|
Size: 16074
Comment:
|
← Revision 86 as of 2017-02-17 08:58:31 ⇥
Size: 6969
Comment:
|
| Deletions are marked like this. | Additions are marked like this. |
| Line 1: | Line 1: |
| #acl Known:read All:read | #acl EditorGroup:read,write,revert Known:read All:read #pragma section-numbers 2 |
| Line 5: | Line 6: |
| {{{#!wiki note If you find any information that is out-dated, incorrect or incomplete, do not hesitate to [[#Support|Open a ticket]]. }}} |
|
| Line 6: | Line 13: |
| '''GRIDUI''' (Grid [[User Interface]]) cluster is the interactive gateway for the computing resources and projects IFCA is involved on. It is a SSH [[http://www.linuxvirtualserver.org/|LVS]], based on [[http://linuxsoft.cern.ch/|Scientific Linux CERN]] versions 5. For legacy applications some nodes on Scientific Linux CERN 4 are still available, however the maintenance of these machines is made in a best-effort basis. The login machines are as follow: | |
| Line 8: | Line 14: |
| {{{#!rhtml <table border="1" cellpadding="2" cellspacing="0" align="center"> <tr align=center bgcolor=#A0A0A0><td>Hostname</td><td>Distribution</td><td>Architecture</td></tr> <tr><td><strong>gridui.ifca.es</strong></td><td rowspan="2">Scientific Linux CERN SLC release 5.5 (Boron)</td><td rowspan="2" align="center">x86_64</td></tr> <tr><td><strong>griduisl5.ifca.es</strong></td></tr> <tr><td><strong>griduisl4.ifca.es</strong></td><td>Scientific Linux CERN SLC release 4.7 (Beryllium)</td><td align="center">i386</td></tr> </table> |
The '''GridUI''' (Grid User Interface) cluster is the interactive gateway to the [[http://grid.ifca.es|Advanced Computing and e-Science]] resources at IFCA. This cluster is comprised of a pool of machines reachable through a single entry point. The connections to the internal machines are managed by a director node that tries to ensure that proper balancing is made across the available nodes at a given moment. Please note that this cluster is ''not intended for the execution of CPU intensive tasks'', for this purpose use any of the available computing resources. Every process spawned is limited to a maximum CPU time of 2 hours. Login on these machines is provided via [[http://en.wikipedia.org/wiki/Secure_Shell|Secure Shell]]. Outgoing SSH connections are not allowed by default from this cluster. Inactive SSH sessions may be closed after 12h. It is highly recommended that you set up [[Cluster/Usage/SSHKeyManagement| SSH Keys]] for authentication, instead of using your username and password. || '''Hostname''' || '''Operating System''' || '''SSH server key fingerprint''' || || `gridui.ifca.es`, `griduisl6.ifca.es` || Scientific Linux 6.X || `29:80:9b:28:e7:8a:00:fe:6c:60:ef:e6:a6:71:33:bd` || == Authentication and user accounts == See [[Cluster/SSO]]. == Access to Scientific Linux 5 machines == After the [[https://grid.ifca.es/sl5-user-interfaces-deprecation-plan2.html|Scientific Linux 5 deprecation]] interactive access to Scientific Linux 5 is still possible trough the batch system. In order to request a SLC5 machine you must append the complex `scientificlinux5` to your request: {{{#!highlight console numbers=disable user@cloudprv-10-0:~ $ qsub -l scientificlinux5=true (...) |
| Line 16: | Line 51: |
| Login on these machines is provided via [[http://en.wikipedia.org/wiki/Secure_Shell|Secure Shell]]. You can use the standard port 22 or the 22000. | |
| Line 18: | Line 52: |
| Please note that this cluster is not intended for the execution of CPU intensive tasks, for this purpose use any of the available computing resources. | If you want an interactive session, append the complex to your `qlogin` request: |
| Line 20: | Line 54: |
| Outgoing SSH connections are not allowed by default from this cluster. Inactive SSH sessions will be closed after 24h. | {{{#!highlight console numbers=disable |
| Line 22: | Line 56: |
| == Authentication == Authentication is centralized via secured LDAP. All the changes made to a user account in one node take immediate effect in the whole cluster. There is also a secured web interface, that allows a user to change his/her details, available at https://cerbero.ifca.es/. If you need to reset your account password, please contact the system administrators. With this username you should be able to access also the ticketing system at http://support.ifca.es/ == Grid resources == As the cluster is based on several Grid User Interfaces it allows users to access either EGEE-III, int.eu.grid, [[http://www.euforia-project.eu/EUFORIA/|EUFORIA]], [[https://web.lip.pt/wiki-IBERGRID/|IBERGRID]] and [[http://grid.csic.es|GRID-CSIC]] infrastructures. To set up the correct environment variables, please `source` any of the environment scripts located under `/gpfs/csic_projects/grid/etc/env/`. For example, to use the I2G infrastructure: {{{ % source /gpfs/csic_projects/grid/etc/env/i2g-env.sh |
