[Beowulf] Working for DUG, new thead

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Bill Abbott

2018-06-13 17:53:24 UTC

linux, mostly

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.beo

Prentice Bisbal

2018-06-13 18:04:14 UTC

I would think that "low-level" means "close to the hardware"
(troubleshooting drivers, low-level hardware debugging, etc. ) , and
that junior-level would mean "basic Linux system admin skills", but
they're in Australia, so they might use these terms differently.

Prentice

Post by Bill Abbott
linux, mostly

> Stuart Midgley works for DUG? They are currently
> recruiting for an HPC manager in London... Interesting...
Recruitment at DUG wants to call me about Low Level HPC. I have at
least until 6pm.
I am excited but also terrified. My background is C and now
JavaScript, mostly online course work and telnet MUDs.
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C17cc89690dbb4ce5ae9a08d5d15038ec%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645064845177853&sdata=RkclgLTrmOepLzAURyHcfJPnSpADzhMVU4kJz6iI3%2FA%3D&reserved=0

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe)

Bill Abbott

2018-06-13 18:07:53 UTC

s/linux/cabling/g

Post by Prentice Bisbal
I would think that "low-level" means "close to the hardware"
(troubleshooting drivers, low-level hardware debugging, etc. ) , and
that junior-level would mean "basic Linux system admin skills", but
they're in Australia, so they might use these terms differently.
Prentice

Post by Bill Abbott
linux, mostly

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C5ba251a20aa24880fd2708d5d15820d2%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645098799494306&sdata=tltKQcflRPcjtTi5x2VZ0NISIQTmx4O09zZ7OEcJJyM%3D&reserved=0

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.be

John Hearns via Beowulf

2018-06-13 18:26:50 UTC

Jonathan, if you have taken the interest to join this list then there is no
need to be terrified.
I have learned that people who are enthusiastic are quite rare.
Also interviews are a two way street - this is your opportunity to find out
about the role.
Hopefully you will be enthused about it and wish to join the company.

Post by Bill Abbott
s/linux/cabling/g

Post by Prentice Bisbal
I would think that "low-level" means "close to the hardware"
(troubleshooting drivers, low-level hardware debugging, etc. ) , and that
junior-level would mean "basic Linux system admin skills", but they're in
Australia, so they might use these terms differently.
Prentice

Post by Bill Abbott
linux, mostly

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%
2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%
7C01%7Cbabbott%40rutgers.edu%7C5ba251a20aa24880fd2708d5d1582
0d2%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C63664509879
9494306&sdata=tltKQcflRPcjtTi5x2VZ0NISIQTmx4O09zZ7OEcJJyM%3D&reserved=0

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%
2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%
7C01%7Cbabbott%40rutgers.edu%7C5ba251a20aa24880fd2708d5d1582
0d2%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C63664509879
9494306&sdata=tltKQcflRPcjtTi5x2VZ0NISIQTmx4O09zZ7OEcJJyM%3D&reserved=0

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

Andrew Latham

2018-06-13 18:27:23 UTC

Such a broad topic. I would assume things like DHCP, TFTP, Networking, PXE
and IPMI which come to mind. Troubleshooting tools, configuration
management, version control, monitoring, issue tracking and many other
processes are at play. I would love to hear any interview questions which
could cover this array of topics sanely.

Suggested question: Tell me the best hardware horror story?

On Wed, Jun 13, 2018 at 12:08 PM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
- Andrew "lathama" Latham -

Bill Abbott

2018-06-13 18:37:13 UTC

One of my standard interview questions is to say ok, you start on Monday
and you're placed in charge of a web/db server, tell me what you do your
first week.

What I want to hear is security, backups, log checking, monitoring,
performance, functionality, etc., but most of all I want to know how
they think and if they can come up with a coherent plan.

Another version is "A user says the cluster is slow. What do you do?"

Bill

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at
least until 6pm.
I am excited but also terrified. My background is C and now
JavaScript, mostly online course work and telnet MUDs.
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905753024&sdata=sr%2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy0%3D&reserved=0>
--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905763036&sdata=S%2BJM%2BswwtGQFNk8VTDEbWQQnXhpoTQx5CYJR5Wxyuu4%3D&reserved=0

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/

Andrew Latham

2018-06-13 18:46:44 UTC

Bill's question is good and I have heard it many times. Practice answering
that.

Related examples:

Tell me how you would setup and install 100 new servers?
How do you describe to onsite support staff (smart hands) to move a network
cable to another port on the same system?
At 2am you are trouble shooting an issue from home and you lose connection
to the node. What do you check first?

