bvandeusen/FabledCurator
1
0
Fork 0
Code Issues 1 Pull Requests Actions Packages Projects Releases 35 Wiki Activity

feat(agent): raise worker cap to 32 + size the HTTP pool for it (#114)
CI / lint (push) Successful in 3s
Details
CI / frontend-build (push) Successful in 19s
Details
CI / backend-lint-and-test (push) Successful in 28s
Details
CI / integration (push) Successful in 3m35s
Details

Browse Source 
At 8 workers the GPU sat at ~5% util / <5GB VRAM — the pipeline is I/O-bound
(downloading + decoding images over HTTP), so the GPU starves until many workers
overlap that I/O. Raise MAX_CONCURRENCY 8→32 and make the UI worker control a
number input (reaching 32 by ±1 was tedious); the cap is reported via /status so
the UI clamps to it. Also size the shared requests pool (pool_maxsize=64) — the
default 10 would have throttled 32 workers + spammed "connection pool is full".

Verified by running; watch GPU util/VRAM climb as you dial up.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01Ttrj5P7upUTueSfoJcxEqa
This commit is contained in:
Bryan Van Deusen
2026-06-29 19:41:52 -04:00
parent 3abbe58450
commit b7fd69815e
3 changed files with 23 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files
agent/fc_agent/app.py
+12 -5
View File
@@ -75,9 +75,11 @@ _PAGE = """<!doctype html><html><head><meta charset=utf-8>
<div class=row> <div class=row>
workers workers
<button class=step onclick=setc(-1)>−</button> <button class=step onclick=setc(-1)>−</button>
<b id=conc style=margin:0+.5rem>1</b> <input id=conc type=number min=1 value=1
style="width:3.5rem;font:700 16px system-ui;text-align:center;background:#222;color:#e8e8e8;border:1px solid #444;border-radius:6px;padding:.3rem"
onchange="setv(this.value)">
<button class=step onclick=setc(1)>+</button> <button class=step onclick=setc(1)>+</button>
<span class=cap style=color:#9aa>(more = faster + more GPU)</span> <span class=cap style=color:#9aa>(more = overlap I/O, fill the GPU) max <b id=capn>8</b></span>
</div> </div>
<div class=row> <div class=row>
<span class=stat><span class=n id=state>stopped</span><br>state</span> <span class=stat><span class=n id=state>stopped</span><br>state</span>
@@ -89,16 +91,21 @@ _PAGE = """<!doctype html><html><head><meta charset=utf-8>
<div class=bar><i id=gpubar style=width:0%></i></div> <div class=bar><i id=gpubar style=width:0%></i></div>
<div class=q id=queue></div> <div class=q id=queue></div>
<script> <script>
let CAP=8
async function act(p){await fetch('/'+p,{method:'POST'});refresh()} async function act(p){await fetch('/'+p,{method:'POST'});refresh()}
async function setc(d){ function setc(d){ setv((parseInt(conc.value||'1'))+d) }
const v=Math.max(1,Math.min(8,parseInt(conc.textContent||'1')+d)) async function setv(v){
v=Math.max(1,Math.min(CAP,parseInt(v)||1)); conc.value=v
await fetch('/concurrency',{method:'POST',headers:{'Content-Type':'application/json'}, await fetch('/concurrency',{method:'POST',headers:{'Content-Type':'application/json'},
body:JSON.stringify({value:v})});refresh() body:JSON.stringify({value:v})});refresh()
} }
async function refresh(){ async function refresh(){
const s=await (await fetch('/status')).json() const s=await (await fetch('/status')).json()
CAP=s.max_concurrency||8; capn.textContent=CAP
state.textContent=s.state; active.textContent=s.active; done.textContent=s.processed state.textContent=s.state; active.textContent=s.active; done.textContent=s.processed
err.textContent=s.errors; conc.textContent=s.concurrency; fc.textContent=s.fc_url err.textContent=s.errors; fc.textContent=s.fc_url
if(document.activeElement!==conc) conc.value=s.concurrency
conc.max=CAP
cfg.textContent=s.configured?'set':'MISSING' cfg.textContent=s.configured?'set':'MISSING'
if(s.gpu){ if(s.gpu){
gpu.textContent=`GPU — ${s.gpu.util_pct}% util · VRAM ${s.gpu.mem_used_mb}/${s.gpu.mem_total_mb} MB · ${s.gpu.temp_c}°C` gpu.textContent=`GPU — ${s.gpu.util_pct}% util · VRAM ${s.gpu.mem_used_mb}/${s.gpu.mem_total_mb} MB · ${s.gpu.temp_c}°C`
agent/fc_agent/client.py
+6
View File
@@ -4,6 +4,7 @@ The agent's ONLY contact with FC — lease/submit/heartbeat/fail + fetch image
bytes, all over HTTP with the bearer token. No DB/Redis. bytes, all over HTTP with the bearer token. No DB/Redis.
""" """
import requests import requests
from requests.adapters import HTTPAdapter
class FcClient: class FcClient:
@@ -12,6 +13,11 @@ class FcClient:
self.agent_id = agent_id self.agent_id = agent_id
self.s = requests.Session() self.s = requests.Session()
self.s.headers["Authorization"] = f"Bearer {token}" self.s.headers["Authorization"] = f"Bearer {token}"
# Many worker threads share this Session; the default pool (10) would
# throttle them + spam "connection pool is full". Size it for the cap.
adapter = HTTPAdapter(pool_connections=64, pool_maxsize=64)
self.s.mount("http://", adapter)
self.s.mount("https://", adapter)
def lease(self, batch_size: int) -> list[dict]: def lease(self, batch_size: int) -> list[dict]:
r = self.s.post( r = self.s.post(
agent/fc_agent/worker.py
+5 -1
View File
@@ -17,7 +17,10 @@ from .client import FcClient
from .config import Config from .config import Config
from .crops import crop_region from .crops import crop_region
MAX_CONCURRENCY = 8 # Generous cap: the pipeline is usually I/O-bound (downloading + decoding images
# over HTTP), so the GPU stays underused until many workers overlap that I/O.
# Push it up while watching the GPU util + VRAM in the UI.
MAX_CONCURRENCY = 32
class _Slot: class _Slot:
@@ -74,6 +77,7 @@ class Worker:
return { return {
"state": "running" if self._running else "stopped", "state": "running" if self._running else "stopped",
"concurrency": self._target, "concurrency": self._target,
"max_concurrency": MAX_CONCURRENCY,
"workers": len(self._slots), "workers": len(self._slots),
"active": self._active, "active": self._active,
"processed": self.processed, "processed": self.processed,