user@cloudprv-10-0:~ $ qlogin -l scientificlinux5=true (...) JSV "/nfs4/opt/gridengine/util/resources/jsv/jsv-IFCA.tcl" has been started JSV "/nfs4/opt/gridengine/util/resources/jsv/jsv-IFCA.tcl" has been stopped Your job 1822278 ("QLOGIN") has been submitted waiting for interactive job to be scheduled ... Your interactive job 1822278 has been successfully scheduled. Establishing builtin session to host cloudprv-02-9.ifca.es ... user@cloudprv-02-9:~$ cat /etc/redhat-release Scientific Linux SL release 5.5 (Boron) user@cloudprv-02-9:~$ |
| Line 33: | Line 67: |
| The available enviroments are: {{{#!rhtml <table border="1" cellpadding="2" cellspacing="0" align="center"> <tr align=center bgcolor=#A0A0A0> <td>Filename</td> <td>Allows access to</td> </tr> <tr> <td>euforia-env.{csh,sh}</td> <td>EUFORIA</td> </tr> <tr> <td>i2g-env.{csh,sh}</td> <td>int.eu.grid</td> </tr> <tr> <td>ibergrid-env.{csh,sh}</td> <td>IBERGRID and EGI</td> </tr> <tr> <td>ngi-env.{csh,sh}</td> <td>EGI and IBERGRID NGI</td> </tr> </table> }}} More information on seting up the Grid UI is available locally at the [[DUS: Setting up the User Interface account|corresponding DUS page]]. == PBS Cluster == PBS cluster was <strong>decommissioned</strong> in December 2009. Please use the new [[#SGE_Cluster|SGE Cluster]] instead. |
|
| Line 65: | Line 69: |
| The SGE Cluster is based on `Scientific Linux CERN SLC release 5.5` machines, running on x86_64. The exact number or resources available is shown on the [[http://monitor.ifca.es/ganglia/?c=SGE%20Worker%20Nodes&m=&r=hour&s=descending&hc=4|monitorization page]]. | |
| Line 67: | Line 70: |
| ==== Job submission ==== Local submission is allowed for certain projects. As stated below, there are some shared areas that can be accessed from the computing nodes. The underlying batch system is [http://gridengine.sunsource.net/ Sun Grid Engine]. Please note that the syntax for job submission (<tt>qsub</tt>) and monitoring (<tt>qstat</tt>) is similar to the one you might be accustomed to from PBS, but there are important differences. Refer to the following sources for information: |
The SGE Cluster is based on `Scientific Linux CERN SLC release 6.2` machines, running on x86_64. |
| Line 70: | Line 72: |
| * Grid Engine [[http://gridengine.sunsource.net/documentation.html|documentation page]] * Some useful [[http://gridengine.sunsource.net/project/gridengine/howto/howto.html|HOWTOS]] * SGE [[http://gridengine.sunsource.net/project/gridengine/howto/basic_usage.html|basic usage]]. | Local submission is allowed for certain projects. As stated below, there are some shared areas that can be accessed from the computing nodes. The underlying batch system is [[https://arc.liv.ac.uk/trac/SGE|Son of Grid Engine]] 8.0.0d. Refer to the following sources for information: |
| Line 72: | Line 74: |
| Users should submit their jobs directly to their project by using `qstat -P <project>`. In other words, instead of: | * Using [[/GridEngine]]. * Some useful [[http://arc.liv.ac.uk/SGE/howto/|HOWTOS]]. * SGE [[http://grid.ifca.es/wiki/Cluster/SGE/howto/basic_usage.html|basic usage]]. |
| Line 74: | Line 78: |
| {{{ % qsub -q lhidra ''jobfile'' |
{{{#!wiki important '''IFCA Gridengine documentation has moved''' The specific documentation for IFCA has been moved [[/GridEngine|to a separate section]]. |
| Line 77: | Line 83: |
| you should write (note the dot between the "<tt>l</tt>" and the project name): | |
| Line 79: | Line 84: |
| {{{ % qsub -P l.hidra ''jobfile'' }}} If a project name is not specified, the job will fall in the ''catch-all low-priority'' queue. Submission without specifying a project is allowed, in order to perform <strong>short</strong> test jobs. However, it is in the best interest of the user to use her project in general, so that the job can take full advantage of more capable (longer, more CPUs, higher priority) queues. |
== Shared areas == |
| Line 84: | Line 86: |
