Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[ML] Module setup with dynamic model memory estimation #60656

Merged
merged 17 commits into from
Mar 24, 2020
Merged

[ML] Module setup with dynamic model memory estimation #60656

merged 17 commits into from
Mar 24, 2020

Conversation

darnautov
Copy link
Contributor

@darnautov darnautov commented Mar 19, 2020 •
edited
Loading

Summary

Part of #60386

  • /api/ml/modules/setup/{moduleId} kibana endpoint now supports a new parameter estimateModelMemory to perform an actual model memory estimation instead of using hardcoded values from the JSON configuration of the module.
  • cardinalityCheckProvider now supports caching of cardinality field values based on the index, time field and provided time range. It still can be improved taking into account query and bucket span, and also having some expiration time, but for now, it suffices to reduce unnecessary cardinality checks.

For now recognizer module wizard has estimateModelMemory disabled (0e3c9e1) and is going to use hardcoded values from the JSON config until all the required job is done on elasticsearch side.

Checklist

@darnautov darnautov self-assigned this Mar 19, 2020
@elasticmachine
Copy link
Contributor

Pinging @elastic/ml-ui (:ml)

@darnautov darnautov marked this pull request as ready for review March 20, 2020 13:25
@darnautov darnautov requested a review from a team as a code owner March 20, 2020 13:25
@darnautov darnautov mentioned this pull request Mar 20, 2020
Closed
7 tasks
@darnautov
Copy link
Contributor Author

@elasticmachine merge upstream

jgowdyelastic
jgowdyelastic approved these changes Mar 23, 2020
View reviewed changes
Copy link
Member

@jgowdyelastic jgowdyelastic left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Added some small comments, but generally LGTM

x-pack/plugins/ml/common/types/modules.ts Outdated Show resolved Hide resolved
x-pack/plugins/ml/server/models/calculate_model_memory_limit/calculate_model_memory_limit.ts Outdated Show resolved Hide resolved
x-pack/plugins/ml/server/models/data_recognizer/data_recognizer.ts Outdated Show resolved Hide resolved
x-pack/plugins/ml/server/routes/modules.ts Outdated Show resolved Hide resolved
@kibanamachine
Copy link
Contributor

💚 Build Succeeded

History

To update your PR or re-run it, just comment with:
@elasticmachine merge upstream

peteharverson
peteharverson approved these changes Mar 23, 2020
View reviewed changes
Copy link
Contributor

@peteharverson peteharverson left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Tested latest edit and LGTM

@darnautov darnautov merged commit 66b5efd into elastic:master Mar 24, 2020
@darnautov darnautov deleted the ML-60386-modules branch March 24, 2020 07:46
@darnautov darnautov mentioned this pull request Mar 24, 2020
Merged
darnautov added a commit to darnautov/kibana that referenced this pull request Mar 24, 2020
@darnautov
[ML] Module setup with dynamic model memory estimation (elastic#60656)
b45ff40 
* [ML] add estimateModelMemory to the setup endpoint

* [ML] wip caching cardinality checks

* [ML] refactor

* [ML] fix a fallback time range

* [ML] fix typing issue

* [ML] fields_aggs_cache.ts as part of fields_service

* [ML] fix types, add comments

* [ML] check for MML overrides

* [ML] disable estimateModelMemory

* [ML] fix typing

* [ML] check for empty max mml

* [ML] refactor, update types, fix jobsForModelMemoryEstimation

* [ML] fix override lookup

* [ML] resolve nit comments

* [ML] init jobsForModelMemoryEstimation
gmmorris added a commit to gmmorris/kibana that referenced this pull request Mar 24, 2020
@gmmorris
Merge branch 'master' into alerting/tls-warning
1b3d0b4 
* master: (34 commits)
  [APM] add service map config options to legacy plugin (elastic#61002)
  [App Arch] migrate legacy CSS to new platform (core_plugins/kibana_react) (elastic#59882)
  Migrated styles for "share" plugin to new platform (elastic#59981)
  [ML] Module setup with dynamic model memory estimation (elastic#60656)
  Drilldowns (elastic#59632)
  Upgrade mocha dev-dependency from 6.2.2 to 7.1.1 (elastic#60779)
  [SIEM] Overview: Recent cases widget (elastic#60993)
  [ML] Functional tests - stabilize df analytics clone tests (elastic#60497)
  [SIEM] Updates process and TLS tables to use ECS 1.5 fields (elastic#60854)
  Migrate doc view part of discover (elastic#58094)
  Revert "[APM] Collect telemetry about data/API performance (elastic#51612)"
  fix(NA): log rotation watchers usage (elastic#60956)
  [SIEM] [CASES] Build lego blocks case details view (elastic#60864)
  Create Painless Lab app (elastic#57538)
  [SIEM] Move Timeline Template field to first step of rule creation (elastic#60840)
  [Reporting/New Platform Migration] Use a new config service on server-side (elastic#55882)
  [Alerting] allow email action to not require auth (elastic#60839)
  [Maps] Default ES document layer scaling type to clusters and show scaling UI in the create wizard (elastic#60668)
  [APM] Collect telemetry about data/API performance (elastic#51612)
  Implement Kibana Login Selector (elastic#53010)
  ...
darnautov added a commit that referenced this pull request Mar 24, 2020
@darnautov
[ML] Module setup with dynamic model memory estimation (#60656) (#61034)
158d9dc 
* [ML] add estimateModelMemory to the setup endpoint

* [ML] wip caching cardinality checks

* [ML] refactor

* [ML] fix a fallback time range

* [ML] fix typing issue

* [ML] fields_aggs_cache.ts as part of fields_service

* [ML] fix types, add comments

* [ML] check for MML overrides

* [ML] disable estimateModelMemory

* [ML] fix typing

* [ML] check for empty max mml

* [ML] refactor, update types, fix jobsForModelMemoryEstimation

* [ML] fix override lookup

* [ML] resolve nit comments

* [ML] init jobsForModelMemoryEstimation
@peteharverson peteharverson mentioned this pull request Mar 27, 2020
Closed
5 tasks
@darnautov darnautov linked an issue Mar 27, 2020 that may be closed by this pull request
[ML] Improvements to model memory limit handling for data recognizer modules #48510
Closed
5 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@peteharverson peteharverson peteharverson approved these changes

@jgowdyelastic jgowdyelastic jgowdyelastic approved these changes

@alvarezmelissa87 alvarezmelissa87 Awaiting requested review from alvarezmelissa87

Assignees

@darnautov darnautov

Labels
Feature:Anomaly Detection ML anomaly detection :ml release_note:enhancement v7.7.0 v8.0.0
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

[ML] Improvements to model memory limit handling for data recognizer modules
5 participants
@darnautov @elasticmachine @kibanamachine @peteharverson @jgowdyelastic