Menu

#Concurrent

17 posts

Feed·
16 of 17 posts
Notes from a 2 AM On-Call: Why Our Bifrost Gateway Silently Dropped 40% of Streaming Requests
🖼️
0Read More

Notes from a 2 AM On-Call: Why Our Bifrost Gateway Silently Dropped 40% of Streaming Requests

DEV Community: cloud·James LIN·20 days ago
#70yoi1bz
#devtocloud#devto#dev#streaming#nginx

Postmortem of a production incident where maximhq/bifrost's SSE streaming silently failed under concurrent load—and the 3-line nginx buffer fix that resolved it.

15s
Notes from a 2 AM On-Call: Ollama's Model Hot-Swap Race Condition Under Concurrent Requests
🖼️
0Read More

Notes from a 2 AM On-Call: Ollama's Model Hot-Swap Race Condition Under Concurrent Requests

DEV Community: go·Kevin Tian·24 days ago
#qyuDFuwe
#devtogo#devto#dev#model#ollama

Debugging silent model unload failures when Ollama handles concurrent inference requests during model switching—and the mutex guard that fixed it.

15s
Pro teams can now run up to 500 concurrent builds - Vercel
📰
0Read More

Pro teams can now run up to 500 concurrent builds - Vercel

Vercel News·David Lee·25 days ago
#Z2ldCbM8
#vercel#builds#concurrent#teams#demand

Pro now supports up to 500 concurrent builds on Vercel, up from 12. Large repos and monorepos ship faster without waiting on queued builds.

15s
Pricing for on-demand concurrent builds reduced by over 50% - Vercel
📰
0Read More

Pricing for on-demand concurrent builds reduced by over 50% - Vercel

Vercel News·Mariano Cocirio·6 months ago
#MMpnTDeF
#vercel#demand#builds#concurrent#build

Pricing for on-demand concurrent builds has been reduced by more than 50%. Billable intervals have also been lowered from 10 minutes to 1 minute.

15s