| A scratch area is defined for every job as the environment variable `$TMPDIR`. This area is cleared after the job has exited. | The `$HOME` directories are shared between the UIs and the computing nodes. There is a ''projects'' shared area (located at `/gpfs/csic_projects/`), also accessible from the UI and the computing nodes. If your group does not have this area, please open an [[http://support.ifca.es|Incidence ticket]]. |
| Line 86: | Line 88: |
| ==== Wall Clock time ==== A default wall clock time of 72 hours is enforced by default in all jobs submitted to the cluster. Should you require a higher value, please set it by yourself by requesting a new `h_rt` value in the form `hours:minutes:seconds`. Please note that requesting a high value may impact negatively in your job scheduling and execution. Please try to be as accurate as possible when setting this required value. For example, a job requiring 23h should be sent as follows: |
=== Usage === |
| Line 89: | Line 90: |
| {{{ % qsub -P l.hidra -l h_rt=24:00:00 ''jobfile'' }}} ==== Memory management ==== When requesting memory for a job you must take into account that per-job memory is limited in the default queues to a [http://en.wikipedia.org/wiki/Resident_set_size Resident Set Size] (h_rss) of 5 GB. If you need to use more memory, you should request the special resource '''''highmem'''''. Please notice that your group may not be able to request that flag by default. If you need to do so, please [http://wiki.ifca.es/e-ciencia/index.php/GRIDUI_Cluster#Support Open a ticket] requesting it. |
The shared directories '''are not intended for scratch''', use the temporal areas of the local filesystems instead. In other words, instruct every job you send to copy the input from the shared directory to the local scratch (`$TMPDIR`), execute all operations there, then copy the output back to some shared area where you will be able to retrieve it comfortably from the UI. |
| Line 95: | Line 92: |
| However, it is <strong>highly recommended</strong> that you tune your memory requirements to some realistic values. Special emphasis is made in the following resources: | As mentioned above, the contents of `$TMPDIR` are removed after job execution. |
| Line 97: | Line 94: |
| * `h_rss` * `mem_free` |
=== Disk quotas === |
| Line 100: | Line 96: |
| ===== h_rss ===== The first one (`h_rss`) refers to the '''hard resident set size limit'''. The batch system will make sure a given job does not consume more memory than the value assigned to that variable. This means that <strong>any job above the requested `h_rss` limit will be killed (SIGKILL) by the batch system</strong>. It is recommended to request this resource as a top limit for your application. If you expect your job to consume no more than a peak value of 3GB you should request those 3GB as its resident set size limit. This request will not produce a penalty on the scheduling of your jobs. ===== mem_free ===== The second one (`mem_free`) refers to the free RAM necessary for the job to run. The batch system will allow jobs to run only if sufficient memory (as requested by `mem_free`) is available for them. It will also subtract that amount of memory from the available resources, once the job is running. This ensures that a node with 16 GB of memory will not run jobs totalling more than 16 GB. The default value is 1.8 GB per slot. Please note that breaking the `mem_free` limit will not automatically kill your job. Its aim is just to ensure that your job has available the memory you requested. Also note that this value is not intended to be use to reflect the memory peaks of your job. This request will impact the scheduling of your jobs, so it is highly recommended to tune it to fit your application memory usage. This limit is defined in the `/opt/gridengine/default/common/sge_request` file. If your application is always expected to use the same values, you can override that file by creating a `$HOME/.sge_request` file. For further details, please check the `sge_request` manual page. ===== Examples ===== * A job that needs to have 4 GB of memory assigned to it: {{{ % qsub -l mem_free=4G ''jobfile'' }}} * A job that might peak at 4 GB, but in its execution normally needs 3 GB: {{{ % qsub -l h_rss=4G,mem_free=3G ''jobfile'' }}} * A job that might reach 4 GB, and also needs 4 GB: {{{ % qsub -l h_rss=4G,mem_free=4G ''jobfile'' }}} For serial jobs requiring more than 5 GB of memory, submission requesting the '''''highmem''''' flag is necessary. Using this flag, the `h_rss` limit will be unset, but the requirement tuning described above still