.
Then the question is if you’re still buying the Mellanox NICs, why not go the
whole hog, particularly as you may grow outside of a single switch.
Matt.
> On 27 Feb 2025, at 19:19, Brice Goglin wrote:
>
> Hello
>
> While meeting vendors to buy our next cluster, w
sult.
I’m about to rebuild it and the ice driver, I’m just confused at how it went
from PSM3 complaining about an interface to nothing at all. LDD shows all the
correct libraries are being found, lsmod shows the correct modules in the
kernel.
Matt.
t I’ve seen, adding OpenStack is the exact opposite of that.
Matt.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
https://beowulf.org/cgi-bin/mailman/listinfo/beowulf
NVIDIA has a version of HPL floating around, but will only supply it under NDA,
and you’re definitely not allowed to share the version you have. Not that that
doesn’t happen of course, but NVIDIA would definitely prefer you didn’t.
Matt.
—
Matt Wallis
ma...@madmonks.org
> On 15 Aug 2
Fortran assumes no aliasing. C/C++ assumes possible aliasing.
Makes a world of difference when optimising. C++ can catch up with Fortran
with extensive template trickery which was first widely demonstrated with
Blitz++ which benchmarked relative to Fortran.
On Fri, 30 Nov 2018 at 02:22, Stu Midgle
and it does eliminate the whole question of
cleaning up scratch space, when the job is over, the scratch space is gone.
I have another system based on BeeGFS coming online in the second half
of the year, that I can't talk about right now, but I will be looking
for new adjectives
res all over the world. Rather than hook them all up
to some common file server, just build a system to package up the data and the
application and send it where ever can process it the quickest. Send it 3 times
to make sure it gets done, then pull back
at it only uses the swap when to do otherwise
would be to crash the machine.
This might be a good article to read http://www.linuxjournal.com/article/10678
Matt.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change you
ugh I may be missing something and it has nanosecond support
hidden in there somewhere which is what I was asking about.
Kind regards,
--Matt.
On 10 August 2014 14:19, Lux, Jim (337C) wrote:
> Nanoseconds?
> You need something like a GPS (run of the mill is good to 10-50 ns) or
> IE
Can it do better than millisecond time stamps yet? My network stuff needs
nanoseconds or better...
--Matt.
On 7 August 2014 12:48, Walid wrote:
> Hi,
>
> I am interested to see if any one is already using ELK for HPC related
> logs. i would like to know more about what metrics,
On 2 Jul 2014, at 4:19 pm, Jonathan Aquilina wrote:
> How would the same arguments apply if you are just dealing with dns
> servers web servers databases etc.
If you're just dealing with standard services then you're not really doing high
performance clustering, but fault tolerance.
In a typi
rupted environment variables.
Eeep. I did mean 2012, not 2010, I can read, honest.
Good to hear the list is still active I guess.
Matt.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or
gh, has anyone not been put off enough by lua to use
Lmod?
I'm thinking of setting up a build environment with it because Environment
Modules
seems to have been slow to update to newer versions of TCL, and they don't
appear to
be activel
ation.
And then when it comes time to debug why the application doesn't work,
commercial compiler suites often come with very good debugging tools.
Matt.
--
Matt Wallis
ma...@madmonks.org
___
Beowulf mailing list, Beowulf@beowulf.org sponsored
o with the dam walls but I'm not sure if people really need
much more than what ptp can already do.
--Matt.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
r a few financial market uses, such a device
is not really useful for bewoulf or HPC as the MISD model doesn't seem
to be of much practical use unless you can get something cute for
virtually free to suit occaisonal use like those mentioned earlier
such as the PAPERS or integrated Blue Gene.
would be huge...
Nope, not a dumb hub, even dumber ;-) No collisions just a tree of
optical couplers frantically splitting the photon streams. The only
real trick, albeit pretty minor, is ensuring the signal integrity is
within budget and suitable for non-thinking plug and play.
Regards,
uch small amounts of nanos at
extra cost doesn't quite seem to fit a beowult style budget anyhow.
Just for HFTs and financial exchanges I guess...
Regards,
--Matt.
>
> -Kevin
>
>
>> _
>> www.zeptonics.com
>> ___
ot of ports though 1024 ports seems a bit too big.
Any ideas on the list about use of low latency broadcast for specific
applications in HPC? Are there codes that would benefit?
Regards,
Matt.
_
www.zeptonics.com
___
Beowulf mailing li
On Tue, 16 Jun 2009, Mike Davis wrote:
In my experience, Sysadmins don't want beer or luxurious offices they want
the tools that they need, proper managerial support, and respect.
And I'm sure not getting any of that in my current job. And the pay scale
is much lower.
