*All* resources have cost attributes. I'm quite sure Netflix, for example,
spends a huge sum on computing resources, although to my knowledge they
don't have a mainframe -- at least not one of their own. (Maybe they
should!) I would also point out that the world's top wealthiest people
often
Thank you for clarifying the threadability of an IFL CP.
I worked w/ a fellow who went to work in one of the labs in Vermont working
on "hyperthreading" for the pSeries CPUs... I seem to recall him talking
about efforts for the zSeries CPs but that seemed a long ask.
-soup
On Wed, Nov 8, 2017
z/OS supports ZIIPs with SMT-2 as well.
-Original Message-
From: Alan Altmark [mailto:alan_altm...@us.ibm.com]
Sent: Wednesday, November 8, 2017 5:52 PM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: How many Intel cores does an IFL emulate
Thanks for advertising, Soup! :-)
In fact, IFLs
Thanks for advertising, Soup! :-)
In fact, IFLs are the only engines on Z that have SMT-2, where each core in
the LPAR splits in two, appearing to the host as twice the number of CPUs.
Alan
Sent from my iPhone using IBM Verse
On Nov 8, 2017, 8:36:39 PM, soup...@gmail.com wrote:
Come to think of it, I think an IFL is a single zSeries core. z/VM can
fake it, though that's "overcommitting" the CPU resource.
Despite my NOT looking at the zSeries Principles of Operation I somehow
doubt the zSeries has hyperthreading... as the z/VM OS would be dispatch
threads. On the
On Wednesday, 11/08/2017 at 09:55 GMT, Philipp Kern
wrote:
> I thought unused CP capacity was literally costing OpEx whereas IFLs
> would be CapEx? If it's both OpEx, then yes, agreed. (And I suppose
> originally there might have been deals about giving some IFLs with
>
Hi,
On 08.11.2017 16:42, Alan Altmark wrote:
> On Wednesday, 11/08/2017 at 03:17 GMT, Philipp Kern
> wrote:
>> I idly wonder how actual utilization of IFLs looks like in the field.
>> It's clear to me that CPs run hot because otherwise you're wasting
>> moneys. I.e. if
On Wednesday, 11/08/2017 at 03:17 GMT, Philipp Kern
wrote:
> I idly wonder how actual utilization of IFLs looks like in the field.
> It's clear to me that CPs run hot because otherwise you're wasting
> moneys. I.e. if overcommitment is just a fact of life like it is in the
>
Victor,
As others have mentioned, it depends. I often get asked the same question
and usually they want to compare one intel server/application against an IFL
based solution. But without understanding the load characteristics of the
application it's next to impossible to give a clear answer. Even
and my two cents:
1) don't forget that energy efficiency and hence cost is yet another
factor for comparison.
Scaled up, Z does rather well in this respect.
2) CICS vs RISC: the distiction is so blurred that to even talk in these
terms is difficult. Most Z instructions are implemented in H/W. We
10 matches
Mail list logo