On Mon, May 3, 2021 at 6:50 PM Li Wang wrote:
> ...
>
> Please let me know if you have any thoughts or questions.
>
My thought is that you have done some nice work.
>> because the tests were run with Prometheus enabled, which is new in 3.6
and has significant negative perf impact.
Interesting, let's see what the numbers are without Prometheus involved. It
could be that the increased latency we observed in CommitProcessor is just
a symptom rather than the
Hi,
Thanks Ted and Enrico again for the inputs and discussions. I would like to
share some updates from my side.
1. The main issue is that Prometheus summary computation is expensive and
locked/synchronized. To reduce the perf impact, I changed the
PrometheusLabelledSummary.add() operation to be
Li Wang created ZOOKEEPER-4289:
--
Summary: Reduce the performance impact of Prometheus metrics
Key: ZOOKEEPER-4289
URL: https://issues.apache.org/jira/browse/ZOOKEEPER-4289
Project: ZooKeeper
Hi Michael,
Thanks for your additional inputs.
On Mon, May 3, 2021 at 3:13 PM Michael Han wrote:
> Hi Li,
>
> Thanks for following up.
>
> >> write_commitproc_time_ms were large
>
> This measures how long a local write op hears back from the leader. If it's
> big, then either the leader is
Hi Li,
Thanks for following up.
>> write_commitproc_time_ms were large
This measures how long a local write op hears back from the leader. If it's
big, then either the leader is very busy acking the request, or your
network RTT is high.
How does the local fsync time (fsynctime) look like