[MODBULKED-22] Establish performance baseline for item locations bulk updates Created: 05/Feb/22  Updated: 09/Aug/22  Resolved: 09/Aug/22

Status: Closed
Project: mod-bulkedit
Components: None
Affects versions: None
Fix versions: None

Type: Story Priority: P3
Reporter: Magda Zacharska Assignee: Oleksandr Bozhko
Resolution: Done Votes: 0
Labels: None
Remaining Estimate: Not Specified
Time Spent: Not Specified
Original estimate: Not Specified

Issue links:
Blocks
is blocked by MODEXPW-166 Bulk Edit: 500 Server Error when edit... Closed
Defines
defines UXPROD-3543 Bulk Edit performance Closed
Gantt End to Start
has to be done after MODBULKED-21 Establish bulk edit performance basel... Closed
Relates
relates to MODEXPW-164 Increase maximal upload file size Closed
Sprint: Firebird - Sprint 145, Firebird - Sprint 146, Firebird - Sprint 144
Story Points: 3
Development Team: Firebird

 Description   

Purpose/Overview:

Establish a baseline for bulk edit app - item records update.

Requirements/Scope:

  1. Create jmeter tests that can be executed against: https://bulk-edit-perf.ci.folio.org/
  2. The tests cover:
    1. uploading file with identifiers: item barcodes, item UUIDs, holdings UUIDs, item HRIDs, item former ids and item accession numbers
    2. updating permanent location 
    3. updating temporary location
    4. saving preview file
    5. saving matching records file
  3. the test will start with 100 item records edits and then progress to 1000 records edits and then 10 000 and 100 000 records edits
  4. Impact of the bulk edit on the storage modules documented: CPU, memory, db timeouts

Approach:

Acceptance criteria:

  • The tests are written and can be executed on demand
  • The findings are documented on the wiki

Additional information:
Coordinate with devops to get memory/cpu usage



 Comments   
Comment by Magda Zacharska [ 03/May/22 ]

Removing blocking Jira because the query search for items is not in scope for Morning Glory. I also removed story points as the scope of the story changed.

Comment by Denis [ 10/Jun/22 ]

suggestion: the steps above should be applied for each identified type

Comment by Magda Zacharska [ 12/Jun/22 ]

Removing estimates as the scope has changed.

Comment by Oleksandr Bozhko [ 21/Jun/22 ]

Magda Zacharska - during the testing of 100_000 identifiers it was found that there is a maximum size of file that can be uploaded (1048576 bytes), and 100_000 exceeds this limit. Based on that, MODEXPW-164 Closed story was created.

Comment by Magda Zacharska [ 21/Jun/22 ]

Oleksandr Bozhko - what is the max number of records that can be uploaded right now? We might mark it as a known issue for Morning glory and increase the number in Nolana.

Comment by Oleksandr Bozhko [ 21/Jun/22 ]

Magda Zacharska - 90_000 records can be uploaded.

Comment by Oleksandr Bozhko [ 21/Jun/22 ]

Magda Zacharska - needs to add clarification: 90_000 records can be uploaded only for accession numbers, barcode, and hrid; for item uuid and holding uuid no more than 15_000 is allowed. Should all records be reduced to 15_000?

Comment by Magda Zacharska [ 21/Jun/22 ]

Oleksandr Bozhko - I think we can assume that for Morning Glory we will support bulk edit of 15k records at once. The number will increase in later releases.

Comment by Oleksandr Bozhko [ 24/Jun/22 ]

Magda Zacharska - the report is incomplete and was created only for 100 records due to the issues described in MODEXPW-166 Closed

Comment by Magda Zacharska [ 24/Jun/22 ]

Hi Oleksandr Bozhko - I'm marking this story as blocked for now. If there is another way of handling it - please let me know. Also, the performance environment will not be available during the next sprint as it will be used for UAT.

Comment by Magda Zacharska [ 20/Jul/22 ]

Removing former item ids due to MODEXPW-194 Closed that will be addressed in later releases.

Comment by Magda Zacharska [ 27/Jul/22 ]

As noted in MODBULKED-23 Closed , Oleksandr Bozhko please include performance baseline for 1 to 5 parallel users.

Comment by Oleksandr Bozhko [ 09/Aug/22 ]

Magda Zacharska - the report is completed and can be found here

Generated at Thu Feb 08 22:30:34 UTC 2024 using Jira 1001.0.0-SNAPSHOT#100246-sha1:7a5c50119eb0633d306e14180817ddef5e80c75d.