[
https://issues.apache.org/jira/browse/CASSANDRA-15093?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17268966#comment-17268966
]
Lorina Poland commented on CASSANDRA-15093:
-------------------------------------------
Note for 4.0 GA docs: I have incorporated the information in this ticket in the
new hardware.adoc file.
> update hardware recommendations
> -------------------------------
>
> Key: CASSANDRA-15093
> URL: https://issues.apache.org/jira/browse/CASSANDRA-15093
> Project: Cassandra
> Issue Type: Improvement
> Components: Documentation/Website
> Reporter: Jon Haddad
> Priority: Normal
>
> The recommendations on
> http://cassandra.apache.org/doc/latest/operating/hardware.html are pretty out
> of date. We should improve the following:
> * GC recommendations. We (The Last Pickle) routinely use 16GB heaps with
> 8-10GB of new gen, Survivor ratio of 4-6 and max tenuring threshold of 6. It
> works far better than G1GC
> * The instance sizes in AWS are pretty outdated. M1s are years out of date.
> i3's work well with NVMe disks, EBS works OK if you want easy backups and
> replacements.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
