Greetings; (Posted to VMESA-L and VSE-L and LINUX-390)
- - Now in its fifth year! - - Now includes VSE and linux/390!
I have set up a public service web page at
http://www.eskimo.com/~wix/vm/
for posting positions available and wanted for VM, VSE and linux/390.
Please visit the
The article is negative about the iTanium, but my colleagues in high
performance computing tell me that this is a *very* capable
cpu which is
going head to head with the Power4.
I think Intel has stumbled:
http://www.theregister.co.uk/content/61/28894.html
I think I'd have to agree with
POWER4 is clearly better positioned at present.
This is arguable. For the small class of problem where 64-bit system
like an Itanium is really the only option, IA64 does well, although the
PowerPC does about equally well (modulo operating system availability,
which is getting better with
AMD may steal some of Intels play because their 64 bit chip will also be
backwards compatible. Everyone associates Intel with cheap commodity
processing. Itanium does not fit that bill. They will need to do some
marketing to inform corporate America why they should spend more money on a
On Thursday, 02/13/2003 at 08:37 CST, Ryan Ware [EMAIL PROTECTED]
wrote:
I do like Intel's approach of brand new, not backward
compatible. Backwards compatible is another phrase for compromise.
The phenomenal success of S/360, S/370, S/370-XA, S/370-ESA, S/390,
zSeries has been attributed
Oh, I agree with you. I just meant from a sheer engineering standpoint not
being backward compatible usually leads to a better new product.
-Original Message-
From: Alan Altmark [SMTP:[EMAIL PROTECTED]]
Sent: Thursday, February 13, 2003 8:51 AM
To: [EMAIL PROTECTED]
Subject:
Greetings;
I am trying to install lvm2. I downloaded the .deb file but could
not get apt-get to install it no matter what I tried. I finally
got some results with dpkg -i ... but it said there was a dependency
on libc6. I downloaded that. Same difficulties as with lvm2. I seem
to be in a
That depends on what you mean by better. Leaving out backwards compatibility makes
for a cleaner design at first, but then the new design becomes the legacy over time,
and the clean design becomes yet another cluttered design.
My opinion is that Intel is going to have a hard time getting
You are probably right. I was just getting at it (Itanium) being cleaner.
HP has to back Itanium, they've abandoned other chips. IBM is in the
enviable position of being big enough to make chips, buy chips, etc.
And yes overtime as new things age they get crufty and cluttered, but
usually the
Sorry - blasted session IDs get in the way.
Go to http://www.computerweekly.co.uk
Enter Linux storage as the search commands.
Currently the second hit - Want to see the future of storage? Look to Linux
Check out the author's name.
Exactly the same arguments that billg and Motorola have just
Linux to drive Motorola cellphones:
http://www.eweek.com/article2/0,3959,887381,00.asp
Bill worried:
http://212.100.234.54/content/4/29308.html
--
Phil Payne
http://www.isham-research.com
+44 7785 302 803
+49 173 6242039
One of my Linux Guests got hosed last night. I need to know if there are
some error log files I can look at to see what caused the this. I'm running
z/VM 4.3.0 in an IFL and using SuSe 2.4.7. Thanks in advance.
--- Legal Disclaimer: The information contained in this communication may be
another person told me to ust boot.local
to put the programs to start in there
will try it tonight
-Original Message-
From: Post, Mark K [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, February 12, 2003 11:51 AM
To: [EMAIL PROTECTED]
Subject: Re: starting programs at ipl
How did you
Hi, - sorry x-posted to VM, VSE and LINUX/390 Lists -
I4m envolved with a team to study the following scenario:
- migration of VM/VSE (both 2.4.0) to z/VM + VSE 2.6.x (eventually 2.7)
- migration from 9221-421 to MP3000 H50
- migration from RAMAC II (3390-3 format) to IBM 2105 (1.2TB)
-
Look into /var/log In addition you should do cp spool
cons maint start on you LINUX servers.