-- Mat
duplication in install trees, particularly
when you just want to install the latest. I've managed to get some
massive savings with NIM on AIX and some lesser but stll very good savings
with CentOS by building parallel trees and hard linking the files.
-- Matt
It's not what I know tha
On Mon, 4 May 2009, Prentice Bisbal wrote:
In an ideal world, the code/Makefiles/configure scripts would written
such that any standards-compliant compiler would work, but that's not
always the case.
That's the nice thing about standards, there are so many of them to choose
from.
On Wed, 15 Apr 2009, Per Jessen wrote:
Matt Lawrence wrote:
Sure it is. The only issue you may run into is that the partition
containing /boot can't be on software RAID, grub & lilo don't know how
to deal with RAID partitions.
That's not correct - lilo has no such pro
ed in this cluster, so many
decisions were made in too much of a hurry.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
n only two drives available.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsu
into is that the partition
containing /boot can't be on software RAID, grub & lilo don't know how to
deal with RAID partitions. My solution has been to put /boot on Compact
Flash that is plugged into the motherboard IDE connector via a $5 adapter
card.
-- Matt
It's not
t a Dell fanboy, but I will have to admit that my notebook has been
a very good value for the money I spent.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
various support organizations.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscrib
re problems and can tell them we're running what he believes is RHEL...
Except for the storage group, who have no clue at all about Linux and very
bad attitudes, I haven't had any real troubles getting support from Dell.
Maybe it's a difference in approach.
-- Matt
It's not w
CentOS and the centosplus repository includes kernels that
support xfs. Still wind up installing to ext3, but big data filesystems
can easily be xfs.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beow
rFS,
which is a pity, I really like their architecture and the developers.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode o
it's a bit larger than they need or
would grow into.
The Linksys 48 port managed switches seem to be very unreliable. I
recommend avoiding them.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
B
keyboard on the remote machine).
3. contains smartmontools, parted, fdisk, etc., but not any desktop pieces.
I've been doing this with CentOS for quite a while. It works very well.
-- Matt
It's not what I know that counts.
It's what I ca
On Fri, 24 Oct 2008, Robert G. Brown wrote:
Awe, surely I could be "Best HPC software product or technology"
(rgb-bot, after all:-). Or maybe "Best HPC cluster solution" -- "Ask
rgb..." ;-)
Maybe, but I would still rather vote for cymk instead..
-- Matt
It
n for preferring it over any other.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beow
t, the guy was
downright snotty about it.
Also, I recommend staying away from the MD3000 storage unit. It has
issues such as not being able to present LUNs larger thn 2GB, not
supporting RAID6 and requiring a GUI interface to be able to install the
command line tools.
-- Matt
It's
twork that is seperate
for management and BMC traffic. Since our systems have two ethernet ports
on the motherboard, the only cost was the switches and cables. For the
price, I think it is very worthwhile.
-- Matt
It's not what I know that counts.
It's what I can rem
a lot of unecessary
stuff that goes on, like converting everything to and from XML for each
message.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To c
fair amount of room cooling may be needed as well.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe)
ugh to let me do those runs in the timeframe I require otherwise.
Please don't! If you tuned that thing on, it would probably blow the
breaker for my office as well.
I'll bet we could put something together that is faster for less than half
the price.
-- Matt
It's not what
rrier?
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
suite and why they should implement a thorough
automatic build process is beyond the skill of many of us mere software
developers.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beow
DE->CF adapter and putting /boot on
the Compact Flash. I picked up several adapters for around $5 each a
while back and I use CF cards that are smaller than I feel like using in
my camera (nothing smaller than 4GB in my camera case).
-- Matt
It's not what I know that
doing changes by hand. I was figuring to find the
appropriate sections of the CentOS spec file and use them. As always,
other suggestions are welcome.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
ed looking at how to make that work, but the 9K lines of spec
file for the CentOS kernel rpm are rather daunting.
Since y'all have obviously been dealing with these sorts of problems
longer than I have, I could really use some help here.
-- Matt
It's not what I know that counts.
It&#x
en someone would shut down a workstation that was serving a
filesystem and everything would crash. Just like dominos.
Like I said, a sordid history.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
I trust enough that if he warns me of something, I
make a real effort to doublecheck if it is currently a problem. It
doesn't mean he is always right, just that I think the research effort is
a really good idea.
-- Matt
It's not what I know that counts.
It's
ve minutes today going "where the heck is mkfs.xfs and the man
pages?".
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (dig
, I would rate him much higher than any of the folks I work with as
far as being a sysadmin.
So, any good info on kernel configuration when I go to build a new rpm?
There are a huge number of options and you have obviously gone through
them much more recently than I have.
-- Matt
It's not w
. He
cited lots of fragmentation problems that routinely locked up his systems.
I am willing to be convinced otherwise, but he is a very sharp fellow.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailin
ly it is of use.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
etake when you
say "I'm waiting for the power strip to boot up".
