On Tuesday, 09/04/2007 at 10:45 EDT, Rick Troth [EMAIL PROTECTED] wrote:
I cannot offer MONEY to the IBM team(s) which support 31-bit
Linux for S/390. But to abandon a platform is counter to
the whole effort behind Linux. (Not that platforms aren't lost.)
Huh? I thought Linux was about
On Sep 5, 2007, at 10:41 AM, Alan Altmark wrote:
If IBM actively encouraged distributors to build and ship an S/390
version
of the new distros, while at the same time not supporting any of
its own
software on it, the message would be, umm, mixed at best;
misleading at
worst. And I'm not sure
On 9/4/07, Rick Troth [EMAIL PROTECTED] wrote:
I cannot wait for Barton and company to measure it!
Don't hold your breath... Most of the measurements we do is because
of real customer requirements to reduce their TCO or improve
performance without increasing cost. And most of our customers do
Rob van der Heij wrote:
On 9/4/07, Rick Troth [EMAIL PROTECTED] wrote:
[snip]
PS The wonderful thing about Linux on z/VM performance is that there
often are multiple factors involved, and you cannot really predict the
effect without measuring. As for running 32-bit applications in a
On Tue, Sep 4, 2007 at 7:32 AM, in message
[EMAIL PROTECTED], Mario
Held [EMAIL PROTECTED] wrote:
-snip-
And given the situation that the recent distributions are only available
as 64-bit flavor the question is 'Do I stay with my old 31-bit kernel version
or am I interested in the new
On Tue, 4 Sep 2007, Mark Post wrote:
I want to point out that the situation that the recent distributions
are only available as 64-bit was at the insistence of IBM itself.
I personally still think that 31-bit versions are ...
Well ... that is sad. Can you substantiate that?
I cannot offer
On Tue, Sep 4, 2007 at 10:44 PM, in message
[EMAIL PROTECTED], Rick Troth
[EMAIL PROTECTED] wrote:
On Tue, 4 Sep 2007, Mark Post wrote:
I want to point out that the situation that the recent distributions
are only available as 64-bit was at the insistence of IBM itself.
I personally still
Assuming that I have a workload which will run effectively in 31 bit
mode, does anybody know if it will run more efficiently in 31 bit mode
than in 64 bit mode. That is, does 64 bit mode on System z have more
overhead (either hardware or software) than 31 bit mode? Or is this
another it depends
On Mon, 3 Sep 2007, Mario Held wrote:
If you have the source of your application available you can compile the
application with and without the '-m31' parameter and test the performance
in your special case. Then sometimes the answer is it depends of
the scenario.
But if the distributors are
Rick Troth wrote:
But if the distributors are only shipping 64-bit kernels, then '-m31'
still does not completely answer the question about 31-bit performance.
The rest of Linux will be running 64-bit, skewing the results.
-- R;
Not necessarily. Granted, the process will be running in
I wrote:
But if the distributors are only shipping 64-bit kernels, then '-m31'
still does not completely answer the question about 31-bit performance.
The rest of Linux will be running 64-bit, skewing the results.
What I was trying to say is that a 31-bit program running on a
64-bit Linux
Morning, John.
Yes, for the most part, running in 64-bit mode will not be quite as
efficient as running the same application in 31 bit mode. CP has more
memory to keep track of, and that means larger page tables, TLBs etc.
If the guest is Linux, then Linux also has more work to do in it's
On Tuesday, 08/28/2007 at 09:58 EDT, Dave Jones [EMAIL PROTECTED]
wrote:
Yes, for the most part, running in 64-bit mode will not be quite as
efficient as running the same application in 31 bit mode. CP has more
memory to keep track of, and that means larger page tables, TLBs etc.
If the guest
Well, maybe. What about the case where, say, a WAS server is running 31-bit and
is fighting with itself with multiple address spaces for heap storage, and all
must below the 2G line? ... and can be thrashing around doing this management
all the time? That's not good.
I really don't say a few
On 8/29/07, David Kreuter [EMAIL PROTECTED] wrote:
Well, maybe. What about the case where, say, a WAS server is running 31-bit
and is fighting with itself with multiple address spaces for heap storage,
and all must below the 2G line? ... and can be thrashing around doing this
management
64-bit has more storage overhead for sure. measurably impacts working sets and
thus paging.
McKown, John wrote:
Assuming that I have a workload which will run effectively in 31 bit
mode, does anybody know if it will run more efficiently in 31 bit mode
than in 64 bit mode. That is, does 64
Assuming that I have a workload which will run effectively in 31 bit
mode, does anybody know if it will run more efficiently in 31 bit mode
than in 64 bit mode. That is, does 64 bit mode on System z have more
overhead (either hardware or software) than 31 bit mode? Or is this
another it depends
17 matches
Mail list logo
