fix: revert extractor to 1 replica (single GPU bottleneck)

This commit is contained in:
Celes Renata
2026-04-20 12:16:51 +00:00
parent 7fc54a6023
commit 740ddc1c54
+2 -3
View File
@@ -33,8 +33,7 @@ ingress:
trino: stonks-paper-trino.celestium.life
tradingEngine: stonks-paper-trading.celestium.life
## Scale extractor to 3 replicas — Ollama handles concurrent requests
## and the backlog of SEC filings needs higher throughput
## Scale extractor — single GPU bottleneck, 1 pod is optimal
services:
extractor:
replicas: 3
replicas: 1