Skip to content

Need to rethink the query plans #106

@jaredpar

Description

@jaredpar

Pretty much all of my query plans, save for ModelBuild, require join operations on large tables. These joins will occur in such a way that a full index scan occurs and that is expensive. Even doing queries like looking for count of ModelTestRun that occurs before a given date causes a full index scan of ModelTestRun which has unacceptable performance.

The tables need to be optimized for the queries that can be run against them. Need to avoid full index scan here.

Need to take this one table at a time right now and work out from there.

Once I finish GCing the old rows and get the DB down to a reasonable size it should be possible to add the necessary new columns here to begin optimizing the queries.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions