[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16384319#comment-16384319
]
Michael Burman commented on CASSANDRA-14281:
Okay, I've managed to get to the stage where
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16382387#comment-16382387
]
Michael Burman commented on CASSANDRA-14281:
It's not the LongAdder/AtomicLong issue, but
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16382261#comment-16382261
]
Chris Lohfink commented on CASSANDRA-14281:
---
Can use a LongAdder array instead of an
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16382251#comment-16382251
]
Michael Burman commented on CASSANDRA-14281:
I tried those tricks, but I don't really see
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16382095#comment-16382095
]
Chris Lohfink commented on CASSANDRA-14281:
---
The decaying EH is an estimated value anyway
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381982#comment-16381982
]
Michael Burman commented on CASSANDRA-14281:
Actually, thinking about it.. why do we need
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381909#comment-16381909
]
Michael Burman commented on CASSANDRA-14281:
It's not the atomic addition that's slow,
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381474#comment-16381474
]
Chris Lohfink commented on CASSANDRA-14281:
---
Code can be written to do it theres a
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381176#comment-16381176
]
Michael Burman commented on CASSANDRA-14281:
The biggest slowdown as far as I could see
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381141#comment-16381141
]
Chris Lohfink commented on CASSANDRA-14281:
---
the update(long)'s rescaleIfNeeded acquires
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16381131#comment-16381131
]
Michael Burman commented on CASSANDRA-14281:
You're right about the EstimatedHistogram,
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16380418#comment-16380418
]
Chris Lohfink commented on CASSANDRA-14281:
---
[
https://issues.apache.org/jira/browse/CASSANDRA-14281?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16380402#comment-16380402
]
Chris Lohfink commented on CASSANDRA-14281:
---
> dropwizard histograms can't be merged
13 matches
Mail list logo