On 03/07/2014, Joseph Landman wrote:
>
>
>
> Now just get all those writing rfps to stop with the madness. The number
> and it's pursuit are one of the more significant causes of entropy (as in
> waste heat generation ) as I have ever seen. A dubious metric at best with
> extremely limited, if a
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 03/07/14 07:37, James Cuff wrote:
> Repeat after me:
>
> "Top 500 scores mean nothing!"
I was at a talk about the SKA project on Tuesday night here in
Melbourne, and it touched on just this problem.
They need about 7PF of FFTs alone.
Unfortun
To me it seems that the top lines of the Top500 list have become more
about nation-state appendage measuring than about real science.
China spiked the ball squarely in jump-the-shark territory. Tianhe-2
will likely end up like most of the venues built for the Beijing Summer
Olympics.
On 7/2/14,
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 02/07/14 16:19, Jonathan Aquilina wrote:
> How would the same arguments apply if you are just dealing with
> dns servers web servers databases etc.
Sorry, I thought we were still talking about dealing with multiple
compiler versions in HPC.
The t
On Jul 2, 2014, at 7:31 PM, James Cuff wrote:
Let this be a lesson to us all.
Repeat after me:
"Top 500 scores mean nothing!"
But ... But ... Shiny ... Precious ...
Please pardon brevity and typos ... Sent from my iPhone
As a community we have to stop this madness! 100k / day. Sigh.
N
> On Jul 2, 2014, at 5:55 PM, James Cuff wrote:
>
>
> 404: E_NO_SCIENCE
>
James wins the internets today!
Please pardon brevity and typos ... Sent from my iPhone
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing
To change yo
Well, to think that it's been an issue even with game consoles like PS3,
and people have yet not learned... :/
2014-07-02 18:37 GMT-03:00 James Cuff :
>
> Let this be a lesson to us all.
>
> Repeat after me:
>
> "Top 500 scores mean nothing!"
>
> As a community we have to stop this madness! 100
Let this be a lesson to us all.
Repeat after me:
"Top 500 scores mean nothing!"
As a community we have to stop this madness! 100k / day. Sigh.
J.
On Wednesday, July 2, 2014, Jeff Johnson
wrote:
> $100,000 (USD) per day to operate, users get billed.
>
> "It's like a giant with a super b
404: E_NO_SCIENCE
On Wednesday, July 2, 2014, "C. Bergström" wrote:
> On 07/ 3/14 04:17 AM, Jeff Johnson wrote:
>
>> If they want to spend a bazillion dollars to run hpl faster than anyone
>> else who am I to stop them.
>>
>> If however they want to do real science perhaps they need to archit
On 7/2/14, 2:35 PM, Joseph Landman wrote:
> Please pardon brevity and typos ... Sent from my iPhone
>
>> On Jul 2, 2014, at 5:31 PM, Jeff Johnson
>> wrote:
>>
>> $100,000 (USD) per day to operate, users get billed.
>>
>>"It's like a giant with a super body but without the software to
>>su
Please pardon brevity and typos ... Sent from my iPhone
> On Jul 2, 2014, at 5:31 PM, Jeff Johnson
> wrote:
>
> $100,000 (USD) per day to operate, users get billed.
>
>"It's like a giant with a super body but without the software to
>support its thinking soul," Chi said. Some users would
$100,000 (USD) per day to operate, users get billed.
"It's like a giant with a super body but without the software to
support its thinking soul," Chi said. Some users would need years or
even a decade to write the necessary code, he added.
Ouch...
On 7/2/14, 2:21 PM, "C. Bergström" w
On 07/ 3/14 04:28 AM, James Cuff wrote:
404: E_NO_SCIENCE
lol - check my last email - they didn't pull it and the original link
just got cut - probably by original email client
___
Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Comput
On 07/ 3/14 04:17 AM, Jeff Johnson wrote:
If they want to spend a bazillion dollars to run hpl faster than anyone
else who am I to stop them.
If however they want to do real science perhaps they need to architect
something more manageable.
They should bust that thing up into 3 or 4 clusters.
O
On 07/ 3/14 04:17 AM, Jeff Johnson wrote:
If they want to spend a bazillion dollars to run hpl faster than anyone
else who am I to stop them.
If however they want to do real science perhaps they need to architect
something more manageable.
They should bust that thing up into 3 or 4 clusters.
40
If they want to spend a bazillion dollars to run hpl faster than anyone
else who am I to stop them.
If however they want to do real science perhaps they need to architect
something more manageable.
They should bust that thing up into 3 or 4 clusters.
On 7/2/14, 2:11 PM, Douglas Eadline wrote:
>
China's world-beating supercomputer fails to impress some potential clients
http://www.scmp.com/news/china/article/1543226/chinas-world-beating-supercomputer-fails-imp
--
Doug
--
Mailscanner: Clean
___
Beowulf mailing list, Beowulf@beowulf.org spo
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
18 matches
Mail list logo
