~/talk

Monitoring Energy Consumption and Building Energy Efficient Systems the Cloud Native Way

links:none on file
slides:none on file

Kepler measures per-pod power by probing kernel counters via eBPF and exports Prometheus metrics that Grafana renders; a model server (TensorFlow Keras + Flask) trains two linear models for CPU core and DRAM energy and serves predictions. Integration works on MicroShift with cgroup v2 and kernel headers; a live run of a Monte Carlo CPU-bound pod shows pod energy tracks core energy closely while DRAM energy is negligible, confirming the approach and enabling scheduler policies.

AI-written summary · about

  1. 0:00Start
  2. 1:54Energy Measurement Methodology: Ideal vs Reality
  3. 8:24About Kepler Model Server
  4. 9:55Implementation of Kepler Model Server
  5. 12:20Peaks
~/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.