Post by Bill Abbott
One of my standard interview questions is to say ok, you start on Monday
and you're placed in charge of a web/db server, tell me what you do your
first week.
What I want to hear is security, backups, log checking, monitoring,
performance, functionality, etc., but most of all I want to know how
they think and if they can come up with a coherent plan.
Another version is "A user says the cluster is slow. What do you do?"
Bill

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905753024&sdata=sr%2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy0%3D&reserved=0

--
- Andrew "lathama" Latham -

John Hearns via Beowulf

2018-06-13 18:49:49 UTC

Bill's question re. "the cluster is slow"is fantastic.
That covers people skills in addition to technical skills.

ps. best of luck. You never know who you might end up working with ;-)

Post by Andrew Latham
Bill's question is good and I have heard it many times. Practice answering
that.
Tell me how you would setup and install 100 new servers?
How do you describe to onsite support staff (smart hands) to move a
network cable to another port on the same system?
At 2am you are trouble shooting an issue from home and you lose connection
to the node. What do you check first?

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%
7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%
7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905753024&sdata=sr%
2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy0%3D&reserved=0>

--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

Douglas Eadline

2018-06-13 19:46:07 UTC

Add to that "Please identify the situations that justify using a LART?"

https://en.wiktionary.org/wiki/LART

--
Doug

Stuart Midgley works for DUG?Â They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at
least until 6pm.
I am excited but also terrified. My background is C and now
JavaScript, mostly online course work and telnet MUDs.
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905753024&sdata=sr%2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy0%3D&reserved=0>
--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905763036&sdata=S%2BJM%2BswwtGQFNk8VTDEbWQQnXhpoTQx5CYJR5Wxyuu4%3D&reserved=0

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
--
MailScanner: Clean

--
Doug
--
MailScanner: Clean

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf

Jonathan Engwall

2018-06-13 21:59:29 UTC

Thanks for the great feedback.

LART? What is that?

I am doing a quick start, and just by chance I have been watching an MIT
lecture on SSE.

This is from a YouTuber who calls himself Biscuit. Quick and beautiful with
pragmas and the Mandelbrot's Set.
Crossing fingers now...

Post by Douglas Eadline
Add to that "Please identify the situations that justify using a LART?"
https://en.wiktionary.org/wiki/LART
--
Doug

Stuart Midgley works for DUG?Ã They are currently
recruiting for an HPC manager in London... Interesting...

Post by Bill Abbott

Post by Andrew Latham
--
- Andrew "lathama" Latham -
_______________________________________________

Computing

Post by Bill Abbott

Post by Andrew Latham
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=

http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%
7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%
7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905763036&sdata=S%
2BJM%2BswwtGQFNk8VTDEbWQQnXhpoTQx5CYJR5Wxyuu4%3D&reserved=0

Post by Bill Abbott
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
--
MailScanner: Clean

--
Doug
--
MailScanner: Clean
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

Prentice Bisbal

2018-06-19 16:01:58 UTC

Despite the source (just kidding, Bill!) I'm going to have to support
this line of questioning.

HPC (and general IT), consists of systems with many different layers,
and it takes the correct personality type with the good analytical
skills to be able to trouble shoot things effectively. I really don't
care about questioning an interviewee about technical minutiae in an
interview. Most of that can be easily googled these days. The real
value/skill is in coming up with a plan of attack: knowing where to
start troubleshooting, what to google, how to correctly interpret those
google results, and then now to apply what you find online to fix your
problem.

If a sys admin position involves shell prorgramming/scripting, knowing
the details of a specific programming language or processor are
secondary, but thinking like a programmer is skill not everyone has or
can develop.Just last week a wrote a Lua script without knowing a thing
about Lua. I looked at the example scripts, and then googled to fill in
the blanks. I think most SysAdmins do stuff like that on a regular basis.

If a position involves writing optimized code, I wouldn't apply this
logic - getting really good performance can require knowing the minutiae
of a specific language, but most HPC sys admins don't do that level of
programming.

I also look for people who have hobbies that involve understanding how
things go together and work: working on bicycles or cars, electronics,
etc. If you can understand how other things go together or work,
computers, are not that big of a jump.

Prentice

Post by Bill Abbott
One of my standard interview questions is to say ok, you start on
Monday and you're placed in charge of a web/db server, tell me what
you do your first week.
What I want to hear is security, backups, log checking, monitoring,
performance, functionality, etc., but most of all I want to know how
they think and if they can come up with a coherent plan.
Another version is "A user says the cluster is slow. What do you do?"
Bill

Post by Andrew Latham
Such a broad topic. I would assume things like DHCP, TFTP,
Networking, PXE and IPMI which come to mind. Troubleshooting tools,
configuration management, version control, monitoring, issue tracking
and many other processes are at play. I would love to hear any
interview questions which could cover this array of topics sanely.
Suggested question: Tell me the best hardware horror story?
On Wed, Jun 13, 2018 at 12:08 PM Jonathan Engwall
     > Stuart Midgley works for DUG? They are currently
     > recruiting for an HPC manager in London... Interesting...
    Recruitment at DUG wants to call me about Low Level HPC. I have at
    least until 6pm.
    I am excited but also terrified. My background is C and now
    JavaScript, mostly online course work and telnet MUDs.
    Any suggestions are very much needed.
    What must a "low level HPC" know on day 1???
    Jonathan Engwall
    _______________________________________________
    To change your subscription (digest mode or unsubscribe) visit
    http://www.beowulf.org/mailman/listinfo/beowulf
<https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905753024&sdata=sr%2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy0%3D&reserved=0>
--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b69b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645112905763036&sdata=S%2BJM%2BswwtGQFNk8VTDEbWQQnXhpoTQx5CYJR5Wxyuu4%3D&reserved=0

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) v

John Hearns via Beowulf

2018-06-19 18:38:29 UTC

If a sys admin position involves shell prorgramming/scripting, knowing the

details of a specific programming language or processor are secondary, but
thinking like a >programmer is skill not everyone has or can develop.Just
last week a wrote a Lua script without knowing a thing about Lua. I looked
at the example scripts, and then >googled to fill in the blanks. I think
most SysAdmins do stuff like that on a regular basis.

Nooo.. I would never do that.
At one stage I had a whole shelf of Oreilly books at work. Still have them
at home, but a bunch went to Textbooks for Africa recently.
My local library gladly accepted some rather heavy tomes on parallel
programming. Goodness knows what the residents of SE London make of them. I
rather hope that some youngster is inspired by one of the books.

Despite the source (just kidding, Bill!) I'm going to have to support this
line of questioning.
HPC (and general IT), consists of systems with many different layers, and
it takes the correct personality type with the good analytical skills to be
able to trouble shoot things effectively. I really don't care about
questioning an interviewee about technical minutiae in an interview. Most
of that can be easily googled these days. The real value/skill is in coming
up with a plan of attack: knowing where to start troubleshooting, what to
google, how to correctly interpret those google results, and then now to
apply what you find online to fix your problem.
If a sys admin position involves shell prorgramming/scripting, knowing the
details of a specific programming language or processor are secondary, but
thinking like a programmer is skill not everyone has or can develop.Just
last week a wrote a Lua script without knowing a thing about Lua. I looked
at the example scripts, and then googled to fill in the blanks. I think
most SysAdmins do stuff like that on a regular basis.
If a position involves writing optimized code, I wouldn't apply this logic
- getting really good performance can require knowing the minutiae of a
specific language, but most HPC sys admins don't do that level of
programming.
I also look for people who have hobbies that involve understanding how
things go together and work: working on bicycles or cars, electronics, etc.
If you can understand how other things go together or work, computers, are
not that big of a jump.
Prentice

Post by Bill Abbott
One of my standard interview questions is to say ok, you start on Monday
and you're placed in charge of a web/db server, tell me what you do your
first week.
What I want to hear is security, backups, log checking, monitoring,
performance, functionality, etc., but most of all I want to know how they
think and if they can come up with a coherent plan.
Another version is "A user says the cluster is slow. What do you do?"
Bill

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at
least until 6pm.
I am excited but also terrified. My background is C and now
JavaScript, mostly online course work and telnet MUDs.
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
<https://na01.safelinks.protection.outlook.com/?url=http%3A%
2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%
7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b6
9b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C63664511290
5753024&sdata=sr%2FGxwT6uV5hOD029A6Pgbc7%2BGrciWMP7jq77Ap1sy
0%3D&reserved=0>
--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%
2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%
7C01%7Cbabbott%40rutgers.edu%7C3dccc87b23f6445e544e08d5d15b6
9b1%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C63664511290
5763036&sdata=S%2BJM%2BswwtGQFNk8VTDEbWQQnXhpoTQx5CYJR5Wxyuu
4%3D&reserved=0
_______________________________________________

To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

Lux, Jim (337K)

2018-06-19 20:57:12 UTC

From: Beowulf <beowulf-***@beowulf.org> on behalf of "***@beowulf.org" <***@beowulf.org>
Reply-To: John Hearns <***@googlemail.com>
Date: Tuesday, June 19, 2018 at 11:40 AM
To: "***@beowulf.org" <***@beowulf.org>
Subject: Re: [Beowulf] Working for DUG, new thead

If a sys admin position involves shell prorgramming/scripting, knowing the details of a specific programming language or processor are secondary, but thinking like a >programmer is skill not everyone has or can develop.Just last week a wrote a Lua script without knowing a thing about Lua. I looked at the example scripts, and then >googled to fill in the blanks. I think most SysAdmins do stuff like that on a regular basis.

Nooo.. I would never do that.
At one stage I had a whole shelf of Oreilly books at work. Still have them at home, but a bunch went to Textbooks for Africa recently.
My local library gladly accepted some rather heavy tomes on parallel programming. Goodness knows what the residents of SE London make of them. I rather hope that some youngster is inspired by one of the books.

