-
Notifications
You must be signed in to change notification settings - Fork 3.8k
monitor: Reduce overhead #1926
Copy link
Copy link
Closed
Labels
area/monitorImpacts monitoring, access logging, flow logging, visibility of datapath traffic.Impacts monitoring, access logging, flow logging, visibility of datapath traffic.kind/enhancementThis would improve or streamline existing functionality.This would improve or streamline existing functionality.priority/mediumThis is considered important, but not urgent.This is considered important, but not urgent.
Milestone
Metadata
Metadata
Assignees
Labels
area/monitorImpacts monitoring, access logging, flow logging, visibility of datapath traffic.Impacts monitoring, access logging, flow logging, visibility of datapath traffic.kind/enhancementThis would improve or streamline existing functionality.This would improve or streamline existing functionality.priority/mediumThis is considered important, but not urgent.This is considered important, but not urgent.
We are encoding / decoding a lot. I think this is causing the node monitor to show up in perf and causes the CPU spikes. Could we recycle some of the arrays and shrink / grow them instead of creating new ones?
Options: