Bug 4330 – Job Run Queue Processing Algorithm Inefficient

Bug 4330 - Job Run Queue Processing Algorithm Inefficient

Summary:

Job Run Queue Processing Algorithm Inefficient

Status:	RESOLVED FIXED

Product:	GRAM
Component:	wsrf managed execution job service
Version:	4.0.1
Platform:	PC Linux

Importance:	P3 normal
Target Milestone:	4.0.2
Assigned To:	Peter Lane

URL:
Keywords:

Depends on:
Blocks:	4197
	Show dependency tree / graph

Reported:	2006-04-06 09:37 by Peter Lane
Modified:	2006-04-06 09:38 (History)

Attachments
Add an attachment (proposed patch, testcase, etc.)

Note

You need to log in before you can comment on or make changes to this bug.

Description From Peter Lane 2006-04-06 09:37:54

The job run queue processing can get bogged down in relatively long staging
tasks. This causes short tasks to have to wait often for a large number of
these slow tasks to finish before they can be processed. This causes a big slow
down especially when a non-staging job is requested and has to wait for staging
tasks. Some sort of queue prioritization needs to be implemented. Also, staging
states should be skipped entirely if not needed to optomize non-staging job
processing.

------- Comment #1 From Peter Lane 2006-04-06 09:38:49 -------

Fix in HEAD, globus_4_0_branch, and globus_4_0_community.