This is the mail archive of the
cygwin
mailing list for the Cygwin project.
Re: Somebody having access to a Windows machine with > 64 CPUs?
- From: "Brian J. Johnson" <bjohnson at sgi dot com>
- To: cygwin at cygwin dot com
- Date: Tue, 11 Aug 2015 12:57:53 -0500
- Subject: Re: Somebody having access to a Windows machine with > 64 CPUs?
- Authentication-results: sourceware.org; auth=none
- References: <20150810160011 dot GC13029 at calimero dot vinschen dot de> <6E689F0D-F196-4D1B-B6FA-782A27AF936C at etr-usa dot com> <20150810165604 dot GE13029 at calimero dot vinschen dot de> <87lhdjrs54 dot fsf at Rainer dot invalid>
On 08/10/2015 12:05 PM, Achim Gratz wrote:
Corinna Vinschen writes:
I was referring to Windows 7 because that's the first OS (including
it's 2008R2 server version) which supports more than 64 CPUs and the
OS calls required to use and fetch info on them,
GetLogicalProcessorInformationEx.
I think the only practical use of this possibility might have been in
the HPC space. I don't know anybody running such a system, though.
In-memory databases also benefit from really large systems. Lots of RAM
requires lots of sockets, which bring along lots of cores. Eg.
http://www.sgi.com/company_info/newsroom/press_releases/2015/july/sap_hana.html
I believe KVM will let you define a VM with > 64 CPUs, and recent
versions provide some control over the NUMA layout of the virtual
sockets/cores/threads. Maybe that will give you what you're looking
for, Corinna.
--
Brian J. Johnson
--------------------------------------------------------------------
My statements are my own, are not authorized by SGI, and do not
necessarily represent SGI’s positions.
--
Problem reports: http://cygwin.com/problems.html
FAQ: http://cygwin.com/faq/
Documentation: http://cygwin.com/docs.html
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple