[
https://issues.apache.org/jira/browse/CASSANDRA-20250?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17925572#comment-17925572
]
Dmitry Konstantinov commented on CASSANDRA-20250:
-------------------------------------------------
{quote}I would guess there are still some long adders in cache, and that we may
to some approximation be measuring the additional cost of hitting L3).
{quote}
Yes, I targeted L1/L2 originally, I agree - for L3 the array definitely should
be bigger.
My server CPU is quite old:
{code:java}
Intel(R) Xeon(R) CPU E5-2680 v4
L1 Instruction Cache: 32 KB per core
L1 Data Cache: 32 KB per core
L2 Cache: 256 KB per core
L3 Cache: 35 MB{code}
{quote}I guess the next thing might be to try integrating these classes into
Cassandra?
{quote}
Yes, I will try to do an e2e version this week.
Regarding int[]/delta approach
For the initial 2 long[] option I think I understand how to do it: we have 2
thread-local arrays: long[] current, long[] lastReported and one global long[]
summaries, when we read a value we add to summaries a delta between current and
lastReported and then update lastReported.
like:
increment() -> current[metricId]++;
getCount() -> current = current[metricId]; summaries[metricId] += current -
lastReported[metricId]; lastReported[metricId] = current;
But so far I cannot find a way to do it for a counter using int[] deltas
without an extra-syncrhonization...
Did you describe it for the general counters case or it was about a more
specific case for meters only, when reset to 0 is done by writing threads
themselves?
> Provide the ability to disable specific metrics collection
> ----------------------------------------------------------
>
> Key: CASSANDRA-20250
> URL: https://issues.apache.org/jira/browse/CASSANDRA-20250
> Project: Apache Cassandra
> Issue Type: New Feature
> Components: Observability/Metrics
> Reporter: Dmitry Konstantinov
> Assignee: Dmitry Konstantinov
> Priority: Normal
> Attachments: 5.1_profile_cpu.html,
> 5.1_profile_cpu_without_metrics.html, async_profiler_cpu_profiles.zip,
> cpu_profile_insert.html, jmh-result.json, vmstat.log,
> vmstat_without_metrics.log
>
>
> Cassandra has a lot of metrics collected, many of them are collected per
> table, so their instance number is multiplied by number of tables. From one
> side it gives a better observability, from another side metrics are not for
> free, there is an overhead associated with them:
> 1) CPU overhead: in case of simple CPU bound load: I already see like 5.5% of
> total CPU spent for metrics in cpu framegraphs for read load and 11% for
> write load.
> Example: [^cpu_profile_insert.html] (search by "codahale" pattern). The
> framegraph is captured using Async profiler build:
> async-profiler-3.0-29ee888-linux-x64
> 2) memory overhead: we spend memory for entities used to aggregate metrics
> such as LongAdders and reservoirs + for MBeans (String concatenation within
> object names is a major cause of it, for each table+metric name combination a
> new String is created)
>
> The idea of this ticket is to allow an operator to configure a list of
> disabled metrics in cassandra.yaml, like:
> {code:java}
> disabled_metrics:
> - metric_a
> - metric_b
> {code}
> From implementation point of view I see two possible approaches (which can be
> combined):
> # Generic: when a metric is registering if it is listed in disabled_metrics
> we do not publish it via JMX and provide a noop implementation of metric
> object (such as histogram) for it.
> Logging analogy: log level check within log method
> # Specialized: for some metrics the process of value calculation is not for
> free and introduces an overhead as well, in such cases it would be useful to
> check within specific logic using an API (like: isMetricEnabled) do we need
> to do it. Example of such metric:
> ClientRequestSizeMetrics.recordRowAndColumnCountMetrics
> Logging analogy: an explicit 'if (isDebugEnabled())' condition used when a
> message parameter is expensive.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
