On 16/11/2012 2:16 AM, Anne Lynn Wheeler wrote:
IBM has z196 benchmark with peak of 2m IOPS with 104 FICON channels,
14 storage subsystems, and 14 system assist processors. It mentions
that the 14 SAPs are capable of peak 2.2m SSCH/sec running at 100% cpu
busy, but recommends SAPs run at 70%
re:
http://www.garlic.com/~lynn/2012o.html#25 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
attached is item from today I did in linkedin mainframe ... work I had
done for channel extender in 1980 then also start to show up for
fibre-channel in the late 80s. Then nearly
enhancement to FICON
http://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012m.html#5 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012m.html#11 Blades versus z
/BIPS)
past posts in this thread:
http://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#57 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#59
On Tue, 11 Sep 2012 13:33:26 +0800, Timothy Sipples1 sipp...@sg.ibm.com wrote:
Almost everyone here would. These are mainframes, with PR/SM and LPARs
proven to Common Criteria EAL5+ certification standards. Regardless of the
operating system(s) running in particular LPARs.
I think I know what an
On 11/09/2012 5:41 PM, Bruno Sugliani wrote:
Everyone WOULD ... but how many do ? On this list for example? just curious
There is a difference between theory and real life.
My 20 years in IBM level 2 support center gave me some scars.
Several customers have been very sorry to use a single CP
In a7ur48l9vhqq4g7p4akhsbhejdmea3i...@4ax.com, on 09/10/2012
at 11:31 AM, Clark Morris cfmpub...@ns.sympatico.ca said:
Does i-o on system z take less cpu and memory resource than i-o on
Intel servers?,
Probably less CPU, not less memory.
--
Shmuel (Seymour J.) Metz, SysProg and JOAT
. of Tennessee
http://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012m.html#4 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
As I've referenced numeruous times (ever since early 90s fibre
l...@garlic.com (Anne Lynn Wheeler) writes:
oops, late 80s is 25yrs ago ... not 35yrs ... finger slip.
re:
http://www.garlic.com/~lynn/2012m.html#2 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off
Thank you Mark for this very honest answer
Those of us who tried, and those of us who got the bills from both world, know
better than have a definite answer
It is quite funny to see that people very often speak of a single engine IFL
machine and compare it to a several core x86 installation
W dniu 2012-09-10 13:29, Bruno Sugliani pisze:
[...]
I do not believe anyone
here would run production on a single engine box or run Dev and Test
on that single engine machine. [...]
Well, I ran *two* production LPARs plus dev LPAR plus sandbox LPAR on
*single* CPU. Later, we've had two
Clark Morris is entitled to have and to express his [not at all
disinterested] views.
His question
begin snippet
Does i-o on system z take less cpu and memory resource than i-o on
Intel servers?, p servers?
end snippet/
is, however, challenging to this entitlement.
The answer to this question
On 9/10/2012 10:14 AM, Anne Lynn Wheeler wrote:
jwgli...@gmail.com (John Gilmore) writes:
The answer to this question is yes. z/Architecture channel-based I/O
is very different from that used by, for example, Intel servers. In
particular it uses many fewer cp[u] cycles, and its permits many
On 9/10/2012 at 10:31 AM, Clark Morris cfmpub...@ns.sympatico.ca wrote:
On 4 Sep 2012 08:12:48 -0700, in bit.listserv.ibm-main you wrote:
On 9/3/2012 at 05:05 PM, Richard Hintz rjhi...@gmail.com wrote:
Do you have something or can point me to something that shows comparative
metrics for
Lynn's most recent response is unsatisfactory, in substance evasive.
Let us for the sake of the argument stipulate, though this is not
usually the case, that some non-mainframe server can perform some
single I/O operation faster than some mainframe.
