Nigel Jones

Results 313 comments of Nigel Jones

Also https://github.com/jjohannes/java-module-dependencies

YouTube video: https://www.youtube.com/watch?v=0xijKEeb5ug&list=PLPZy-hmwOdEWUR5OaOGs863rNS2DaIyXS&index=2

Note #3031 removes some spring dependency from the data platform server

Currently we have: ``` ./open-metadata-conformance-suite/open-metadata-conformance-suite-server/pom.xml: - uses spring-web ./open-metadata-implementation/common-services/ffdc-services/pom.xml: - spring-core - spring-rest-client-connector ./open-metadata-implementation/access-services/data-engine/data-engine-server/pom.xml: - spring-core - spring-web - ./open-metadata-implementation/access-services/glossary-view/glossary-view-server/pom.xml: - spring-core ./open-metadata-implementation/access-services/glossary-view/glossary-view-client/pom.xml - spring-core ./open-metadata-implementation/access-services/asset-catalog/asset-catalog-client/pom.xml - spring-core ./open-metadata-implementation/access-services/asset-catalog/asset-catalog-server/pom.xml -...

Assigning to re-assess the situation

Repeated... if we start with all the servers configured, started, and stabilized/idle, then with graph around 1:50 is typical, whilst inmemory takes around 22s. This is running locally (MBP 2016)...

Taking a look with the YourKit profiler (CPU sampling & async sampling of the multiple chassis) it's not clear there's any particular hotspots when specifically running this notebook This would...

Whilst sampling (regular or async) didn't identify any delays, a full trace does seem to indicate a lot of time (in my case 34s - and the only execution was...

Discussed on status call 25 Aug - Agreed to repeat these tests after a large set of changes is integrated in the next week.

This should be retested after performance changes made in 2.7, plus new support in our demo environment for graph/persistence