/
Check in items workflow Test Report (Fameflower)

Check in items workflow Test Report (Fameflower)

Jul 10, 2020

     

     

     

     

     

     

     

     

     

     

     

    https://folio-org.atlassian.net/browse/PERF-37

    Overview

    In this workflow we are checking the performance of the check in items workflow running in the Fameflower release https://folio-org.atlassian.net/browse/PERF-37.  We tested it with 1, 5, 8, and 20 virtual users for 30 minutes.  

     

    Backend:

    • mod-inventory-storage-19.1.2

    • mod-inventory-14.1.3

    • mod-circulation-18.0.9

    • mod-circulation-storage-11.0.8

    • okapi-2.38.0

    • mod-authtoken-2.4.0

    • mod-permissions-5.9.0

    • mod-data-export-1.1.1

    • mod-source-record-storage-3.1.4

    • mod-source-record-manager-2.1.4

    Frontend:

    • folio_inventory-2.0.2

    • folio_circulation-2.0.0

    • Item Check-in (folio_checkin-2.0.1)

    Environment:

    • 55 back-end modules deployed in 110 ECS services

    • 3 okapi ECS services

    • 8 m5.large  EC2 instances

    • 1  db.r5.xlarge AWS RDS instance

    • WARNING logging level

    High Level Summary

    1. Overall check in time is 781 ms for a typical use case of 8 users, 1127.57 ms for 20 users

    2. Slow APIs taking more than 100ms to run:

      1. POST checkin-by-barcode

      2. GET inventory/items

    3. mod-circulation-storage log warnings for missing indexes. Adding the missing indexes could improve performance while stop logging these warning. See the Recommended Improvements for the JIRAs created to address these indexes. 

    4. JVM profiling shows JSON de/serialization operations one of the slowest operations, totaling more CPU time than other calls.  Since FOLIO modules retrieve and stores JSON objects, making sure that serializing and deserializing JSON efficient is essential, see Recommended Improvements

    Test Runs

    Test

    Virtual Users

    Duration

    OKAPI log level

    Profiled

    Ramp up (total time

    in seconds)

    1. 

    1

    30 min

    WARNING

    No

    1

    2. 

    5

    30 min

    WARNING

    No

    50

    3. 

    8

    30 min

    WARNING

    No

    80

    4. 

    20

    30 min

    WARNING

    No

    200

    5. 

    8

    30 min

    WARNING

    Yes

    80

     

    Results

    1. Overall check in items workflow time in seconds

      For 30 minutes

      Profiler

      Average 

      75th %tile

      1 user

      OFF

      379.88 ms

      379.88 ms

      5 users

      OFF

      715.4 ms

      789.13 ms

      8 users

      OFF

      781 ms

      822.02 ms

      20 users

      OFF

      1127.57 ms

      1203.27 ms

      8 users

      ON

      1352.67 ms

      1415.58 ms

    2. Slow APIs taking more than 100 ms to return (Profiling OFF)

      API

      1 user (75th %tile)

      5 users (75th %tile)

      8 users (75th %tile)

      20 Users (75th %tile)

      API

      1 user (75th %tile)

      5 users (75th %tile)

      8 users (75th %tile)

      20 Users (75th %tile)

      POST checkin-by-barcode 

      379.88 ms

      470.19 ms

      508.61 ms

      750.88 ms

      Get inventory/items

      168.20 ms

      183.79 ms

      208.23 ms

      289.92 ms

    JVM Profiling

    • Overall slow methods (between the modules profiled: mod-circulation, mod-circulation-storage, mod-inventory, mod-inventory-storage)

     

    Database