|+-
|| Ketchens, LeMarr T. |
|| (RyTull) |
|| LeMarr.Ketchens@ryerso|
|
We were hit by the 2 Gig file limit restriction when restoring an
Oracle 9i DB export file of 3.5 Gig into an ReiserFS LVM.
mkreiserfs with operand -v 2 created neither syntax errors
nor any better results. How can we find out if LFS is supported
in our system ? We use S/390 SLES7 GA 2.4.7 kernel
Hi all, i have a problem, we have a z800, the configuration is:
1 cp 80 MIPS
1 IFL
8 Gb storage
3 partitions:
-os/390 2.6
-os/390 2.6
-z/vm 4.3
840 gb (shark)
the cp is dedicated to both os/390, and the ifl to z/vm, 2gb to
both os/390, and 6 gb to z/vm.
Redhat 7.2 as a
This is not the real performance of Linux under VM. This is the real
performance of S/390 processors in general. S/390 processors are _slow_
compared to the newer generations of Intel hardware. Don't use them for CPU
intensive tasks if you can avoid it. Compressing and decompressing data are
Ralph,
That's certainly one way to do it. It's entirely possible it's not the
_right_ way to do it. The reason I asked what distribution you're running,
and whether you installed an RPM or built it from source, etc. is related to
this. Most (if not all) Linux distributions generate RPMs that
On Thursday 13 February 2003 16:09, Alex wrote:
-the hmc indicates that the ifl is at 99% utilization.
Our HMC on the z900 always shows 99% for the IFL, even if it's not
doing anything substantial. Running a CPU hot all the time was an
old performance enhancing trick I remember from the 4381
Cool, question. I don't have anything to offer except what you're seeing is
normal if you base normal on your machine and my machine as they both
seem to act similarily. So, I'll be watching the responses with as much
interest as yourself.
Leland
In the (relatively) dark ages of VM/HPO and the 4381 MP models, they
called this Active Wait. I'm sure somebody got a nice oxymoron award
for that one... ;)
-dan.
On Thu, 13 Feb 2003, Joe Poole wrote:
On Thursday 13 February 2003 16:09, Alex wrote:
-the hmc indicates that the ifl is at 99%
Typically on the HMC we exclude key 3 from the display. That gives a better
representation of what VM is actually doing.
On Thursday 13 February 2003 03:59 pm, you wrote:
In the (relatively) dark ages of VM/HPO and the 4381 MP models, they
called this Active Wait. I'm sure somebody got a nice
Thanks, Dan. Memory fades..
JP
On Thursday 13 February 2003 16:59, you wrote:
In the (relatively) dark ages of VM/HPO and the 4381 MP models, they
called this Active Wait. I'm sure somebody got a nice oxymoron
award for that one... ;)
-dan.
On Thu, 13 Feb 2003, Joe Poole wrote:
On
I recall that the UTS developers at Amdahl noticed something similar. I think that VM
gives hot machines higher priority. If I recall correctly they made the VM aware
version of UTS run hot just to improve response times.
-Original Message-
From: Joe Poole [mailto:[EMAIL PROTECTED]]
Greetings;
I am trying to install lvm2. I downloaded the .deb file but could
not get apt-get to install it no matter what I tried. I finally
got some results with dpkg -i ... but it said there was a dependency
on libc6. I downloaded that. Same difficulties as with lvm2. I seem
to be in a
You are probably right. I was just getting at it (Itanium) being cleaner.
HP has to back Itanium, they've abandoned other chips. IBM is in the
enviable position of being big enough to make chips, buy chips, etc.
Surely HP is too. As I understand it, Itanium is very much an HP design too.
If you want proper help, supply the information people ask for. I think
boot.local won't do anything.
another person told me to ust boot.local
to put the programs to start in there
will try it tonight
-Original Message-
From: Post, Mark K [mailto:[EMAIL PROTECTED]]
Sent:
We were hit by the 2 Gig file limit restriction when restoring an
What imposed the 2 Gig file limit?