It turns out that this stipulation does not
different from what CKD
emulation has to do.
re:
http://www.garlic.com/~lynn/2012m.html#2 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012m.html#3 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
but major server apps rdbms
oops, late 80s is 25yrs ago ... not 35yrs ... finger slip.
in the 80s, it was recognized that the half-duplex channel paradigm (not
just ibm mainframe) ... introduced a lot of end-to-end latency overhead
chatter. there were several serial, asynchronous efforts launched in the
late 80s ... all
On 10 Sep 2012 08:11:00 -0700, in bit.listserv.ibm-main you wrote:
Clark Morris is entitled to have and to express his [not at all
disinterested] views.
His question
begin snippet
Does i-o on system z take less cpu and memory resource than i-o on
Intel servers?, p servers?
end snippet/
is,
Bruno Sugliani writes:
I do not believe anyone here would run production on a single
engine box or run Dev and Test on that single engine machine.
Almost everyone here would. These are mainframes, with PR/SM and LPARs
proven to Common Criteria EAL5+ certification standards. Regardless of the
software (Websphere, Java, C/C++)?
re:
http://www.garlic.com/~lynn/2012l.html#87 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#88 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html
On 9/7/2012 at 06:56 PM, Mike Schwab mike.a.sch...@gmail.com wrote:
10 or 20 Linux servers consolidated onto 1 x86-64 blade server.
300 Linux servers consolidated onto 1 zIFL.
Now that looks reasonable. A full speed z processor is still 15 to 30
times faster than Virtual x86-64.
Um, no,
/2012l.html#81 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#87 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#88 Blades versus z was Re: Turn Off
Another Light - Univ
W dniu 2012-09-06 23:16, Mike Schwab pisze:
On Thu, Sep 6, 2012 at 3:55 PM, R.S. mike.a.sch...@gmail.com CAN'T YOU SNIP.MY
ADDRESS?.com.pl wrote:
I meant it's hard to justfiy the choice: to buy IFL (plus rest of mainframe)
or x64 servers. I meant Linux on IFL is *much* more expensive than on
On Thu, 6 Sep 2012 16:16:03 -0500, Mike Schwab mike.a.sch...@gmail.com wrote:
You can buy and power 300 servers cheaper than one IFL? IBM says that
many servers would cost 10 times as much to power and cool.
As Mandy Rice-Davies[1] famously said: they would, wouldn't they?
Roger Bowler
Hercules
On 9/7/2012 at 04:40 AM, Roger Bowler ibm-m...@snacons.com wrote:
On Thu, 6 Sep 2012 16:16:03 -0500, Mike Schwab mike.a.sch...@gmail.com
wrote:
You can buy and power 300 servers cheaper than one IFL? IBM says that
many servers would cost 10 times as much to power and cool.
As Mandy
/~lynn/2012l.html#81 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
note part of the enormous growth in servers was that they were viewed as
nearly zero cost item ... so people costs to manage multiple
applications was greater than just having server per application. some
re:
http://www.garlic.com/~lynn/2012l.html#81 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#87 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
BladeCenter blade servers
http://www-03.ibm.com/systems/bladecenter
To make long story short: IFL can be cost effective because of ISV
licenses. Linux is (in theory) free, but Oracle and other are not.
Assuming free software only it's very hard to justify spendings on IFL.
BTW: Licensing models do change. How can one be sure that Oracle will
keep the model
On 9/6/2012 at 03:16 AM, R.S. r.skoru...@bremultibank.com.pl wrote:
To make long story short: IFL can be cost effective because of ISV
licenses. Linux is (in theory) free, but Oracle and other are not.
Let's not keep propagating the myth that Linux is free as in no cost. The
classic
W dniu 2012-09-06 18:46, Mark Post pisze:
On 9/6/2012 at 03:16 AM, R.S.
r.skoru...@snip.it!.COM.PL wrote:
To make long story short: IFL can be cost effective because of ISV
licenses. Linux is (in theory) free, but Oracle and other are not.
