harshit.cloud
ツ
Senior SRE
Home
Blog
TIL
Tags
Graph
Toggle theme
Back to tags
#scale-to-zero
1 blog post.
Blog posts
Scaling GPU inference to zero and back
Idle GPUs at six dollars an hour are a bonfire. Scaling to zero saves the money, but the first user back waits minutes unless you kill the cold start.
Related tags
#gpu
#inference
#kubernetes
#autoscaling
#scale-to-zero — 1 post · Infra Magician