This is the shortest path to understanding the current Janus implementation.
- ../README.md
- ../GETTING_STARTED.md
- ../START_HERE.md
- JANUSQL.md
- QUERY_EXECUTION.md
- BASELINES.md
- NESTED_HISTORICAL_SUBQUERIES.md
- HTTP_API_CURRENT.md
- ANOMALY_DETECTION.md
- QUICK_REFERENCE.md
- PAPER_BENCHMARKING.md
- PAPER_ARTIFACT_MAP.md
- PAPER_ARCHITECTURE.md
- PAPER_SUBMISSION_PACKAGE.md
-
- query structure
- supported window types
USING BASELINE <window> LAST|AGGREGATE- how live and historical queries are derived
-
- registration and parsed metadata
start_query()flow- historical workers
- live workers and MQTT subscription
- result multiplexing and runtime status
-
- what baseline bootstrap does
LASTvsAGGREGATE- async warm-up behavior
- what state is and is not retained
-
NESTED_HISTORICAL_SUBQUERIES.md
- nested historical subquery syntax
- planning pipeline and diagnostics
- supported and rejected shapes
- comparison with explicit
DEFINE BASELINE - microbenchmark runner
-
- current REST endpoints
- WebSocket result flow
- request and response shapes
- persisted query lifecycle status
-
- when extension functions are enough
- when baseline state helps
- recommended query patterns
-
- paper-facing benchmark harnesses
- current paper claim destinations
- benchmark output filenames
-
- claim to benchmark mapping
- output files and caveats
- figure and table destinations
-
- unified execution diagram
- decomposed Oxigraph baseline diagram
- nested historical subquery lowering diagram
-
- include and exclude lists
- benchmark artifacts to preserve
- benchmark artifacts to keep local only
-
- common local commands
- query lifecycle endpoints
- replay endpoints
- smoke-test flow
- Maintained dashboard repository:
https://github.com/SolidLabResearch/janus-dashboard