== some poor kid is going to read Beckerâs book, find archives of this list, find RGBs notes, and go hunting for DEC ethernet cards.. May the computing gods have mercy on his soul..

That said, google is my friend when writing in a new language thatâs similar to an old language (Ruby, Iâm looking at you)

Jonathan Engwall

2018-06-13 19:28:01 UTC

Last month I knocked the CMOS off of the motherboard of my 2950. I ordered
a new one, waited, wrestled the old one free of the hidden riser (riser A).
And then reversed the process with the new...
Then I realized I had knocked the CMOS off of that board too! But then I
noticed the faded green caused by a GPU blowing down on the board,
loossening the glue in the CMOS's general vicinity: it was the board I had
just removed.
The new one was sitting beside me the whole time!
And it now posts and boots just fine.

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
- Andrew "lathama" Latham -
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

Lux, Jim (337K)

2018-06-14 03:46:37 UTC

Naahh. We should work to scare the whatevers out of him..

Theyâre going to ask about how you designed the adaptive equalizer in your last 10GEthernet ASIC. What specific problems did you have modifying the NetGear FA310TX device driver to accommodate this change. Oh, and by the way, we need a polynomial time, scalable algorithm to solve the generalized knapsack problem. And, what *are* your opinions about temperature profiles in the reflow soldering process, vis a vis the formation of tin whiskers.

A look at the list archives from 2000 should do nicely.

From: Beowulf <beowulf-***@beowulf.org> on behalf of Andrew Latham <***@gmail.com>
Date: Wednesday, June 13, 2018 at 11:28 AM
Cc: "***@beowulf.org" <***@beowulf.org>
Subject: Re: [Beowulf] Working for DUG, new thead

Such a broad topic. I would assume things like DHCP, TFTP, Networking, PXE and IPMI which come to mind. Troubleshooting tools, configuration management, version control, monitoring, issue tracking and many other processes are at play. I would love to hear any interview questions which could cover this array of topics sanely.

Suggested question: Tell me the best hardware horror story?

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
- Andrew "lathama" Latham -

Fred Youhanaie

2018-06-13 18:49:43 UTC

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

This is one of the links in Stu's "HPC positions" email from June 11:

https://www.dug.com/careers/positions_vacant/low_level_hpc_developer

Could this be what you're looking for?

Cheers,
Fred
_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit

Bill Abbott

2018-06-13 19:01:36 UTC

I thought it was for sysadmin, not developer. Disregard most of what I
said.

Bill

https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.dug.com%2Fcareers%2Fpositions_vacant%2Flow_level_hpc_developer&data=02%7C01%7Cbabbott%40rutgers.edu%7C63fcf2cdf2a14198e8c208d5d15e7e5b%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C1%7C636645126139750738&sdata=3BEsI0Hj5GkbSJLDI7jGH7I12mf3fSZ3AU2j9pi36GM%3D&reserved=0
Could this be what you're looking for?
Cheers,
Fred
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cbabbott%40rutgers.edu%7C63fcf2cdf2a14198e8c208d5d15e7e5b%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C1%7C636645126139750738&sdata=y%2FXISnWl%2B9cbhOS7Y3pkwjbHyC6W5NW7scuMWC4EO%2FA%3D&reserved=0

_______________________________________________
Beowulf mailing list, ***@beowulf.org sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.beo

Ryan Novosielski

2018-06-13 19:23:21 UTC

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

As others have said, it does depend on what âlow levelâ means â eg. entry level or low level in the hardware sense, as Prentice said. In general, though, Iâd say that you should have some working knowledge of what a job scheduler is, be it SLURM or SGE ob PBS or whatever, the major components, the concepts behind it. Itâs almost a given that theyâre using a job scheduler. You should also probably at least have a concept of how MPI works, and what sort of interconnects and storage one tends to encounter in HPC.

They also may have some information on their website about their systems, depending who they are, which might help you target your review.

--
____
|| \\UTGERS, |---------------------------*O*---------------------------
||_// the State | Ryan Novosielski - ***@rutgers.edu
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'

Jonathan Engwall

2018-06-13 19:32:03 UTC

Yes. Very good idea, it is a mining company in Ausralia.

On Jun 13, 2018, at 1:07 PM, Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at least

until 6pm.

I am excited but also terrified. My background is C and now JavaScript,

mostly online course work and telnet MUDs.

Any suggestions are very much needed.
What must a "low level HPC" know on day 1???

