-
Type: Improvement
-
Status: Resolved
-
Priority: Major
-
Resolution: Fixed
-
Affects Version/s: 9.10, 10.1
-
Fix Version/s: 10.3
-
Component/s: Web UI
-
Release Notes Summary:Allows search initialisation to skip aggregation for large repositories
-
Epic Link:
-
Backlog priority:1,000
-
Upgrade notes:
-
Sprint:nxGang Sprint 10.2.9, nxGang Sprint 10.3.1, nxGang Sprint 10.3.2, nxGang Sprint 10.3.3, nxGang Sprint 10.3.8
-
Story Points:1
When opening the search drawer for the first time, user must wait for the first query to complete before being able to actual search for something.
This is costly for several reasons:
- computing aggregates on all the content of a large repository is costly and long on the elasticsearch side (30s on a 500M doc repository without user concurrency)
- making user wait before being able to search is poor UX
This is not the same things as disabling auto search. This issue iss only about opening the drawer for the first time after login.
Proposed solution (Tiago)
Allow attribute on search to disable first call, as this call server the porpose of getting ES buckets.
- depends on
-
ELEMENTS-785 Add skipAggregates boolean property on nuxeo-page-provider element
- Resolved
- is related to
-
ELEMENTS-701 Add option to skip aggregation computation in nuxeo-page-provider
- Resolved
-
NXP-25158 Add a way to do not compute page provider aggregation on demand
- Resolved