Task Buffer for PANDA

This document describes the status and progress of the Task Buffer for Panda.

Functions

• Store jobs supplied by the PandaClient interface
• Provide job status information to PandaExecutorInterface and other clients
• Update jobs when requested by PandaDataService, PandaBrokerage, or PandaJobDispatcher

DB Implementation

API

* Store jobs into database

  storeJobs(jobs): 
     parameters
       jobs: list of JobSpecs
     return
       list of tuples (pandaID,jobDefID,jobName).
       pandaID is None when the job has not been stored.

* Update job information

  updateJobs(jobs):
     parameters
       jobs: list of JobSpecs

* Update job status only

  updateJobStatus(jobIDs,status):
     parameters
       jobIDs: list of pandaIDs
       status: list of status

* Get jobs

  getJobs(nJobs,siteName)
     parameters
       nJobs: number of jobs in integer
       siteName: site name in string
     return
       list of JobSpecs

* Get job status only

  getStatus(jobIDs)
     parameters
       jobIDs: list of pandaIDs
     return
       list of status

* Get overall job information

  peekJobs(jobIDs)
     parameters
       jobIDs: list of pandaIDs
     return
       list of JobSpecs

* Query PandaID

  queryPandaIDs(jobDefIDs)
     parameters
       jobDefIDs: list of jobDefIDs
     return
       list of PandaIDs. Each PandaID is for the latest attempt.
       None is returned when corresponding PandaID is not found.

Development Team

Responsible: KaushikDe