As others have said, it does depend on what âlow levelâ means â eg. entry
level or low level in the hardware sense, as Prentice said. In general,
though, Iâd say that you should have some working knowledge of what a job
scheduler is, be it SLURM or SGE ob PBS or whatever, the major components,
the concepts behind it. Itâs almost a given that theyâre using a job
scheduler. You should also probably at least have a concept of how MPI
works, and what sort of interconnects and storage one tends to encounter in
HPC.
They also may have some information on their website about their systems,
depending who they are, which might help you target your review.
--
____
|| \\UTGERS, |---------------------------*
O*---------------------------
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'

Stu Midgley

2018-06-14 01:13:13 UTC

OR... we could answer your questions directly...

Stu (who has worked at DUG for 12 years, built and designed the systems...
was CTO and now Systems Architect)

On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall <

Post by Jonathan Engwall
Yes. Very good idea, it is a mining company in Ausralia.

On Jun 13, 2018, at 1:07 PM, Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at

least until 6pm.

I am excited but also terrified. My background is C and now JavaScript,

mostly online course work and telnet MUDs.

Any suggestions are very much needed.
What must a "low level HPC" know on day 1???

As others have said, it does depend on what âlow levelâ means â eg. entry
level or low level in the hardware sense, as Prentice said. In general,
though, Iâd say that you should have some working knowledge of what a job
scheduler is, be it SLURM or SGE ob PBS or whatever, the major components,
the concepts behind it. Itâs almost a given that theyâre using a job
scheduler. You should also probably at least have a concept of how MPI
works, and what sort of interconnects and storage one tends to encounter in
HPC.
They also may have some information on their website about their systems,
depending who they are, which might help you target your review.
--
____
|| \\UTGERS,
|---------------------------*O*---------------------------
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'
_______________________________________________

To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

--
Dr Stuart Midgley
***@gmail.com

Stu Midgley

2018-06-14 01:39:27 UTC

seismic processing - oil'n'gas not hard rock.

On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall <

Post by Jonathan Engwall
Yes. Very good idea, it is a mining company in Ausralia.

--
Dr Stuart Midgley
***@gmail.com

Jonathan Engwall

2018-06-14 02:02:31 UTC

64 bit words have me wondering. I have been interested in monolithic lately
too.
Well, I hope to get a call...and did. I think it went well. She did not
find my pip3 story exciting.
And thanks again ever one.

Post by Stu Midgley
seismic processing - oil'n'gas not hard rock.
On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall <

Post by Jonathan Engwall
Yes. Very good idea, it is a mining company in Ausralia.

--
Dr Stuart Midgley

Joe Landman

2018-06-14 02:24:30 UTC

Post by Stu Midgley
seismic processing - oil'n'gas not hard rock.

But ... you work with heavy metal (supercomputers that is) :D

Good to see you active here, BTW!

Post by Stu Midgley
On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall
Yes. Very good idea, it is a mining company in Ausralia.
--
Dr Stuart Midgley
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf

--
Joe Landman
e: ***@gmail.com
t: @hpcjoe
c: +1 734 612 4615
w: https://scalability.org
g: https://github.com/joelandman
l: https://www.linkedin.com/in/joelandman

Stu Midgley

2018-06-14 02:34:04 UTC

seismic processing is a MASSIVE user of compute resources. We run single
processing steps that can take months on a 10PFlop machine...

Post by Stu Midgley
seismic processing - oil'n'gas not hard rock.
But ... you work with heavy metal (supercomputers that is) :D
Good to see you active here, BTW!
On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall <

Post by Jonathan Engwall
Yes. Very good idea, it is a mining company in Ausralia.

--
Dr Stuart Midgley
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
--
Joe Landman
c: +1 734 612 4615
w: https://scalability.org
g: https://github.com/joelandman
l: https://www.linkedin.com/in/joelandman
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

--
Dr Stuart Midgley
***@gmail.com

Lux, Jim (337K)

2018-06-14 03:53:25 UTC

And you drill through hard rock to get to the hydrocarbons.

But Stu is right â itâs one of the more interesting massively parallel (but not necessarily embarrassingly) problems, because the propagation medium is anisotropic. By comparison SAR (Synthetic Aperture Radar), tomography,and finite element electromagnetics processing are a doddle. Some mechanical FEM codes are tricky, and non-linear flow (shock waves, etc.) are complex.

Everyone who deals with these sorts of problems is all about efficiency and âgood approximationsâ â you typically start with a linearized approximation, and then iterate with something to get the nonlinear solution â all of this takes lots oâ cycles.

From: Beowulf <beowulf-***@beowulf.org> on behalf of Stu Midgley <***@gmail.com>
Reply-To: "***@gmail.com" <***@gmail.com>
Date: Wednesday, June 13, 2018 at 7:35 PM
To: "***@gmail.com" <***@gmail.com>
Cc: "***@beowulf.org" <***@beowulf.org>
Subject: Re: [Beowulf] Working for DUG, new thead

seismic processing is a MASSIVE user of compute resources. We run single processing steps that can take months on a 10PFlop machine...