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (dige
g is a pack rat.
No, he is merely satisfying his genetic tendancy toward archivism.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (d
erhaps you should consider getting time on someone else's cluster. For
something that only requires three nodes, there should be quite a number
of places to run.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
__
On Thu, 26 Jun 2008, Joe Landman wrote:
Not just "my cluster is bigger than your cluster" but "my cluster is
WYY bigger than your cluster".
TACC's PDU field is bigger than my cluster.
-- Matt
It's not what I know that counts.
It's
stems just to say
it has been done. Should be pretty easy using Hercules/390.
Since my boss reads this list, I'm not sure if he will be amused or
appalled at the idea. probably both.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
__
ch { |n| addr += sprintf("%02.2X",
n.to_i) }
p addr
begin
base = "/tftpboot/pxelinux.cfg/#{addr}"
File.rename(base, base + ".done")
rescue SystemCallError
puts "Error occurred renaming #{base}"
end
session.close
end
Yeah, lots of stuff
ol that was (and it did take a few
minutes to eject all those disks afterwards), but it took less time
than rebooting each node, for sure, and I had a desk full of spare
floppy disks for two or three years after that.
Matt
- --
812.855.7318 voice
Research Technologies - High-Performance Syst
gets egg on their face.
Yes even "modern" BIOSes are bad but localboot first is a sensible
default.
I will have to disagree. Changing the BIOS settings in a single head node
is preferable to having to connect to 126 compute nodes and change their
BIOS settings.
-- Matt
It's
student worker is on vacation sailing in the Carribean, so I get the
abuse
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mo
CF->SATA converters and boot from 4 GB flash
cards. We have built working 4GB flash images of Ubuntu, Suse, Openfiler,
and one of the RHELs.
Actually, I use CF->IDE adapters. About $5 each. Incredibly useful to
get Linux booted far enough to understand LVM & RAID.
-- Matt
It&#x
would be appreciated.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailma
delaminate and the adhesive fails when they are used on cables.
Also, I would be quite happy if the list moderator were to take me off
moderation. Hint, hint
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
__
Also, make sure you have a really good HVAC person on site
as well, it would be really bad to cook all of the equipment.
Please post some detailed results, I want to hear what worked really well
and what had problems (and those solutions).
-- Matt
It's not what I know that counts.
It&#x
On Sat, 3 May 2008, Alex Younts wrote:
The machine will be running Redhat Enterprise 4.
That's getting to be a bit dated, but still very well supported and
extremely stable.
-- Matt
It's not what I know that counts.
It's what I can remember
start at providing performance information back
to the users. Particularly since I can implement this in an hour or so
and have it completely automated.
-- Matt
It's not what I know that counts.
It's what I can remember in time to use.
_
tools, wrapping it up in
scripts to run automatically would be really easy.
My goal is to make it really easy to provide useful information to users
without them having to constantly ask me.
Thoughts?
-- Matt
It's not what I know that counts.
It's what I can remember in t
I is linked dynamically with my
> parallel application, MPI_Get_count return a wrong value.
I'm guessing your machine is suffering from version hell and your
LD_LIBRARY_PATH environment variable doesn't match your Makefile.
We use modules and someone else figures
putation (and also those of you who hire undergrads), I'd be grateful for
>your thoughts. One specific question I have is what text covers scientific
>programming and touches on MPI using the C language.
Can't help with texts. I still use Luc Chamberland's Fortran 90
Reference G
only show up with large processor
counts (making one window per process impossible), and the third is
that, being remote, the fancy gui based debuggers run like a ruptured
turtle over the internet.
We can also run our code in serial, so that's where we do most of our
d
isn't have 10 slower
'datawulf' nodes better than 1 at the same speed (with them added up)
for data access? surely the optimal speed is a processor that can
handle the intensive IO streaming and a kernel + a bit of overhead for
good luck.
another idea i have floating - with lower speed cpu's
also - this uses intellipower, intellipark and intelliseek, which are
basic things... i doubt these will be of any use in data centers where
drives are under constant strain. in an archive systems these, with good
power managment and resorce managment, are good things.
also - are the figures av
hi -
at home our server we call green. it's not low power. but it is 100%
recycled from someone who didn't want it... it's a P2 mobo with a P3
XU02 SLOT it consumes about 28W while under normal net usage. it has
2 HDD's a 20GB 'low' consumption laptop drive - adapted for the 3.5"
bay, and
room... power, UPS, mains, EPO, cooling, node
turn off and WOL... all in one program (part of the job scheduler i
should think). this would really be a step into the custom and not
commodity hardware tho.
but yea, DC electronics is gross on a la
nd simplest, but
doesn't use the machine very efficiently. I think it'll only get worse
with more cores.
