On Wed, 2006-05-17 at 12:07 -0500, Tom Duerbusch wrote:
#4 Context switching. Seems like when you switch from one task to
another on some processors, all of cache is invalidated. Doesn't seem
to be so with the mainframe. I assume there is a point, where we
thrash cache, but it seems like
John Campbell wrote:
[GRAIN TYPE=SALT MODE=Stand Up Philosopher]
A long time ago in a website far, far away, I once sent a note to the
fellow talking about mainframes.
In fact, some years after I wrote the original author a note and then
forgot about it, I discovered that my reply had somehow
Mike Lovins wrote:
Can some give me the directions on the proper steps to change the system
name and the IP address of my test Linux system that is going to replace
a current system I am shutting down. I need to use the same system name
and IP address because of the TSM application that I have.
On Iau, 2006-05-18 at 10:03 +0200, Martin Schwidefsky wrote:
On x86 it is the translation-lookaside-buffers (TLBs) which get flushed
each time the control register 1 is loaded. Switching between threads is
[%cr3 not 1 but thats by the way]
fine because the use the same translation table.
Hello list,
Has anyone used heartbeat across LPARs to set up an HA solution? We are
close but it's not quite working.
We set up two ITDS6 LDAPs that are peer to peer master servers so their
databases stay in sync. We set up heartbeat to give the two servers a
single Virtual IP Address (VIPA).
On Thursday, 05/18/2006 at 10:03 ZE2, Martin Schwidefsky
[EMAIL PROTECTED] wrote:
The cache is a different story. Mainframes have the advantage of a
shared level 2 cache compared to x86. If a process migrates from one
processor to another, the cache lines of the process just have to be
loaded
John Summerfied writes:
...
I have not heard of any failed Intel or AMD CPUs in a very long time.
Accompanying system components such as RAM, disks, NICs, yes, but not
the CPU itself. The systems I use are built to be cheap; one can have
greater reliability for a greater price. I imagine that
Ghosts? That was what we called non-terminal driven BASIC users in
Throroughbred BASIC, useful for handling some background jobs.
Beyond maintaining the interpreter's ability to drive them I never quite
understood all that they were good for, but, then, I was working on the
internals of the
The shared L2 reduces the penalty for those situations when you can't avoid
dispatching on a new engine. That is when the system is very busy. This
is one of the reasons for the difference in utilization. As the machine
gets busier other machines are forced into L2-L2 or remote L3-localL1
Yes tagging works, but you will find that the system z holds a lot more
translations in a two tiered TLB and has tagging as well. Thus the System
z does not have to retranslate as often.
Joe Temple
Distinguished Engineer
Sr. Certified IT Specialist
[EMAIL PROTECTED]
845-435-6301 295/6301
I've always wanted to ask this question and you seem like someone who would
know. First and foremost I am not a computer engineer!
Here's the situation.
When the PC first came out I taught myself some basic assembler for the x86
chip. One thing that I remember to move 255 bytes from one
On May 18, 2006, at 3:13 AM, John Summerfied wrote:
I have not heard of any failed Intel or AMD CPUs in a very long time.
Accompanying system components such as RAM, disks, NICs, yes, but not
the CPU itself.
I'll be happy to give you a couple of mine.
OK, so it was thermal failure caused by
I'm sure this is a completely different class of processor, but I recently
purchased an AMD Athlon 3200+ XP. After chaning the memory chips, power
supply, heatsink/fan, video card, and removing all other cards I finally
decided the chip was faulty. It would attempt to boot up and after the
Many times when it works like that for a little bit and then freezes up,
you're experiencing a thermal issue. There could be any number of
reasons for that and a couple of the common ones it seems you looked at
(fans, heastink).
And yes, the Athlon is not in the same class as the Opteron. :)
Yes, that was my first thought too. I bought a heavy duty Thermaltake
heatsink/fan that was rated for the processor. I made sure I had the thermal
grease applied correctly and in sufficent amount. I went into the BIOS setup
and monitored the CPU temperature for over an hour and it was well
Has anyone used heartbeat across LPARs to set up an HA solution? We
are
close but it's not quite working.
