Skip to content

About

Agent workspace for optimising ijson, the data structure under RedisJSON, with RedisJSON as the usage-pattern reference

Resources

Stars

0 stars

Watchers

0 watching

Forks

Repository files navigation

json-agent-workspace

Agent workspace for optimising the data structure underneath RedisJSON: ijson and the dependencies it builds on.

Status: scaffold. No benchmark suite, baseline or experiment exists yet.

Layout

Path What
ijson/ The data structure, pinned to the revision RedisJSON builds (379dadf). Primary target.
ciborium/ RedisJSON's CBOR fork, a dependency of ijson.
RedisJSON/ The consumer, pinned at 9ca1c88. Reference only: it defines which operations matter.
docs/USAGE-PATTERN.md How RedisJSON calls ijson, and the call paths to benchmark.
docs/DEPENDENCIES.md Which dependencies are in scope, and which RedisJSON-org repos are not.
docs/PERFORMANCE-MCP.md, docs/TMA-REPORTING.md Required per-session and per-experiment rules.
experiments/, EXPERIMENTS.md The experiment ledger.
AGENTS.md, .claude/ Agent instructions.

Getting started

git clone --recurse-submodules https://github.com/redis-performance/json-agent-workspace.git
cd json-agent-workspace/ijson && cargo test --lib -- --test-threads=1

Next steps

  1. Profile RedisJSON command workloads to weight the call paths in docs/USAGE-PATTERN.md.
  2. Build and freeze an ijson microbenchmark suite that mirrors those call paths.
  3. Record a baseline (time and memory) per architecture, then start experiments.

About

Agent workspace for optimising ijson, the data structure under RedisJSON, with RedisJSON as the usage-pattern reference

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors