[ Skip to the content ]

Institute of Formal and Applied Linguistics Wiki


[ Back to the navigation ]

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
grid [2018/11/01 14:34]
vodrazka [Installation]
grid [2019/02/14 17:49] (current)
dusek [Advanced usage]
Line 160: Line 160:
   * **act_mem_free** (or amf) is a ÚFAL-specific option, which specifies the real amount of free memory (at the time of scheduling). You can specify it when submitting a job and it will be scheduled to a machine with at least this amount of memory free. In an ideal world, where no jobs are exceeding their ''​mem_free''​ requirements,​ we would not need this options. However, in real world it is recommended to use this option with the same value as ''​mem_free''​ to protect your job from failing with out-of-memory error (because of naughty jobs of other users).   * **act_mem_free** (or amf) is a ÚFAL-specific option, which specifies the real amount of free memory (at the time of scheduling). You can specify it when submitting a job and it will be scheduled to a machine with at least this amount of memory free. In an ideal world, where no jobs are exceeding their ''​mem_free''​ requirements,​ we would not need this options. However, in real world it is recommended to use this option with the same value as ''​mem_free''​ to protect your job from failing with out-of-memory error (because of naughty jobs of other users).
   * **h_vmem** is equivalent to setting ''​ulimit -v'',​ i.e. it is a hard limit on the size of virtual memory (see RLIMIT_AS in ''​man setrlimit''​). If your job exceeds this limit, memory allocation fails (i.e., malloc or mmap will return NULL), and your job will probably crash on SIGSEGV. TODO: according to ''​man queue_conf'',​ the job is killed with SIGKILL, not with SIGSEGV. Note that ''​h_vmem''​ specifies the maximal size of **allocated_memory,​ not used_memory**,​ in other words it is the VIRT column in ''​top'',​ not the RES column. SGE does not use this parameter in any other way. Notably, job scheduling is not affected by this parameter and therefore there is no guarantee that there will be this amount of memory on the chosen machine. The problem is that some programs (e.g. Java with the default setting) allocate much more (virtual) memory than they actually use in the end. If we want to be ultra conservative,​ we should set ''​h_vmem''​ to the same value as ''​mem_free''​. If we want to be only moderately conservative,​ we should specify something like h_vmem=1.5*mem_free,​ because some jobs will not use the whole mem_free requested, but still our job will be killed if it allocated much more than declared. The default effectively means that your job has no limits.   * **h_vmem** is equivalent to setting ''​ulimit -v'',​ i.e. it is a hard limit on the size of virtual memory (see RLIMIT_AS in ''​man setrlimit''​). If your job exceeds this limit, memory allocation fails (i.e., malloc or mmap will return NULL), and your job will probably crash on SIGSEGV. TODO: according to ''​man queue_conf'',​ the job is killed with SIGKILL, not with SIGSEGV. Note that ''​h_vmem''​ specifies the maximal size of **allocated_memory,​ not used_memory**,​ in other words it is the VIRT column in ''​top'',​ not the RES column. SGE does not use this parameter in any other way. Notably, job scheduling is not affected by this parameter and therefore there is no guarantee that there will be this amount of memory on the chosen machine. The problem is that some programs (e.g. Java with the default setting) allocate much more (virtual) memory than they actually use in the end. If we want to be ultra conservative,​ we should set ''​h_vmem''​ to the same value as ''​mem_free''​. If we want to be only moderately conservative,​ we should specify something like h_vmem=1.5*mem_free,​ because some jobs will not use the whole mem_free requested, but still our job will be killed if it allocated much more than declared. The default effectively means that your job has no limits.
 +    * For GPU jobs, it is usually better to use **h_data** instead of **h_vmem**. CUDA driver allocates a lot of "​unused"​ virtual memory (tens of GB per card), which is counted in ''​h_vmem'',​ but not in ''​h_data''​. All usual allocations (''​malloc'',​ ''​new'',​ Python allocations) seem to be included in ''​h_data''​.
   * It is recommended to **profile your task first** (see [[#​profiling]] below), so you can estimate reasonable memory requirements before submitting many jobs with the same task (varying in parameters which do not affect memory consumption). So for the first time, declare mem_free with much more memory than expected and ssh to a given machine and check ''​htop''​ (sum all processes of your job) or (if the job is done quickly) check the epilog. When running other jobs of this type, set ''​mem_free''​ (and ''​act_mem_free''​ and ''​h_vmem''​) so you are not wasting resources, but still have some reserve.   * It is recommended to **profile your task first** (see [[#​profiling]] below), so you can estimate reasonable memory requirements before submitting many jobs with the same task (varying in parameters which do not affect memory consumption). So for the first time, declare mem_free with much more memory than expected and ssh to a given machine and check ''​htop''​ (sum all processes of your job) or (if the job is done quickly) check the epilog. When running other jobs of this type, set ''​mem_free''​ (and ''​act_mem_free''​ and ''​h_vmem''​) so you are not wasting resources, but still have some reserve.
   * **s_vmem** is similar to ''​h_vmem'',​ but instead of SIGSEGV/​SIGKILL,​ the job is sent a SIGXCPU signal which can be caught by the job and exit gracefully before it is killed. So if you need it, set ''​s_vmem''​ to a lower value than ''​h_vmem''​ and implement SIGXCPU handling and cleanup.   * **s_vmem** is similar to ''​h_vmem'',​ but instead of SIGSEGV/​SIGKILL,​ the job is sent a SIGXCPU signal which can be caught by the job and exit gracefully before it is killed. So if you need it, set ''​s_vmem''​ to a lower value than ''​h_vmem''​ and implement SIGXCPU handling and cleanup.
Line 166: Line 167:
 ===== Advanced usage ===== ===== Advanced usage =====
  
