[
https://issues.apache.org/jira/browse/CASSANDRA-7316?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14090953#comment-14090953
]
Joshua McKenzie commented on CASSANDRA-7316:
--------------------------------------------
Your summary gels with both mine and Ala's findings - I'm +1 on closing as
won't fix. Given that windows users have another way to resolve this that's
vastly lower complexity and the current windows ecosystem doesn't seem to value
locking pages in RAM in general, I think we should leave that ball in their
court to configure as they see fit.
We could reduce friction on that choice and modify either the install scripts
in PowerShell or the start-up scripts (or both) w/an optional flag to use
Get-WmiObject on [Win32_PageFileSetting
|http://msdn.microsoft.com/en-us/library/aa394245(v=vs.85).aspx] and disable
the page file for the user if requested (similar to what we do with tcp
keepalive via install script). Another slightly less invasive option would be
to warn if we query and see the page file enabled, suggesting to users that
they should disable paging on their machines in production.
I kind of like the idea of doing both but I'll need to think on it a bit more
before I'm really sold on it.
> Windows feature parity - lock JVM in memory to prevent swapping
> ---------------------------------------------------------------
>
> Key: CASSANDRA-7316
> URL: https://issues.apache.org/jira/browse/CASSANDRA-7316
> Project: Cassandra
> Issue Type: Improvement
> Reporter: Joshua McKenzie
> Assignee: Joshua McKenzie
> Priority: Minor
> Labels: Windows, perfomance
> Fix For: 3.0
>
>
> Similar to mlockall() in CLibrary.java for linux, it would be nice to lock
> the virtual address space on Windows to prevent page faults.
> One option: Reference API:
> http://msdn.microsoft.com/en-us/library/windows/desktop/aa366895(v=vs.85).aspx
--
This message was sent by Atlassian JIRA
(v6.2#6252)
