[ 
https://issues.apache.org/jira/browse/CASSANDRA-7546?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

graham sanderson updated CASSANDRA-7546:
----------------------------------------

    Attachment: hint_spikes.png
                young_gen_gc.png

I thought I'd attach an image of the correlation between the hint spikes and 
the massive memory allocation. There is a chicken and egg thing here, but if a 
node starts hinting heavily, it may become unresponsive causing a knock on 
effect.

Note on the yellow node, there was a period of about 500 seconds of young gen 
GC, which equates to about 3TB of garbage

> AtomicSortedColumns.addAllWithSizeDelta has a spin loop that allocates memory
> -----------------------------------------------------------------------------
>
>                 Key: CASSANDRA-7546
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-7546
>             Project: Cassandra
>          Issue Type: Bug
>          Components: Core
>            Reporter: graham sanderson
>            Assignee: graham sanderson
>             Fix For: 2.1.1
>
>         Attachments: 7546.20.txt, 7546.20_2.txt, 7546.20_3.txt, 
> 7546.20_4.txt, 7546.20_5.txt, 7546.20_6.txt, 7546.20_7.txt, 7546.20_7b.txt, 
> 7546.20_alt.txt, hint_spikes.png, suggestion1.txt, suggestion1_21.txt, 
> young_gen_gc.png
>
>
> In order to preserve atomicity, this code attempts to read, clone/update, 
> then CAS the state of the partition.
> Under heavy contention for updating a single partition this can cause some 
> fairly staggering memory growth (the more cores on your machine the worst it 
> gets).
> Whilst many usage patterns don't do highly concurrent updates to the same 
> partition, hinting today, does, and in this case wild (order(s) of magnitude 
> more than expected) memory allocation rates can be seen (especially when the 
> updates being hinted are small updates to different partitions which can 
> happen very fast on their own) - see CASSANDRA-7545
> It would be best to eliminate/reduce/limit the spinning memory allocation 
> whilst not slowing down the very common un-contended case.



--
This message was sent by Atlassian JIRA
(v6.2#6252)

Reply via email to