Changes between Initial Version and Version 1 of RemoteJob


Ignore:
Timestamp:
Mar 2, 2010, 11:32:31 AM (14 years ago)
Author:
davea
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • RemoteJob

    v1 v1  
     1= Remote job submission =
     2
     3A group from Universitat Pompeu Fabra
     4has developed a system for remote job submission and monitoring.
     5This system allows scientists to submit jobs (or groups of jobs)
     6from a convenient command-line interface.
     7
     8The system (Perl-based) is in boinc/rboinc/.
     9
     10'''Warning: this system has been used only by its developers.
     11It will take some work to get it working on other projects.'''
     12
     13Powerpoint slides describing the system are
     14[http://boinc.berkeley.edu/rboinc.pdf here].
     15
     16== Notes ==
     17
     18The software should be fairly self-explanatory, but installation may be tricky. Here's a general overview
     19
     20    * boinc_retrieve_server, boinc_submit_server run as cgi. The former, actually, also handles all administrative requests (stop, purge).
     21    * boinc_retrieve, boinc_submit, are the client components (ditto as above for admin requests)
     22    * Exchange of files between client and server is done through WEBDAV http extensions (a scratch area needs be setup for this)
     23
     24    * Wus naming is important and enforced like this: NNN-UUU_GGG-XX-YY-RNDzzzz where
     25          * NN is the name of the workunit (sub-group)
     26          * UU is the submitter id
     27          * GGG is the group
     28          * XX is the current step in the chain
     29          * YY is the total n. of steps
     30          * zzzz is a random number (not needed,actually)
     31    * WUs are kept in a "workflow_directory",  a subdir of the project dir,  as per slide 22 of the Powerpoint.
     32    * Inside each dir a "process" bash file is created, which is executed by the assimilator with the name of the assimilated WU as its argument. It will create_work the next step for execution.
     33    * The main reason for using perl is that  I preferred to use the XML::Simple module for (un-) xml-ing data structures over the network - it was useful for adding features on the fly keeping backwards compatibility
     34    * I implemented basic functions for authentication, but this is not finished yet
     35    * file storage is optimized through hardlinking and pooling. Network transfers are not (but they could be)