Oracle 9i DB export file of 3.5 Gig into an ReiserFS LVM.
mkreiserfs with operand -v 2 created neither syntax errors
nor any better results. How can we find out if LFS is supported
in our
On Thu, 13 Feb 2003, Fargusson.Alan wrote:
I recall that the UTS developers at Amdahl noticed something similar.
I think that VM gives hot machines higher priority. If I recall
correctly they made the VM aware version of UTS run hot just to improve
response times.
Depends on what you mean
and the time went from 1m3.6s to 1m2.039s in the better case, the
people from ibm (they are here yet) can give me an answer about the
poor performance (i consider that its a poor performance, because a
intel piii 128Mb RAM make the tar in about 28s), so i really dont know
if this is the real
Linux has been 64-bit for longer than OS, by a big margin. Some time, somone
will produce cheapish 64-bit machines for the desktop, and they will win the war
as everyone +dog buys one.
Few people need 64-bit addressing. Number of particles in the known universe comes
to mind.
But quite a
OK, all you Linux folks
Here is one I have been racking my pea brain for half the day today
I have SLES7 installed and am at run levlel 5 Now I want to be able to use
my RedHat system to X-window into the S390 Suse(Sles7)
that I just installed last night.
Anybody have an idea on
On Thu, 2003-02-13 at 23:04, Phil Payne wrote:
Linux has been 64-bit for longer than OS, by a big margin. Some time, somone
will produce cheapish 64-bit machines for the desktop, and they will win the war
as everyone +dog buys one.
Few people need 64-bit addressing. Number of particles in
Ken,
If you're talking about accessing a graphical desktop, such as KDE, then the
Distributions Redbook chapter 6, section 6.6 on page 113 talks about how to
do that. If all you want to do is use an xterm session, then you should be
able to ssh to the SLES7 system, and start an xterm. You may
I had 2.2.16 installed and decided to try the SLES7 beta (downloaded
from mirror.mcs.anl.gov). I downloaded the whole file structure, created
a copy on an AIX box to do an ftp install from and created a CD to boot
from. Everything went fine until the package install when the install
could'nt find
Steve,
Try having a directory structure that looks like this:
/path/to/the/files/cd1
/path/to/the/files/cd2
When asked for the FTP server directory, specify /path/to/the/files, and it
will try a few combinations of things to tack on the end of that (like CD1,
cd1, etc.), and it will find the
Mainframes do I/O exceptionally well, but when it comes to compute bound tasks, they
do very poorly. If you think about a tar operation, the compression is a fairly
compute-intensive operation.
We're running a 9672-R56 w/ one IFL. During our initial trial, we found the IFL to be
about the same
We recently had a user with lots of standalone unix experience use ssh to send about 8
GB of data (the entire contents of a 3390-9 minidisk mount-point) from one linux image
to another on the same IFL.
Consider this: He compressed his data to a tar file. He encrypted the data. He sent
the
installed apache from source we are at suse version
-Original Message-
From: John Summerfield [mailto:[EMAIL PROTECTED]]
Sent: Thursday, February 13, 2003 4:39 PM
To: [EMAIL PROTECTED]
Subject: Re: starting programs at ipl
If you want proper help, supply the information people ask for.
One of my Linux Guests got hosed last night. I need to know if there are
some error log files I can look at to see what caused the this. I'm running
z/VM 4.3.0 in an IFL and using SuSe 2.4.7. Thanks in advance.
That's awfully vague. Define hosed.
And before you come back, peruse the logs
Hello from Gregg C Levine
My hacker's dictionary defines hosed, as badly broken, or not
working at all, or if its an operating system, it refuses to boot. In
this case, the person in question can't IPL the blasted thing. Does
that clarify the issue?
---
Gregg C Levine [EMAIL
Wolfe, Gordon W wrote:
He used ssh because it was more secure and compressed it to save bandwidth.
This even though all the communications took place internally to CP
(VCTCA links) and never even hit an ethernet cable. This would be
REAL hard
to put a sniffer on!
Don't forget
43 matches
Mail list logo