I'd be interested in your experience and what you find out.
Matt
___
Beowulf mailing list, Beowulf@beowulf.org
To change yo
rectly meaning no or little analogue electronics used.
personally...
HD is a marketing CON to get nieve people to buy 'HD' products when they
would be better buying a computer monitor with a higher resolution,
colour depth, and refresh rate. although a 42" 'HD' widescreen
a 32 bit system (maybe the last bit is lost because of signed
arithmetic?), I'm guessing the ia-64 compilers can have much larger
arrays.
Fortran90 doesn't say anything about array limits.
Matt
___
Beowulf mailing list, Beowulf@beowulf.org
To c
nds of
problems? Does anyone know how much money they're spending on
maintaining old code as opposed to creating new? I'd like to extend
the tools I've built and find customers that either want these tools
or would pay me to help them, but I first want to find some potential
customers t
ing for
but, I'm not sure how to say this, it's basically just a place to
announce conferences anymore.
Thanks,
Matt
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://ww
h but approximates as "No-thing". It does not
> compute. Or in true computerese, NaN.
--
matt.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
thanks for all the replies first of all,
i don't know the exact scyld distribution. However, i am running mpich 1.2.5.
When i run my program (stripped down to a mere MPI_INIT(...) call) and test it
with valgrind i get something like :
==21799== Use of uninitialised value of size 8
==21799==
Hi,
i hope this is the right mailing list to post to...
Anyway, i was wondering if i could get some advice/direction on how to debug
my mpich program. I am running on a scyld configuration. What i am trying
right now is the following:
mpirun -dbg=gdb -nolocal -np 32 exec
which starts the debu
a game
> console would be wasted as a webserver.
indeed it would. 95% of any machine that is not designed to be a
webserver is lost... a true webserver would be as you say just embedded
chips, storage and net. little need for anything else. the admin can be
ionic head nodes with KillerNic
and playstations for compute nodes...
--
matt.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
fails others will respond anyway.
>
> Apparently they use more reliable hardware for things like the
advertising service
--
matt.
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
htt
owing the resource manager to completely handle job spawning has
provided better post-job cleanup, and more complete job statistics
(cpu-time, mostly) for us.
Do you not have to deal with these sorts of issues? If not, lay some
wisdom on me; I could use it.
Matt
--
Matt Allen| Sys
eed. if you
can get hold of these ITX boards you can run w.e. software you like on
them as they have 10/100 base networking, and you could probably get
gagabit aswell.
i would say that you dont need to have these as speced up as normal
nodes, if their only file serving then a GHz processor
ou boot from net and then go to local storage, does the RAM used
to boot stay off limits to the OS and programs or can it be used for
other purposes once booting is done?
--
matt jones,
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
ine fans to route cold air to the cases that are producing the most
heat. but that would require some enginearing to do effectivly and some
programming of the 25 temp sensors (for each case), 25 power relays, the
AC unit controls, fans and integrating with a job manager for the
switching on and o
ard rack
> mount'.
>
> --
> Leif Nixon -Systems expert
>
> National Supercomputer Centre- Linkoping University
> -
that the admin can work from.
--
matt jones,
this e-mail has been virus scanned,
sent from a workstation called: matts, IP address: 10.64.63.17, Subnet
Mask: 255.255.192.1
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription
in this for a 100Mbps hub you would get 10.24 seconds per node.
so assuming that you dont have any more than one node failing every 11
seconds.. a 100Mbps hub would do adequately.
--
matt jones,
this e-mail has been virus scanned,
sent from a workstation
ewulfs were high cost finatually to sart
out-in... more of high in needs of time to sit, configure and play
around with and get used to.
--
matt jones,
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest mode or unsubsc
depends on how you use it.
--
matt jones,
this e-mail has been virus scanned,
sent from a workstation called: matts, IP address: 10.64.63.17, Subnet
Mask: 255.255.192.1
___
Beowulf mailing list, Beowulf@beowulf.org
To change your subscription (digest
processes. This
requires some user education, though, and there will almost always be
cases where it's not going to work, such as users who require commercial
MPI distributions.
Matt
Larry Felton Johnson wrote:
> My apologies in advance if this is a FAQ, but I'm reading through t
want the motherboards,
memory and hard drives for the project...
and i need to find a proper use for the cluster nce it's going, any
ideas anyone???
matt jones, 15, YR. 11 student at a normal comprehensive school.
Bill Harman wrote:
> One of the things that you can do to get things
e those 1 to 10 of us might get the help we need
to actually build and maintain a cluster. which i suppose would benefit
the other students as they could run models through it.
matt jones, 15, YR. 11 student at a normal comprehensive school.
ps* can anyone advise me to a guide for starting a
97 matches
Mail list logo