Let's not keep propagating the myth that Linux is
On Thu, Sep 6, 2012 at 3:55 PM, R.S. r.skoru...@bremultibank.com.pl wrote:
I meant it's hard to justfiy the choice: to buy IFL (plus rest of mainframe)
or x64 servers. I meant Linux on IFL is *much* more expensive than on x64
servers. Things like power, cooling, floor space, staffing won't
it, but the software licenses could do
it.
BTW: I did not mention RAS, but not everyone will pay for good
RollRoyce, some of us choose Toyota.
re:
http://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#57 Blades versus
Maybe they will start using price per MIPS model?
Then that'd be *even more* financially favorable to the most intensely
virtualized, highest average utilization cores -- the chips with the MIPS
which are most efficiently used, with the fewest idle MIPS per year. Which
would be zEnterprise cores.
For purposes of licensing its software running on Linux, IBM assigns a
metric called PVUs (Processor Value Units) to each type of processor
core. Most IBM software products for Linux are licensed according to the
number of PVUs at a price per PVU. Likewise, annual software subscription
and support
re:
http://www.garlic.com/~lynn/2012l.html#56 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#57 Blades versus z was Re: Turn Off
Another Light - Univ. of Tennessee
http://www.garlic.com/~lynn/2012l.html#59 Blades versus z was Re: Turn
]
On Behalf Of McKown, John
Sent: Tuesday, September 04, 2012 10:28 AM
To: IBM-MAIN@LISTSERV.UA.EDU
Subject: Re: Blades versus z was Re: Turn Off Another Light - Univ. of
Tennessee
Doesn't mean much to us, really. But the post below says that the
government of Quebec installed Oracle on zLinux
From: Mark Post mp...@suse.com Sun, 2 Sep 2012 15:46:35 -0600
Then you haven't looked deeply enough. Software licensing for middleware
significantly favors running on Linux on System z. Many other costs such has
power, cooling, floor space, people, inventory tracking, networking hardware,
On 1 Sep 2012 08:04:00 -0700, in bit.listserv.ibm-main you wrote:
r.skoru...@bremultibank.com.pl (R.S.) writes:
No, with *one* blade cabinet of Dell+Windows. HW cost comparable to
spare HMC and two OSA cards.
as mentioned before:
max. configured z196 with 80 processors is rated at 50BIPs and
mike.a.sch...@gmail.com (Mike Schwab) writes:
Put a Hercules emulator and z/OS on that blade, 50 z/OS MIPS per
hyperthread, so 100 MIPS per core, 1600 MIPS per blade (per
TurboHercules). Perhaps $5,000 per blade? Some blades do have 4
sockets.
re:
http://www.garlic.com/~lynn/2012l.html#51
I was talking about x86-64 bit cores. 16 cores running Hercules and
z/OS would get you about 3 z196 cores of 1,600 MIPS. I don't know
what chip or benchmark you are talking about because that is about 300
times faster that x86-64 chips.
On Sun, Sep 2, 2012 at 3:26 PM, Anne Lynn Wheeler
cfmpub...@ns.sympatico.ca (Clark Morris) writes:
How much work can that z196 do compared with the 4829/hr Amazon cloud
you mentioned? Given the great disparity between costs per
instruction execution, on reading these posts it would seem that
getting to a secure, fault tolerant operating
On 2 Sep 2012 10:31:31 -0700, in bit.listserv.ibm-main you wrote:
On 9/2/2012 at 12:47 PM, Clark Morris cfmpub...@ns.sympatico.ca wrote:
Given the great disparity between costs per
instruction execution, on reading these posts it would seem that
getting to a secure, fault tolerant operating
On 9/2/2012 at 04:51 PM, Clark Morris cfmpub...@ns.sympatico.ca wrote:
The major reason for staying on a mainframe (z, Unisys A, Unisys 2200
follow-on) is the horrendous cost of migration.
That is far from the only reason to stay on the mainframe.
Noting the number of
smaller entities that
43 matches
Mail list logo
