Ticket #3713 (closed: wontfix)

Opened 9 years ago

Last modified 5 years ago

Performance degradation in LoadInstrument

Reported by: Russell Taylor Owned by: Anders Markvardsen
Priority: major Milestone: Release 2.3
Component: Mantid Keywords:
Cc: Blocked By:
Blocking: Tester: Nick Draper

Description (last modified by Russell Taylor) (diff)

Changeset [14408] caused the test LoadInstrumentTestPerformance.test_GEM (i.e. running LoadInstrument on GEM) to slow down by around 15% (graph attached).

Changeset [14377] caused a smaller (~5%) slowdown.

We should see if this performance hit can be avoided.

Change History

Changed 9 years ago by Russell Taylor

comment:1 Changed 9 years ago by Russell Taylor

  • Description modified (diff)

comment:2 Changed 9 years ago by Russell Taylor

  • Owner set to Anders Markvardsen
  • Status changed from new to assigned
  • Milestone changed from Iteration 30 to Iteration 31

comment:3 Changed 9 years ago by Anders Markvardsen

Note at the moment ISIS Jenkins 'Performance Tests Report' link does not work

comment:4 Changed 9 years ago by Nick Draper

  • Milestone changed from Iteration 32 to Iteration 33

Moved to iteration 33 at iteration 32 code freeze

comment:5 Changed 8 years ago by Nick Draper

  • Milestone changed from Release 2.1 to Release 2.2

Moved at end of release 2.1

comment:6 Changed 8 years ago by Nick Draper

  • Milestone changed from Release 2.2 to Release 2.3

Moved at the end of release 2.2

comment:7 Changed 8 years ago by Anders Markvardsen

Hi Russell is this ticket still relevant?

comment:8 Changed 8 years ago by Anders Markvardsen

  • Status changed from assigned to accepted

It is judge reasonable to accept the performance increase of about 5% (in absolute value 2.3 seconds to 2.6 seconds). In summary the changes causing this increase was ​http://trac.mantidproject.org/mantidsvn/changeset/14408 which include changes made to allow new IDF notation: http://www.mantidproject.org/IDF#Using_.3Ccombine-components-into-one-shape.3E

The extra increase is perhaps due to checking if this additional notation is present in IDF, and perhaps this could be done smarter to cause no performance increase. But for now accepted this...

comment:9 Changed 8 years ago by Anders Markvardsen

  • Status changed from accepted to verify
  • Resolution set to wontfix

comment:10 Changed 8 years ago by Nick Draper

  • Status changed from verify to verifying
  • Tester set to Nick Draper

Acceptable

comment:11 Changed 8 years ago by Nick Draper

  • Status changed from verifying to closed

comment:12 Changed 5 years ago by Stuart Campbell

This ticket has been transferred to github issue ​4560

Note: See TracTickets for help on using tickets.