Skip to content

Test and optimize parallel article generation (multi-user queue behavior) #79

@taha-abbasi

Description

@taha-abbasi

Context (from #70)

Hasnat reported that requesting articles from 2 different accounts simultaneously takes noticeably longer.

What to test

  • Queue behavior when 2, 5, 10, 20+ users request articles simultaneously
  • Worker throughput under concurrent load
  • Whether queue is FIFO or has any starvation issues
  • User-facing feedback on queue position and ETA

Taha's note

We'll do some parallel testing to understand queuing and how this scales if 10, or 20, or 100+ people are requesting articles. We have to test this without burning through tons of tokens.

Approach

  • Build a lightweight load test script (mock requests, minimal token usage)
  • Measure queue depth, wait times, completion times
  • Identify bottlenecks before beta launch

Related

Metadata

Metadata

Assignees

No one assigned

    Labels

    P1Importantscribe-launchScribe Launch project boardtestingTesting and QA

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions