitting jobs or making reservations if your '''experiment''' is not '''described in your [[Special:UserReports | User report]]'''.
 +
# You must '''mention Grid'5000 in all publications''' presenting results or contents obtained or derived from the usage of Grid'5000.
  
 
== Good usage rules (Local users) ==
 
== Good usage rules (Local users) ==
Some sites have opened accounts on a local branch of LDAP. Users on these branches only have access to the local cluster. In addition to the above rules, they should
+
Some sites have opened user accounts on a local branch of LDAP. Users on these local branches only have access to the local cluster. In addition to the above rules, they should:
# not have more than one reservation in advance,
+
# not have more than 1 reservation in advance;
# prefer submissions (let oar decide when to run the job) to reservations (specifying a time when the job should run) as much as possible,
+
# prefer submissions (let oar decide when to run the job) to reservations (specifying a time when the job should run) as much as possible;
# limit the size of their jobs to reasonable proportions such as those locally defined by the local site they have access to. (e.g. In Rennes, this would be around 1280h).
+
# limit the size of their jobs to reasonable proportions such as those locally defined by the local site they have access to (e.g. in Rennes, this would be around 1280h).
  
 
== Monitoring ==
 
== Monitoring ==
Platform usage is actively monitored by Grid'5000 staff. In case of unconform use, your account will be locked. Unconform use includes users whose user report is unsufficiently filled with regard to their usage of the resources.
+
Platform usage is actively monitored by Grid'5000 staff. In case of unconform use, your account will be locked. Unconform use includes users whose user report is unsufficiently filled with regard to their resource usage.
 +
 
 
== Mailing Lists ==
 
== Mailing Lists ==
As a Grid'5000 user you are automatically subcribed to the Grid'5000 users' mailing list. The traffic is not very high, so please keep an eye on those emails as they may contain important information.
+
As a Grid'5000 user you are automatically subscribed to the Grid'5000 users mailing list. The traffic is not very high, so please keep an eye on those emails as they may contain important information.

Revision as of 14:19, 25 October 2006


You are a Grid'5000 citizen

In a general manner, please keep in mind that you are not alone on Grid'5000 and that, due to communities overlaps and common research interests, a majority of users have similar scientific deadlines...

Principles

The Grid'5000 platform is intended to support research in all areas of computer science related to grid computing. You should use Grid'5000 in the perspective of large scale experiments (at least 3 sites and 1000 CPUs).

You should NOT use Grid'5000 as a production platform. Any usage of Grid'5000 should lead to progress in Computer Science. You may generate useful results for other communities, as long as the community of computer science researchers learns something from your experiments.

It is a shared tool, used by many people with different and varying needs. The administrators pursue the following objectives, the main one being the first of this list:

  1. Make the tool available to experiments involving a significant number of nodes (in the 1000's). To make this possible, reservation fragmentation must be avoided as much as possible. Never using more than a single cluster for experiments without any justification will lead to investigations by the steering committee to check whether your experiment fits the objective of Grid'5000
  2. Keep the platform available for the development of experiments during the day. Therefore, reservations using all the nodes available on one site during work hours (in France) should be avoided.
  3. Prefer running experiments while administrators are available, because platform availability is still greatly dependant on the administrators solving day to day issues.
  4. Increase the platform's machines usage as much as possible, as long as this usage doesn't interfere with the first three objectives.

Glossary

You can use the platform with two different modes: submissions and reservations.

  • Submission: you submit an experiment when you let the scheduler decide when to run it.
  • Reservation: when you make a reservation, you gain usage of the platform at a specified time. You will then need to launch your experiment interactively.

The different sites participating in Grid'5000 do not have clusters of the same size. Therefore, you should take into account job size when you plan an experiment.

  • Experiment: An experiment is typically composed of one or more of jobs running on local clusters
  • Job size: the size of a job is defined by the cpu time usable by the job: (nb nodes) * (nb procs) * (job walltime). A 2 hours job using 32 bi-processor nodes has a size of 128h.

Good usage rules (All users)

  1. Please try to plan large scale experiments during night-time or week-ends.
  2. Tuesday is the only day where you should feel authorized to use all the machines of a cluster during work hours. On the other days, please leave a few nodes for people developing or preparing an experiment. Between 09:00 and 18:00 (CEST), you should not use more than the equivalent of 2 hours on all the processors of the cluster during a given day (e.g. on a 64 bi-processor cluster, you should not use more than 2*2*64 = 256h between 09:00 and 18:00 CEST).
  3. If you plan to use machines on a Tuesday for anything else than a large scale experiment, please wait after 13:00 (CEST) on the preceding Monday to make your reservation.
  4. You should not have more than 2 reservations in advance, because it kills down resource usage. Please optimize and submit jobs instead.
  5. You should manually release reserved resources (kill interactive session, delete job ID, etc.) when you completed your work before the planned end-time.
  6. You should not be submitting jobs or making reservations if your experiment is not described in your User report.
  7. You must mention Grid'5000 in all publications presenting results or contents obtained or derived from the usage of Grid'5000.

Good usage rules (Local users)

Some sites have opened user accounts on a local branch of LDAP. Users on these local branches only have access to the local cluster. In addition to the above rules, they should:

  1. not have more than 1 reservation in advance;
  2. prefer submissions (let oar decide when to run the job) to reservations (specifying a time when the job should run) as much as possible;
  3. limit the size of their jobs to reasonable proportions such as those locally defined by the local site they have access to (e.g. in Rennes, this would be around 1280h).

Monitoring

Platform usage is actively monitored by Grid'5000 staff. In case of unconform use, your account will be locked. Unconform use includes users whose user report is unsufficiently filled with regard to their resource usage.

Mailing Lists

As a Grid'5000 user you are automatically subscribed to the Grid'5000 users mailing list. The traffic is not very high, so please keep an eye on those emails as they may contain important information.