applies. If your group is allowed to request it, and your job needs 20GB of memory, you can request it as follows: {{{ % qsub -P ''projectname'' -l highmem,mem_free=20G ''jobfile'' }}} * A job that might reach 30 GB and needs 20GB will be submitted as: {{{ % qsub -P ''projectname'' -l highmem,h_rss=30G,mem_free=20G ''jobfile'' }}} * For jobs using [[MPI]], please refer to the [[http://wiki.ifca.es/e-ciencia/index.php/GRIDUI_Cluster#Parallel_jobs|Parallel job submission]] section. ##==== Scratch space ==== ## ##The scratch area for the jobs submitted to the cluster is located under `/tmp/` and is pointed by the `$TMPDIR` variable. ## ##By default, jobs request a 2GB scratch area. Should you need more space, please use the `scract_space` in your resource requiremens: ## ## % qsub -l scratch_space=20G ## ##Please note that this space is dynamic. To check the current disk usage in the nodes you can issue: ## ## % qhost -F scratch_space ##--> ==== Parallel jobs ==== Parallel jobs can be submitted to the parallel environment `mpi`, specifying the number of slots required. SGE will try to spread them over the available resources. {{{ % qsub -pe mpi 8 ''jobfile'' }}} Please note that parallel jobs will be routed to queue ''parallel'' (see [[#Memory_management|previous section]]). Also note that access to that queue is restricted to groups having requested it beforehand. ==== Interactive jobs ==== Interactive, short lived and high priority jobs can be send to a special queue `interactive.q` if the user's project has access to it. {{{ % qsub -q interactive.q ''jobfile'' }}} Execution on this queue is limited to a maximum of 1h of WALL clock time. ==== Resource quotas ==== Some limits may be enforced by the administrators in a user/group/project basis. To check the current resource quotas, the following command must be issued: {{{ % qconf -srqs }}} In order to know the current usage of the quotas defined above, the comand `qquota` must be used: {{{ % qquota -P ''<project> '' }}} ==== Advanced reservation ==== Some users and/or projects might request a reservation of a set of resources in advance. This is called an "Advanced Reservation (AR). If your project needs such a reservation you should make a petition using the [[http://support.ifca.es/|support helpdesk]]. You need to specify the following: * Start datetime, and end datetime (or duration) of the reservation. * Duration of your job(s) (i.e. h_rt for the individual jobs). * Computational resources needed (mem_free, number of slots). Once the request has been made, the system administrators shall give you the ID(s) of the AR created. You can submit your jobs whenever you want by issuing: {{{ % qsub -ar ''<reservation_id>'' ''<other_job_options>'' }}} You can submit your job(s) before the AR starts and also once it is started. However, you should take care of the duration of the reservation and your job' duration. If your job execution exceeds either the `h_rt` that it has requested or the duration of the AR it will be killed by the batch system. You should also take into account that your reservation might not be created in the date and time that you requested if there are no resources available. In this case, it will be created whenever it is possible. To avoid this, please request your reservations well in advance. Since the requested and reserved resources cannot be used for other jobs, those requested resources will be used for accounting purposes as if they were resources used by normal jobs (even in the case that the AR is unused). <strong>Please request only the resources that you need</strong>. If you want to query the existing advance reservations, you can use the `qrstat` command. To query about an specific advance reservation, you can issue: {{{ % qrstat -ar ''<reservation_id>'' }}} == Shared areas == The <tt>home</tt> directories (`/home/$USER`) are shared between the UIs and the computing nodes. There is a ''projects'' shared area (located at `/gpfs/csic_projects/`), also accessible from the UI and the computing nodes. If your group does not have this area, please contact the system administrators. ==== Usage ==== The shared directories <strong>are not intended for scratch</strong>, use the temporal areas of the local filesystems instead. In other words, instruct every job you send to copy the input from the shared directory to the local scratch (<tt>$TMPDIR</tt>), execute all operations there, then copy the output back to some shared area where you will be able to retrieve it comfortably from the UI. As mentioned above, the contents of <tt>$TMPDIR</tt> are removed after job execution. ==== Disk quotas ==== |
