Apple a605b8c43e P3.1: GPU/Queue-aware routing — NCS metrics + scoring-based model selection ...

NCS (services/node-capabilities/metrics.py):
- NodeLoad: inflight_jobs, queue_depth, concurrency_limit, estimated_wait_ms,
  cpu_load_1m, mem_pressure (macOS + Linux), rtt_ms_to_hub
- RuntimeLoad: per-runtime healthy, p50_ms, p95_ms from rolling 50-sample window
- POST /capabilities/report_latency for node-worker → NCS reporting
- NCS fetches worker metrics via NODE_WORKER_URL

Node Worker:
- GET /metrics endpoint (inflight, concurrency, latency buffers)
- Latency tracking per job type (llm/vision) with rolling buffer
- Fire-and-forget latency reporting to NCS after each successful job

Router (model_select v3):
- score_candidate(): wait + model_latency + cross_node_penalty + prefer_bonus
- LOCAL_THRESHOLD_MS=250: prefer local if within threshold of remote
- ModelSelection.score field for observability
- Structured [score] logs with chosen node, model, and score breakdown

Tests: 19 new (12 scoring + 7 NCS metrics), 36 total pass
Docs: ops/runbook_p3_1.md, ops/CHANGELOG_FABRIC.md

No breaking changes to JobRequest/JobResponse or capabilities schema.

Made-with: Cursor

2026-02-27 02:55:44 -08:00

..

agent-e2e-prober

feat(P0/P1/P2): Add E2E agent prober, version pinning, prometheus fixes

2026-01-28 07:06:07 -08:00

ai-security-agent

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

artifact-registry

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

brand-intake

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

brand-registry

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

chandra-inference

feat: implement TTS, Document processing, and Memory Service /facts API

2026-01-17 08:16:37 -08:00

chandra-service

feat: implement TTS, Document processing, and Memory Service /facts API

2026-01-17 08:16:37 -08:00

clan-consent-adapter

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

clan-visibility-guard

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

comfy-agent

services: update comfy agent, senpai md consumer, and swapper deps

2026-02-19 00:14:18 -08:00

config-registry

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

control-plane

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

crewai-service

agents: add planned AISTALK orchestrator and crew profile

2026-02-20 10:24:59 -08:00

crewai-worker

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

docling-service

feat: implement TTS, Document processing, and Memory Service /facts API

2026-01-17 08:16:37 -08:00

image-gen-service

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

index-doc-worker

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

ingest-service

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

integration-service

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

market-data-service

feat: MD pipeline — market-data-service hardening + SenpAI NATS consumer

2026-02-09 11:46:15 -08:00

memory-service

feat: harden memory summary — fingerprint dedup, versioning, prompt injection defense

2026-02-09 10:26:03 -08:00

node-capabilities

P3.1: GPU/Queue-aware routing — NCS metrics + scoring-based model selection

2026-02-27 02:55:44 -08:00

node-worker

P3.1: GPU/Queue-aware routing — NCS metrics + scoring-based model selection

2026-02-27 02:55:44 -08:00

oneok-calc-adapter

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

oneok-crm-adapter

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

oneok-docs-adapter

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

oneok-schedule-adapter

services: add clan consent/visibility and oneok adapter stack

2026-02-19 00:14:12 -08:00

parser-pipeline

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

presentation-renderer

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

rag-service

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

render-pdf-worker

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

render-pptx-worker

feat: Add Alateya, Clan, Eonarch agents + fix gateway-router connection

2026-01-28 06:40:34 -08:00

router

P3.1: GPU/Queue-aware routing — NCS metrics + scoring-based model selection

2026-02-27 02:55:44 -08:00

scheduler

snapshot: NODE1 production state 2026-02-09

2026-02-09 08:46:46 -08:00

senpai-md-consumer

services: update comfy agent, senpai md consumer, and swapper deps

2026-02-19 00:14:18 -08:00

swapper-service

node2: P0 vision restore + P1 security hardening + node-specific router config

2026-02-27 01:27:38 -08:00

vision-encoder

feat: add Vision Encoder service + Vision RAG implementation

2025-11-17 05:24:36 -08:00