[jira] [Commented] (CASSANDRA-14281) LatencyMetrics performance

["Chris Lohfink (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16381474#comment-16381474 ] 

Chris Lohfink commented on CASSANDRA-14281:
-------------------------------------------

Code can be written to do it theres a combineHistograms in TableMetrics i know that does it, but for the decaying one it might need to rescale to normalize both histograms to same weights. Not sure why it would be faster, both cases your incrementing the atomic integer for the bucket? Which HdrHistogram where you using in your tests? Concurrent one?

> LatencyMetrics performance
> --------------------------
>
>                 Key: CASSANDRA-14281
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-14281
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>            Reporter: Michael Burman
>            Priority: Major
>
> Currently for each write/read/rangequery/CAS touching the CFS we write a latency metric which takes a lot of processing time (up to 66% of the total processing time if the update was empty).
> The way latencies are recorded is to use both a dropwizard "Timer" as well as "Counter". Latter is used for totalLatency and the previous is decaying metric for rates and certain percentile metrics. We then replicate all of these CFS writes to the KeyspaceMetrics and globalWriteLatencies.
> For example, for each CFS write we do first write to the CFS's metrics and then to Keyspace's metrics and finally globalMetrics. The way Timer is built is to maintain a Histogram and a Meter and update both when Timer is updated.  The Meter then updates 4 different values (1 minute rate, 5 minute rate, 15 minutes rate and a counter).
> So for each CFS write we actually do 15 different counter updates. And then of course maintain their states at the same time while writing. These operations are very slow when combined.
> A small JMH benchmark doing an update against a single LatencyMetrics with 4 threads gives us around 5.2M updates / second. With the current writeLatency metric (having 2 parents) we get only 1.6M updates / second. 
> I'm proposing to update this to use a small circular buffer HdrHistogram implementation. We would maintain a rolling buffer with last 15 minutes of histograms (30 seconds per histogram) and update the correct bucket each time. When requesting metrics we would then merge requested amount of buckets to a new histogram and parse results from it. This moves some of the load from writing of the metrics to reading them (which is much more infrequent operation), including the parent metrics. It also allows us to maintain the current metrics structure - if we wish to do so.
> My prototype with this approach improves the performance to around 13.8M updates/second, thus almost 9 times faster than the current approach. We also maintain HdrHistogram already in the Cassandra's lib so there's no new dependencies to add (java-driver also uses it). 
> FUTURE:
> This opens up some possibilities, such as replacing all dropwizard Histograms/Meters with the new approach (to reduce overhead elsewhere in the codebase). It would also allow us to supply downloadable histograms directly from the Cassandra or store them to the disk each time a bucket is filled if user wishes to monitor latency history or graph all percentiles. 
> HdrHistogram also provides the ability to "fix" these histograms with pause tracking, such as GC pauses which we currently can't do (as dropwizard histograms can't be merged).



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

---------------------------------------------------------------------
To unsubscribe, e-mail: commits-unsubscribe@cassandra.apache.org
For additional commands, e-mail: commits-help@cassandra.apache.org