|
| Line 207: | Line 98: |
| If you wish to check your quota at a later time, you can use the commands `mmlsquota gpfs_csic` (for user quotas) and `mmlsquota -g `id -g` gpfs_projects` (for group quotas). A script reporting both quotas is located on `/gpfs/csic_projects/utils/bin/rep-user-quotas.py`. A sample output of the latter could be: | If you wish to check your quota at a later time, you can use the commands `mmlsquota gpfs_csic` (for user quotas) and `mmlsquota -g `id -g` gpfs_projects` (for group quotas). A script reporting both quotas is located on `/nfs4/usr/bin/rep-user-quotas.py`. A sample output of the latter could be: |
| Line 222: | Line 113: |
| For a basic interpretation of this output, note that the "Used" column will tell you about how much disk space you are using, whereas "Soft" denotes the limit this "Used" amount should not exceed. The "Hard" column is the value of the limit "Used" plus "Doubt" should not cross. A healthy disk space management would require that you periodically delete unused files in your <tt>$HOME</tt> directory, keeping its usage below the limits at all times. In the event that the user exceeds a limit, a grace period will be shown in the "Grace" column. If the user does not correct the situation within the grace period, she will be banned from writing to the disk. | For a basic interpretation of this output, note that the "Used" column will tell you about how much disk space you are using, whereas "Soft" denotes the limit this "Used" amount should not exceed. The "Hard" column is the value of the limit "Used" plus "Doubt" should not cross. A healthy disk space management would require that you periodically delete unused files in your `$HOME` directory, keeping its usage below the limits at all times. In the event that the user exceeds a limit, a grace period will be shown in the "Grace" column. If the user does not correct the situation within the grace period, she will be banned from writing to the disk. |
| Line 224: | Line 115: |
| For further information you can read the [http://www.nersc.gov/vendor_docs/ibm/gpfs/am3admst119.html mmlsquota command manual page]. | For further information you can read the [[http://www.nersc.gov/vendor_docs/ibm/gpfs/am3admst119.html|mmlsquota command manual page]]. |
| Line 226: | Line 117: |
| == Extra utils == Some extra packages as [[http://python.org|Python 2.6]] and [[http://software.intel.com/en-us/articles/non-commercial-software-development/|Intel Non-Commercial Compilers]] can be found on `/gpfs/csic_projects/utils/`. |
== Extra utils/Software == Some extra packages can be found at `/nfs4/opt/`. This is the location for some pieces of software commonly used like: * Matlab's like `octave` language for numerical analysis. * Data plotting `gnuplot` program. * Profiling and debugging `valgrind` tools. |
| Line 232: | Line 129: |
Before opening a new incidence, please check the [[Cluster/FAQ|Frequently Asked Questions page]] |
|
| Line 235: | Line 135: |
| CategoryDatacenter | CategoryUserSupport |
IFCA Datacenter usage guidelines
Contents
If you find any information that is out-dated, incorrect or incomplete, do not hesitate to Open a ticket.
1. Introduction
The GridUI (Grid User Interface) cluster is the interactive gateway to the Advanced Computing and e-Science resources at IFCA. This cluster is comprised of a pool of machines reachable through a single entry point. The connections to the internal machines are managed by a director node that tries to ensure that proper balancing is made across the available nodes at a given moment.
Please note that this cluster is not intended for the execution of CPU intensive tasks, for this purpose use any of the available computing resources. Every process spawned is limited to a maximum CPU time of 2 hours.
Login on these machines is provided via Secure Shell. Outgoing SSH connections are not allowed by default from this cluster. Inactive SSH sessions may be closed after 12h. It is highly recommended that you set up SSH Keys for authentication, instead of using your username and password.
Hostname
Operating System
SSH server key fingerprint
gridui.ifca.es, griduisl6.ifca.es
Scientific Linux 6.X
29:80:9b:28:e7:8a:00:fe:6c:60:ef:e6:a6:71:33:bd
2. Authentication and user accounts
See Cluster/SSO.
