Summary
Currently, the performance charts on Atlas take 15-20 seconds to load/reload and Atlas Charts is sending "Query Targeting: Scanned Objects / Returned has gone above 1000" warnings indicating we're performing inefficient queries on our perf data. This is because currently all Atlas perf data is stored in a single collection, but each chart only wants a subset of that data which applies to a specific test.
To fix this we should restructure the DB such that each test has its own collection, or index on test names to avoid needless search.

Motivation

• Does this affect any team outside of WT?
  No

• How likely is it that this use case or problem will occur?
  This will happen on every commit moving forward.

• If the problem does occur, what are the consequences and how severe are they?
  The queries are inefficient, but the DB is still small enough that this is not expensive or overly slow. The main problem is email spam sent to Chart admins.

• Is this issue urgent?
  No.

Scope

• Investigate what can be improved to optimise charts.
• Further work can be decided based on the investigation.

Acceptance Criteria (Definition of Done)

Atlas Charts no longer sends warning emails about inefficient queries.{}