On Thu, Jun 14, 2018 at 10:25 AM Joe Landman <***@gmail.com<mailto:***@gmail.com>> wrote:

On 6/13/18 9:39 PM, Stu Midgley wrote:
seismic processing - oil'n'gas not hard rock.

But ... you work with heavy metal (supercomputers that is) :D

Good to see you active here, BTW!

On Thu, Jun 14, 2018 at 3:32 AM Jonathan Engwall <***@gmail.com<mailto:***@gmail.com>> wrote:

Yes. Very good idea, it is a mining company in Ausralia.
--
Dr Stuart Midgley
***@gmail.com<mailto:***@gmail.com>

_______________________________________________

Beowulf mailing list, ***@beowulf.org<mailto:***@beowulf.org> sponsored by Penguin Computing

To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
--
Joe Landman

e: ***@gmail.com<mailto:***@gmail.com>

t: @hpcjoe

c: +1 734 612 4615

w: https://scalability.org

g: https://github.com/joelandman

l: https://www.linkedin.com/in/joelandman
_______________________________________________
Beowulf mailing list, ***@beowulf.org<mailto:***@beowulf.org> sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf

--
Dr Stuart Midgley
***@gmail.com<mailto:***@gmail.com>

Stu Midgley

2018-06-14 01:14:07 UTC

We aren't that scary...

On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
Dr Stuart Midgley
***@gmail.com

Stu Midgley

2018-06-14 01:17:05 UTC

low level HPC means... lots of things. BUT we are a huge Xeon Phi shop and
need low-level programmers ie. avx512, careful cache/memory management (NOT
openmp/compiler vectorisation etc).

On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
Dr Stuart Midgley
***@gmail.com

Joe Landman

2018-06-14 02:32:17 UTC

I'm curious about your next gen plans, given Phi's roadmap.

low level HPC means... lots of things.Â BUT we are a huge Xeon Phi
shop and need low-level programmers ie. avx512, careful cache/memory
management (NOT openmp/compiler vectorisation etc).

I played around with avx512 in my rzf code.
https://github.com/joelandman/rzf/blob/master/avx2/rzf_avx512.cÂ . Never
really spent a great deal of time on it, other than noting that using
avx512 seemed to downclock the core a bit on Skylake.

Which dev/toolchain are you using for Phi?Â I set up the MPSS bit for a
customer, and it was pretty bad (2.6.32 kernel, etc.).Â Flaky control
plane, and a painful host->coprocessor interface.Â Did you develop your
own?Â Definitely curious.

On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall

Stuart Midgley works for DUG?Â They are currently
recruiting for an HPC manager in London... Interesting...

--
Joe Landman
e: ***@gmail.com
t: @hpcjoe
c: +1 734 612 4615
w: https://scalability.org
g: https://github.com/joelandman
l: https://www.linkedin.com/in/joelandman

Stu Midgley

2018-06-14 04:02:47 UTC

Phi is dead... Long live phi...

By which I mean, while the Phi as a chip is going away, its concepts live
on. Massive number of cores, large vectorisation and high speed memory
(and fucking high heat load - we do ~350W/socket). So, while the product
code will disappear, phi lives on.

For KNC I did a lot of customisation to MPSS to get it to work... and we
haven't been able to shift from one of the very early version. We love the
KNC... we get 8 in 2RU which is awesome density (1.1kW/RU)

For KNL its just x86 with a big vectorisation unit (700W/RU).

In both cases you have to be very very careful how you manage memory.

Post by Joe Landman
I'm curious about your next gen plans, given Phi's roadmap.
low level HPC means... lots of things. BUT we are a huge Xeon Phi shop
and need low-level programmers ie. avx512, careful cache/memory management
(NOT openmp/compiler vectorisation etc).
I played around with avx512 in my rzf code.
https://github.com/joelandman/rzf/blob/master/avx2/rzf_avx512.c . Never
really spent a great deal of time on it, other than noting that using
avx512 seemed to downclock the core a bit on Skylake.
Which dev/toolchain are you using for Phi? I set up the MPSS bit for a
customer, and it was pretty bad (2.6.32 kernel, etc.). Flaky control
plane, and a painful host->coprocessor interface. Did you develop your
own? Definitely curious.
On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
Dr Stuart Midgley
***@gmail.com

Ryan Novosielski

2018-06-19 21:48:19 UTC

We bought KNC a long time ago and keep meaning to get them to a place where they can be used and just havenât. Do you mount filesystems from them? We have GPFS storage, primarily, and would have to re-export it via NFS I suppose if we want the cards to use that storage. Iâve seen complaints about the stability of that setup. I didnât try to build the GPFS portability layer for Phi â not sure whether to think it would or wouldnât work (I guess Iâd be inclined to doubt it).

Post by Stu Midgley
Phi is dead... Long live phi...
By which I mean, while the Phi as a chip is going away, its concepts live on. Massive number of cores, large vectorisation and high speed memory (and fucking high heat load - we do ~350W/socket). So, while the product code will disappear, phi lives on.
For KNC I did a lot of customisation to MPSS to get it to work... and we haven't been able to shift from one of the very early version. We love the KNC... we get 8 in 2RU which is awesome density (1.1kW/RU)
For KNL its just x86 with a big vectorisation unit (700W/RU).
In both cases you have to be very very careful how you manage memory.
I'm curious about your next gen plans, given Phi's roadmap.

low level HPC means... lots of things. BUT we are a huge Xeon Phi shop and need low-level programmers ie. avx512, careful cache/memory management (NOT openmp/compiler vectorisation etc).

I played around with avx512 in my rzf code. https://github.com/joelandman/rzf/blob/master/avx2/rzf_avx512.c . Never really spent a great deal of time on it, other than noting that using avx512 seemed to downclock the core a bit on Skylake.
Which dev/toolchain are you using for Phi? I set up the MPSS bit for a customer, and it was pretty bad (2.6.32 kernel, etc.). Flaky control plane, and a painful host->coprocessor interface. Did you develop your own? Definitely curious.

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

--
Joe Landman
c: +1 734 612 4615
https://scalability.org
https://github.com/joelandman
https://www.linkedin.com/in/joelandman
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
--
Dr Stuart Midgley
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cnovosirj%40rutgers.edu%7C89d9a1fe40cd40448a5708d5d1abc4d9%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645458049748846&sdata=dEUacidlV69%2FM8NEdObFNmSOsOObZpPAF4NlfI7joTw%3D&reserved=0

--
____
|| \\UTGERS, |---------------------------*O*---------------------------
||_// the State | Ryan Novosielski - ***@rutgers.edu
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'

Stu Midgley

2018-06-20 02:49:36 UTC

we initially used them as standalone systems (ie. rsh a code onto them and
run it)

today we use them in offload mode (ie. the host would push memory+commands
onto them and pull the results off - all via pragmas ).

our last KNC systems were 2RU with 8x7120 phi's... which is a 2.1kW
system. They absolutely fly...

Post by Ryan Novosielski
We bought KNC a long time ago and keep meaning to get them to a place
where they can be used and just havenât. Do you mount filesystems from
them? We have GPFS storage, primarily, and would have to re-export it via
NFS I suppose if we want the cards to use that storage. Iâve seen
complaints about the stability of that setup. I didnât try to build the
GPFS portability layer for Phi â not sure whether to think it would or
wouldnât work (I guess Iâd be inclined to doubt it).

Post by Stu Midgley
Phi is dead... Long live phi...
By which I mean, while the Phi as a chip is going away, its concepts

live on. Massive number of cores, large vectorisation and high speed
memory (and fucking high heat load - we do ~350W/socket). So, while the
product code will disappear, phi lives on.

Post by Stu Midgley
For KNC I did a lot of customisation to MPSS to get it to work... and we

haven't been able to shift from one of the very early version. We love the
KNC... we get 8 in 2RU which is awesome density (1.1kW/RU)

Post by Stu Midgley
For KNL its just x86 with a big vectorisation unit (700W/RU).
In both cases you have to be very very careful how you manage memory.
I'm curious about your next gen plans, given Phi's roadmap.

Post by Stu Midgley
low level HPC means... lots of things. BUT we are a huge Xeon Phi shop

and need low-level programmers ie. avx512, careful cache/memory management
(NOT openmp/compiler vectorisation etc).

Post by Stu Midgley
I played around with avx512 in my rzf code.

https://github.com/joelandman/rzf/blob/master/avx2/rzf_avx512.c . Never
really spent a great deal of time on it, other than noting that using
avx512 seemed to downclock the core a bit on Skylake.

Post by Stu Midgley
Which dev/toolchain are you using for Phi? I set up the MPSS bit for a

customer, and it was pretty bad (2.6.32 kernel, etc.). Flaky control
plane, and a painful host->coprocessor interface. Did you develop your
own? Definitely curious.

Post by Stu Midgley
On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at

least until 6pm.

Post by Stu Midgley
I am excited but also terrified. My background is C and now JavaScript,

mostly online course work and telnet MUDs.

