Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon. Entire thread

Supercomputing

Name: Anonymous 2008-01-08 22:47

Why isn't C popular in Supercomputing?

Name: Anonymous 2008-01-08 22:58

I don't know.

Name: Anonymous 2008-01-08 23:00

Sepples v0.1 alpha can't stop there. That's FORTRAN country.

Name: Anonymous 2008-01-08 23:02

Why would it be?

Name: Anonymous 2008-01-08 23:10

Because it's a SUPERFAILURE

Name: Anonymous 2008-01-08 23:19

What is popular in "Supercomputing"?

Name: Anonymous 2008-01-08 23:21

>>5
You may hate it, but its essential to know some and will continue to be so until we hit Quantum mode

Name: Anonymous 2008-01-08 23:24

>>6
FORTRAN

Name: Anonymous 2008-01-09 1:08

anything with multicore support out of the box, c can have multicore support but it's harder to write, it all comes down to being harder to do and ppl being fucking lazy

Name: Anonymous 2008-01-09 2:49

>>9
ppl being fucking lazy: true
[parallel programming] being harder to do: WHAT THE FUCK IS WRONG WITH YOU?  NO IT ISN'T!

Name: Anonymous 2008-01-09 4:01

>>10
you have more things to worry about. it contributes to the overall complexity of the code.

Name: Anonymous 2008-01-09 5:28

>>11
No it doesn't.  If you have a codebase and workload of any size, you already have data isolation, and blocking becomes a serious issue.  Going parallel from the get-go is easier.

Name: Anonymous 2008-01-09 5:33

>>11
Depends on the language. Of course you have more things to worry about when you use C and have indiscriminated acces to the computer's memory

Name: 11 2008-01-09 5:56

If you have a codebase and workload of any size, you already have data isolation, and blocking becomes a serious issue. 
Yeh, I was mostly thinking about the dinky little programs that I work with. For most of the projects I've worked on, parallel programming isn't necessary and trying to shoehorn some parallel model into whatever we already had was a useless cognitive burden. I have also been in a couple of projects where hindsight vision showed me where going parallel would have been very useful.

Name: Anonymous 2008-01-11 17:44

>>11
Yes, is it hard (an absolute term, instead of relative to non-concurrent programming) - not as much as people make it out to be. The myth that concurrent programming is too hardTM comes from all the (Windows? is it fair to blame it on windows? probably not) programmers who are so used to thinking non-concurrently that the paradigm shift to concurrent seems overly hard. It's a relative thing, not an absolute thing. Obviously it's harder because it's more complex (as you said), but how much harder is blown out of proportion because people are taught how to program single-threaded non-concurrent applications too well before they are taught how to program concurrently.

Name: Anonymous 2008-01-11 17:46

>>15 grr should have read what I wrote again before posting, I messed up what I was trying to say ;-) I think you can still understand my point though.

Name: Anonymous 2009-03-18 3:27

I'm feeling really keen, for some of that good ol' green

Marijuana MUST be legalized.

Name: Trollbot9000 2009-07-01 9:28

before With L o L L O  L O L  W U T  d GRUNNUR ar.

Name: Anonymous 2011-01-31 20:06

<-- check em dubz

Name: Anonymous 2013-08-31 18:22


'm told that in the Southern hemisphere cyclones are Low pressure and go anticlockwise, the opposite being true for anticyclones in the SH, and the reverse in the Northern hemisphere. In some text on the internet it says that you can tell by looking at the cyclone if you are in the Northern hemisphere. How does that work?

Newer Posts
Don't change these.
Name: Email:
Entire Thread Thread List