-''​qsub **-q** ​troja-all.q''​ +''​qsub **-q** ​cpu-troja.q''​ 
-This way your job is submitted to the Troja queue. The default is ''​ms-all.q''​. You can also use e.g.+This way your job is submitted to the Troja queue. The default is ''​cpu-ms.q''​. You can also use e.g.
 ''​-q '​(troja*|ms*)'''​ to submit on any machine in those two queues (but **don'​t use ''​-q '​*'''​** as this includes also [[:​gpu|gpu.q]]),​ ''​-q '​(troja*|ms*)'''​ to submit on any machine in those two queues (but **don'​t use ''​-q '​*'''​** as this includes also [[:​gpu|gpu.q]]),​
 ''​-q '​*@hector[14]'''​ to submit on hector1 or hector4, ''​-q '​*@hector[14]'''​ to submit on hector1 or hector4,
 ''​-q '​[tm]*@!(hector*|iridium)'''​ to submit on any troja/ms machine except hectors and iridium. ''​-q '​[tm]*@!(hector*|iridium)'''​ to submit on any troja/ms machine except hectors and iridium.
-However, usually you should specify just the queue (troja-all.q vs. ms-all.q), not a particular machine, and instead use ''​-l''​ to specify the needed resources in a general way.+However, usually you should specify just the queue (cpu-troja.q vs. cpu-ms.q), not a particular machine, and instead use ''​-l''​ to specify the needed resources in a general way.
  
 ''​qsub **-l** ...''​ ''​qsub **-l** ...''​
 See ''​man complex''​ (run it on lrc or sol machines) for a list of possible resources you may require (in addition to ''​mem_free''​ etc. discussed above). See ''​man complex''​ (run it on lrc or sol machines) for a list of possible resources you may require (in addition to ''​mem_free''​ etc. discussed above).
  
-''​qsub **-p** -99''​ +''​qsub **-p** -200''​ 
-Define a priority of your job as a number between -1024 and 0. Only SGE admins may use a number higher than 0. In January 2018, we changed the default to -100 (it used to be 0). SGE uses the priority to decide when to start which pending job in the queue (it computes a real number called ''​prior'',​ which is reported in ''​qstat'',​ which grows as the job is waiting in the queue). Note that once a job is started, you cannot "​unschedule"​ it, so from that moment on, it is irrelevant what was its priority. You can ask for a higher priority (-99...0) if your job is urgent and/or will finish soon and you want to skip your colleagues'​ jobs in the queue. You should ask for lower priority (-1024..-101) if you submit many jobs at once or if the jobs are not urgent.+Define a priority of your job as a number between -1024 and 0. Only SGE admins may use a number higher than 0. In January 2018, we changed the default to -100 (it used to be 0). Please, do not use priority between -99 and 0 for jobs taking longer than a few hours, unless it is absolutely necessary for a deadline. In that case, please notify other GPU users. You should ask for lower priority (-1024..-101) if you submit many jobs at once or if the jobs are not urgent. SGE uses the priority to decide when to start which pending job in the queue (it computes a real number called ''​prior'',​ which is reported in ''​qstat'',​ which grows as the job is waiting in the queue). Note that once a job is started, you cannot "​unschedule"​ it, so from that moment on, it is irrelevant what was its priority.
  
 ''​qsub **-o** LOG.stdout **-e** LOG.stderr''​ ''​qsub **-o** LOG.stdout **-e** LOG.stderr''​
Line 193: Line 194:
 ''​qsub **-M** popel@ufal.mff.cuni.cz,​rosa@ufal.mff.cuni.cz **-m** beas''​ ''​qsub **-M** popel@ufal.mff.cuni.cz,​rosa@ufal.mff.cuni.cz **-m** beas''​
 Specify the emails where you want to be notified when the job has been **b** started, **e** ended, **a** aborted or rescheduled,​ **s** suspended. Specify the emails where you want to be notified when the job has been **b** started, **e** ended, **a** aborted or rescheduled,​ **s** suspended.
-The default is now ''​-m ​ea'' ​(TODO check this) and the default email address is forwarded to you (so there is no need to use ''​-M''​). You can use ''​-m n''​ to override the defaults and send no emails.+The default is now ''​-m ​a''​ and the default email address is forwarded to you (so there is no need to use ''​-M''​). You can use ''​-m n''​ to override the defaults and send no emails.
  
 ''​qsub **-hold_jid** 121144,​121145''​ (or ''​qsub **-hold_jid** get_src.sh,​get_tgt.sh''​) ''​qsub **-hold_jid** 121144,​121145''​ (or ''​qsub **-hold_jid** get_src.sh,​get_tgt.sh''​)
Line 220: Line 221:
  
 ''​**qalter**''​ ''​**qalter**''​
-You can change some properties of already submitted jobs, which are still waiting in the queue (//​pending//​).+You can change some properties of already submitted jobs (both waiting in the queue and running). Changeable properties are listed in ''​man qsub''​.
  
 ''​**man** qsub qstat qalter qhold queue_conf sge_types complex''​ ''​**man** qsub qstat qalter qhold queue_conf sge_types complex''​

[ Back to the navigation ] [ Back to the content ]