We hit this a while back on the real OSA hardware -- it's a side effect
of the OSA trying to be too smart for its own good. There is OSA
microcode service that addresses it for the real
Don't forget to check your power supply under load, either. The PS is the
primary point of entry for those dustbunnies, which can really hose a
power supply. I've had just *one* bad AMD chip in the past (not bad for as
many as I've bought) which actually behaved once I cut its multiplier down
to
Hi, all.
Given recent events (the new who's been reading our list), I'm
hesitant on posting all information here, but I still have a problem
with this transfer performance on our z/VM Linux guests.
For background, we are a government shop with a pretty small z800 box.
We have a single IFL engine
On 5/18/06, Peggy Andrews [EMAIL PROTECTED] wrote:
Given recent events (the new who's been reading our list), I'm
hesitant on posting all information here, but I still have a problem
with this transfer performance on our z/VM Linux guests.
I am pretty sure we answered you here and came to the
On May 18, 2006, at 9:58 AM, James Tison wrote:
Don't forget to check your power supply under load, either. The PS
is the
primary point of entry for those dustbunnies, which can really hose a
power supply. I've had just *one* bad AMD chip in the past (not bad
for as
many as I've bought) which
Hi, Rob.
I didn't miss the posts, but I'm not sure what to do with that
information to prove that we can run at the same performance level.
So essentially, are you saying that we can't get the same performance
on a directory to directory copy without investing in 2 IFL engines?
Thanks.
Peggy
Yes, I remember the Tandem Non Stop like it was yesterday. Oh, wait, it was
yesterday. It is called the HP Non Stop now. We have a fairly large one near
our IBM z900. They have something similar to z/OS Unix, which I find
interesting.
-Original Message-
From: Linux on 390 Port
On 5/18/06, Peggy Andrews [EMAIL PROTECTED] wrote:
I didn't miss the posts, but I'm not sure what to do with that
information to prove that we can run at the same performance level.
It is very hard to demonstrate the suitability of the platform with a
few synthetic test cases. The measured
I suspect you are thinking of the prefix instructions used in the block move.
Or maybe you didn't notice that you can use the prefix instruction. In any
case the block move on Intel is about the same as a move character on the
mainframe.
In the 8086 days it took several clock cycles to do
The AMD chips don't have redundant logic, so when something fails you get the
wrong answer.
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of
Richard Pinion
Sent: Thursday, May 18, 2006 7:19 AM
To: LINUX-390@VM.MARIST.EDU
Subject: Re: Who's been reading our
[EMAIL PROTECTED] 05/18/06 2:13 AM
The term mainframe has been around for some years.
Remember Tandem Non Stop? Tandem's big selling-point was
reliability
(fault tolerance) through redundancy. As best I can recall I went to
a
presentation in the early 80s, when we had Amdahl 43xxs and IBM's
On 5/18/06, Fargusson.Alan [EMAIL PROTECTED] wrote:
In the 8086 days it took several clock cycles to do anything. The most
annoying to me was the 60 cycle divide instruction. These days they are much
faster, some instructions being only one clock.
I still have trouble to measure
On Iau, 2006-05-18 at 09:51 -0400, Joseph Temple wrote:
Yes tagging works, but you will find that the system z holds a lot more
translations in a two tiered TLB and has tagging as well. Thus the System
z does not have to retranslate as often.
How many tags does the Z have in the TLBs ?
Here is a description of the z990 TLB from the IBM Journal of Research and
Development Volume 48 number 3/4 2004 article, The IBM eServer z990
Microprocessor. Sorry for the graphics... The TLB1 has 512 entries.
