Enhancement #1245

Improve database performance of fossology (on fossology.usa.hp.com)

Added by Mary Laser over 1 year ago. Updated 10 months ago.

Status:Closed Start date:02/23/2012
Priority:High Due date:
Assignee:Bob Gobeille % Done:

0%

Category:DataBase
Target version:2.1.0
Rank:2 Tester:

Description

Bob - please feel free to replace this with one or more issues detailing areas for improvement. I will target this for 2.0, pending your list of fixes required to improve performance. Depending on the complexity of the fix(es), one or more may be deferred till post-2.0. Thanks.

History

Updated by Bob Gobeille about 1 year ago

  • Status changed from New to In Progress

sql optimization in svn 5568 may fix this.
Unfortunately, there is currently no test machine with enough data (target is 113 M uploadtree recs) to adequately test this.

Updated by Bob Gobeille about 1 year ago

I'll test this as soon as I can by replicating the fo.usa db.

Updated by Mary Laser about 1 year ago

  • Target version changed from 2.0.0 to 2.0.1

In response to #1, a 1.5TB disk was added to newbobg on 3/6.

Targeting this issue for reevaluation post-2.0

Updated by Paul Holland 11 months ago

  • Rank set to 2

Setting priority rank so it can be addressed sooner rather than later.

Updated by Paul Holland 11 months ago

  • IterNum set to 3

Updated by Bob Gobeille 11 months ago

  • Estimate set to 63

Updated by Bob Gobeille 11 months ago

First round of checkins svn 6001

Updated by Bob Gobeille 10 months ago

  • Status changed from In Progress to Resolved

Scalability is probably going to be an ongoing concern. I'm marking this as resolved as the planned round of changes (to tag retrieval, and partitioning the uploadtree table) are done.

Updated by Bob Gobeille 10 months ago

  • IterNum changed from 3 to 6

Updated by Bob Gobeille 10 months ago

  • IterNum changed from 6 to 5

Updated by Mary Laser 10 months ago

  • Status changed from Resolved to Closed
  • Target version changed from 2.0.1 to 2.1.0

In the most recent OSRB meeting, 4 people were able to use/query fossology.usa simultaneously with good response time. I consider this a sufficient test to close this issue.

Also available in: Atom PDF