Skip to content
This repository was archived by the owner on Apr 24, 2024. It is now read-only.
This repository was archived by the owner on Apr 24, 2024. It is now read-only.

Benchmark heatmap algorithm #1257

@chr-schr

Description

@chr-schr

Tasks

  • Benchmark heatmap algorithm with 10 / 100 / 1000 shadings
  • Profile heatmap SQL function with plprofiler
  • Add and benchmark variations
    • v1
    • v2
  • add README.md

Implemented:

  • v1: Fetch plant relations once
    • => Big improvement for plant relationship score calculation
  • v2: Inline everything (because PL/pgSQL functions are "optimization fences")
    • little or negative effect for larger maps; some improvements for smaller maps
    • terrible maintainability

Other ideas:

  • Hard cutoff of effect after some distance (needs discussion)
    • Effect of plant relationships decreases quadratically
  • Return bad values early (needs discussion)
    • e.g. really bad shading cannot be improved with good companions and vice versa
    • may become even more important later on, when more layers are added to the heatmap

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

Status

Current Sprint

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions