~/talk

Failing to Autoscale?

on YouTube 31m
Talkat SREday 2023 London · 2023-09-14..09-15· [machine-learning, kubernetes]

speaker

name:Erwin Daria
linkedin:erwindaria

resources

video:https://youtu.be/qkSO23iSPFc
slides:none on file

summary

Claims that machine learning right-sizes Kubernetes pod CPU/memory requests to stop the cluster autoscaler from adding costly nodes, reporting about 50% reduction in resource requests and an average 8:1 ROI when customers apply recommendations. Implementation consists of a small agent (one container plus a Prometheus scraper) that samples every 15s, retains a year of data but trains on two weeks, emits YAML patch recommendations and an auto-apply with rollback mode, and integrates with HPA/VPA workflows and AWS Carpenter.

AI-written summary · about

chapters (11)

  1. 0:00Introduction and company background
  2. 3:32Auto scaling challenges in Kubernetes
  3. 6:21Over and under provisioning scenarios
  4. 8:06Market opportunity and waste in cloud costs
  5. 10:12StormForge solution overview
  6. 13:02Demo setup and agent installation
  7. 16:32Recommendation generation and implementation
  8. 18:21Recap and value proposition
  9. 21:10Cluster Auto Scaler and AWS partnership
  10. 22:36Closing and future vision
  11. 22:56Q&A session

related talks

~/search
type to search 6,249 conferences, 138,987 speakers and 185,172 talks
~/help
cmdctrl k
search every conference, speaker and series
/
focus the filter search
h
home
c
conferences
s
speakers
e
series
t
light / dark — the status bar button does the same
?
this help
esc
close this, then clear every filter — or just the search, while you are typing in it

Ignored while you are typing. Pages without a filter rail have no search to focus.