The TLB2 has a unique two part design that buffers a large number
translation
Remember Tandem Non Stop? Tandem's big selling-point was reliability
(fault tolerance) through redundancy. As best I can recall I went to a
presentation in the early 80s, when we had Amdahl 43xxs and IBM's
mainframes were 3080 series. I think we had some dodgy disk drives, but
otherwise
FWIW, I used the Xerox Sigma 9 in school. The Univeristy of Southern
Mississippi had the Sigma 9 in 1980 when I graduated. From my perpective as a
student in CS it was a good machine. Do you remember RAD SAT
[EMAIL PROTECTED] 5/18/2006 3:26:32 PM
Remember Tandem Non Stop? Tandem's big
Rob won't come right out and say it, so I will. Contact Velocity
Software and see what kind of terms they'll give your for evaluating
their software. That will help you diagnose just what is going on with
your situation.
Absent of that, some questions that come to mind are:
Are the source and
I have two zLINUX LPARS. One is SLESS 9 and the other is CentOS 4.
We are migrating from an ESS 800 to DS8100. All unit addresses for
DASD will change. Network address will not. Is there any documentation
on how to migrate to the new devices? Each zLINUX system has two 3390
Mod 9 volumes.
Will you have concurrent access to both sets of disk at some point?
That will make the migration almost a nit.
Mark Post
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Henderson.Louis
Sent: Thursday, May 18, 2006 4:30 PM
To: LINUX-390@VM.MARIST.EDU
Yes, I have physical access now. But permission starting next weekend
5/27. Concurrent access for about two weeks. What is the nit?
Louis E. Henderson
SharedServices - Mainframe - x5392
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] On Behalf Of
Post, Mark
I used a Sigma 9 in the late 80s. I ran into the 16 bit address limit a number
of times. Other that that it was OK.
-Original Message-
From: Linux on 390 Port [mailto:[EMAIL PROTECTED] Behalf Of
Richard Pinion
Sent: Thursday, May 18, 2006 12:30 PM
To: LINUX-390@VM.MARIST.EDU
Subject:
Fargusson.Alan wrote:
I suspect you are thinking of the prefix instructions used in the block move.
Or maybe you didn't notice that you can use the prefix instruction. In any
case the block move on Intel is about the same as a move character on the
mainframe.
In the 8086 days it took
Richard Pinion wrote:
Richard
You may be posting in HTML. Better if you don't - shorter emails kinder
to dialup users, less likely to be tagged as spam (some people, would
you believe it? refuse HTML email altogether). It might explain why
Mozilla's quoting is a bit off.
I've always wanted to
Richard Pinion wrote:
I'm sure this is a completely different class of processor, but I recently
purchased an AMD Athlon 3200+ XP. After chaning the memory chips, power
supply, heatsink/fan, video card, and removing all other cards I finally
decided the chip was faulty. It would attempt to
Peggy Andrews wrote:
Hi, all.
Given recent events (the new who's been reading our list),
Nobody will get a balanced view of anything reading a mailing list such
as this. Of course, it's centred on problems we have: a list devoted to
extolling the virtues of its primary topic will quickly
Rob van der Heij wrote:
On 5/18/06, Peggy Andrews [EMAIL PROTECTED] wrote:
I didn't miss the posts, but I'm not sure what to do with that
information to prove that we can run at the same performance level.
It is very hard to demonstrate the suitability of the platform with a
few synthetic
The SLES system will be very easy, the CentOS one a little bit more
work because of the way Red Hat uses file system labels and puts those
in /etc/fstab.
For the sake of illustration, let's assume that your device numbers and
names for the SLES system are:
0300/dev/dasda1 mounted on /
Post, Mark K wrote:
Now, for the CentOS system, the same methodology will be used, except
for the file system labels. Once the two disks have been copied, you'll
need to change the file system labels on the new disks. Otherwise, when
you try to boot either the old system or the new one,
I don't know if you have access to a z/OS system, I did. I had to move
my Linux DASD to a new DASD box. I was able to take the Linux systems
down for a few hours and move the DASD from a z/OS system. I did full
volume copies using a software package, but I think the same thing could
be done
44 matches
Mail list logo