Post by Stu Midgley
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________

Computing

Post by Stu Midgley
To change your subscription (digest mode or unsubscribe) visit

http://www.beowulf.org/mailman/listinfo/beowulf

Post by Stu Midgley
--
Dr Stuart Midgley
_______________________________________________
Beowulf mailing list,
sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

http://www.beowulf.org/mailman/listinfo/beowulf

Post by Stu Midgley
--
Dr Stuart Midgley
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit

https://na01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%7Cnovosirj%40rutgers.edu%7C89d9a1fe40cd40448a5708d5d1abc4d9%7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645458049748846&sdata=dEUacidlV69%2FM8NEdObFNmSOsOObZpPAF4NlfI7joTw%3D&reserved=0
--
____
|| \\UTGERS, |---------------------------*O*---------------------------
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'

--
Dr Stuart Midgley
***@gmail.com

John Hearns via Beowulf

2018-06-20 03:57:34 UTC

This thread is going fast!

I often wonder if that misleading marketing is one of the reasons why the

Xeon Phi has already been canned. I know a lot of people who were excited
for the Xeon Phi, but > I don't know any who ever bought the Xeon Phis once
they came out.

In the UK at my last company we had a customer in the defence sector who
bought lots of Xeon Phi. Great guy, full of enthusiasm and good to work
with (Hello Kirk!)
They were installed with IBM Platform before I joined the company. I
re-installed the cluster with Bright which brought it up to date.
That is the cluster which used Teradici PCOIP to connect via secure fibre
optic links.

we initially used them as standalone systems (ie. rsh a code onto them and
run it)
today we use them in offload mode (ie. the host would push memory+commands
onto them and pull the results off - all via pragmas ).
our last KNC systems were 2RU with 8x7120 phi's... which is a 2.1kW
system. They absolutely fly...

Post by Stu Midgley
Phi is dead... Long live phi...
By which I mean, while the Phi as a chip is going away, its concepts

live on. Massive number of cores, large vectorisation and high speed
memory (and fucking high heat load - we do ~350W/socket). So, while the
product code will disappear, phi lives on.

Post by Stu Midgley
For KNC I did a lot of customisation to MPSS to get it to work... and

we haven't been able to shift from one of the very early version. We love
the KNC... we get 8 in 2RU which is awesome density (1.1kW/RU)

Post by Stu Midgley
low level HPC means... lots of things. BUT we are a huge Xeon Phi

shop and need low-level programmers ie. avx512, careful cache/memory
management (NOT openmp/compiler vectorisation etc).

Post by Stu Midgley
I played around with avx512 in my rzf code.

https://github.com/joelandman/rzf/blob/master/avx2/rzf_avx512.c .
Never really spent a great deal of time on it, other than noting that using
avx512 seemed to downclock the core a bit on Skylake.

Post by Stu Midgley
Which dev/toolchain are you using for Phi? I set up the MPSS bit for a

customer, and it was pretty bad (2.6.32 kernel, etc.). Flaky control
plane, and a painful host->coprocessor interface. Did you develop your
own? Definitely curious.

Post by Stu Midgley
On Thu, Jun 14, 2018 at 1:08 AM Jonathan Engwall <

Stuart Midgley works for DUG? They are currently
recruiting for an HPC manager in London... Interesting...

Recruitment at DUG wants to call me about Low Level HPC. I have at

least until 6pm.

Post by Stu Midgley
I am excited but also terrified. My background is C and now

JavaScript, mostly online course work and telnet MUDs.

Post by Stu Midgley
Any suggestions are very much needed.
What must a "low level HPC" know on day 1???
Jonathan Engwall
_______________________________________________

Computing

Post by Stu Midgley
To change your subscription (digest mode or unsubscribe) visit

http://www.beowulf.org/mailman/listinfo/beowulf

--
Joe Landman
c: +1 734 612 4615
https://scalability.org
https://github.com/joelandman
https://www.linkedin.com/in/joelandman
_______________________________________________

Computing

Post by Stu Midgley
To change your subscription (digest mode or unsubscribe) visit

http://www.beowulf.org/mailman/listinfo/beowulf

Post by Stu Midgley
--
Dr Stuart Midgley
_______________________________________________

Computing

Post by Stu Midgley
To change your subscription (digest mode or unsubscribe) visit

https://na01.safelinks.protection.outlook.com/?url=
http%3A%2F%2Fwww.beowulf.org%2Fmailman%2Flistinfo%2Fbeowulf&data=02%7C01%
7Cnovosirj%40rutgers.edu%7C89d9a1fe40cd40448a5708d5d1abc4d9%
7Cb92d2b234d35447093ff69aca6632ffe%7C1%7C0%7C636645458049748846&sdata=
dEUacidlV69%2FM8NEdObFNmSOsOObZpPAF4NlfI7joTw%3D&reserved=0
--
____
|| \\UTGERS, |---------------------------*
O*---------------------------
|| \\ University | Sr. Technologist - 973/972.0922 (2x0922) ~*~ RBHS Campus
|| \\ of NJ | Office of Advanced Research Computing - MSB C630, Newark
`'

--
Dr Stuart Midgley
_______________________________________________
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf

John Hearns via Beowulf

2018-06-20 04:07:47 UTC