3. Access to Scientific Linux 5 machines
After the [[https://grid.ifca.es/sl5-user-interfaces-deprecation-plan2.html|Scientific Linux 5 deprecation]] interactive access to Scientific Linux 5 is still possible trough the batch system. In order to request a SLC5 machine you must append the complex scientificlinux5 to your request:
user@cloudprv-10-0:~ $ qsub -l scientificlinux5=true (...)
If you want an interactive session, append the complex to your qlogin request:
user@cloudprv-10-0:~ $ qlogin -l scientificlinux5=true (...)
JSV "/nfs4/opt/gridengine/util/resources/jsv/jsv-IFCA.tcl" has been started
JSV "/nfs4/opt/gridengine/util/resources/jsv/jsv-IFCA.tcl" has been stopped
Your job 1822278 ("QLOGIN") has been submitted
waiting for interactive job to be scheduled ...
Your interactive job 1822278 has been successfully scheduled.
Establishing builtin session to host cloudprv-02-9.ifca.es ...
user@cloudprv-02-9:~$ cat /etc/redhat-release
Scientific Linux SL release 5.5 (Boron)
user@cloudprv-02-9:~$
4. SGE Cluster
The SGE Cluster is based on Scientific Linux CERN SLC release 6.2 machines, running on x86_64.
Local submission is allowed for certain projects. As stated below, there are some shared areas that can be accessed from the computing nodes. The underlying batch system is Son of Grid Engine 8.0.0d. Refer to the following sources for information:
Using /GridEngine.
Some useful HOWTOS.
SGE basic usage.
IFCA Gridengine documentation has moved
The specific documentation for IFCA has been moved to a separate section.
5. Shared areas
The $HOME directories are shared between the UIs and the computing nodes. There is a projects shared area (located at /gpfs/csic_projects/), also accessible from the UI and the computing nodes. If your group does not have this area, please open an Incidence ticket.
5.1. Usage
The shared directories are not intended for scratch, use the temporal areas of the local filesystems instead. In other words, instruct every job you send to copy the input from the shared directory to the local scratch ($TMPDIR), execute all operations there, then copy the output back to some shared area where you will be able to retrieve it comfortably from the UI.
As mentioned above, the contents of $TMPDIR are removed after job execution.
5.2. Disk quotas
Disk quotas are enabled on both user and projects filesystems. A message with this information should be shown upon login. If you need more quota on your user space (not in the project shared area), please contact the system administrators explaining your reasons.
If you wish to check your quota at a later time, you can use the commands mmlsquota gpfs_csic (for user quotas) and mmlsquota -g id -g gpfs_projects (for group quotas). A script reporting both quotas is located on /nfs4/usr/bin/rep-user-quotas.py. A sample output of the latter could be:
**********************************************************************
INFORMATION ABOUT YOUR CURRENT DISK USAGE
USER Used Soft Hard Doubt Grace
Space (GB): 3.41 20.00 0.00 0.06 none
Files (x1000): 64 0 0 0 none
GROUP Used Soft Hard Doubt Grace
Space (GB): 0.00 1000.00 1500.00 0.00 none
Files (x1000): 0 0 0 0 none
**********************************************************************For a basic interpretation of this output, note that the "Used" column will tell you about how much disk space you are using, whereas "Soft" denotes the limit this "Used" amount should not exceed. The "Hard" column is the value of the limit "Used" plus "Doubt" should not cross. A healthy disk space management would require that you periodically delete unused files in your $HOME directory, keeping its usage below the limits at all times. In the event that the user exceeds a limit, a grace period will be shown in the "Grace" column. If the user does not correct the situation within the grace period, she will be banned from writing to the disk.
For further information you can read the mmlsquota command manual page.
6. Extra utils/Software
Some extra packages can be found at /nfs4/opt/. This is the location for some pieces of software commonly used like:
Matlab's like octave language for numerical analysis.
Data plotting gnuplot program.
Profiling and debugging valgrind tools.
Please note that these packages are provided as-is, without further support from IFCA staff.
7. Support
Before opening a new incidence, please check the Frequently Asked Questions page
Questions, support and/or feedback should be directed through the use the Helpdesk.