Memory and CPU for a single asset node at the end of the most recent epoch. This page is deliberately one point in time: the series page carries the history.
Bottleneck
A time budget per operation, from four independent sources: the operation's own duration, the daemon's cgroup CPU delta across exactly that operation, goroutine blocking on real dependencies, and summed statement time. Whichever dominates names the class. Rows are clickable for the contributing functions and queries.
| operation | node | verdict | wall | cpu | block | mutex | db |
|---|
Anonymous memory is what the daemon holds. Page cache is the
kernel caching files the daemon read, mostly the proof archive, and is shown
dimmed because it is not the daemon retaining anything. The docker figure is
what docker stats would print, and it folds active page cache in,
which is why it reads higher than anon.
Taken as the difference between the cumulative cgroup counters before and after the run, so it is CPU actually burned by this epoch rather than an instantaneous sample.
| name | flat | share | cumulative |
|---|
From the per-method histogram that --prometheus.perfhistograms
adds. Counts and totals are cumulative since the daemon started, so they show
which part of the API the load actually exercises.
| method | calls | mean | total |
|---|