| Author |
Message |
Scott Moore
Guest
|
 Posted:
Fri Dec 02, 2005 1:15 am Post subject:
Re: The Emperor's new clothes |
|
|
Joe Seigh wrote On 11/21/05 08:41,:
| Quote: | So these processor manufacturers all have these
nice new multi-core cpu's but apart from market
hyperbole (these cpu's will save the environment, etc...)
I don't see them actually doing anything to exploit their
potential. By "them", I mean them not us. We of course
know to do. But what's going on to get all the applications
to start exploiting this? The magic parallelization fairy?
|
It will allow you to compile without your yahoo music service
skipping, or allow you to run virus checks without slowing your
system to a crawl, etc.
Yes, certainly applications are way behind. I notice my web
browser halts all of its open windows when it is waiting to
get to just one site.
I personally see that as a tooling issue, since many programmers
who have tried multithreaded applications realize it mostly
generates massive and difficult to understand bugs.
I am personally for supporting it at the language level (and I
do). |
|
| Back to top |
|
 |
Guest
|
| Posted:
Fri Dec 02, 2005 9:03 am Post subject:
Re: The Emperor's new clothes |
|
|
Brian Hurt <bhurt@AUTO> writes:
| Quote: | FWIW "paralization" doesn't show up in the Dashboard dictionary widget
(Oxford American Dictionaries), but "parallelization" does.
It's an amalgamation of parallel and paralyze. The magic paralization
fairy allows you to deadlock eight cpus simultaneously.
|
Brian, that hurts... :-) |
|
| Back to top |
|
|
Terje Mathisen
Guest
|
| Posted:
Fri Dec 02, 2005 5:15 pm Post subject:
Re: The Emperor's new clothes |
|
|
Scott Moore wrote:
| Quote: | Joe Seigh wrote On 11/21/05 08:41,:
So these processor manufacturers all have these
nice new multi-core cpu's but apart from market
hyperbole (these cpu's will save the environment, etc...)
I don't see them actually doing anything to exploit their
potential. By "them", I mean them not us. We of course
know to do. But what's going on to get all the applications
to start exploiting this? The magic parallelization fairy?
It will allow you to compile without your yahoo music service
skipping, or allow you to run virus checks without slowing your
system to a crawl, etc.
|
Yes! It magically makes my (single/laptop) hard drive multi-threaded as
well! :-)
| Quote: |
Yes, certainly applications are way behind. I notice my web
browser halts all of its open windows when it is waiting to
get to just one site.
|
Usually when said site is loading some code (Java or even .js).
| Quote: |
I personally see that as a tooling issue, since many programmers
who have tried multithreaded applications realize it mostly
generates massive and difficult to understand bugs.
I am personally for supporting it at the language level (and I
do).
|
Good!
Terje
--
- <Terje.Mathisen@hda.hydro.com>
"almost all programming can be viewed as an exercise in caching" |
|
| Back to top |
|
|
Terje Mathisen
Guest
|
| Posted:
Fri Dec 02, 2005 5:15 pm Post subject:
Re: The Emperor's new clothes |
|
|
Brian Hurt wrote:
| Quote: | nospam@ab-katrinedal.dk (Niels Jřrgen Kruse) writes:
FWIW "paralization" doesn't show up in the Dashboard dictionary widget
(Oxford American Dictionaries), but "parallelization" does.
It's an amalgamation of parallel and paralyze. The magic paralization
fairy allows you to deadlock eight cpus simultaneously. Boy, I wish I
was joking.
|
Ouch. We spent last night (literally: start at midnight) migrating a
failing Java-based document server application to a backup box. Due to
programming bug(s) it would deadlock a new thread each time a user
aborted out of a file transfer.
"How many threads do we have currently?"
70
"How many are deadlocked, while also locking all corresponding heap
buffers?"
68?
The problem was of course that as soon as a significant amount of
resources had been lost this way, response time went way up, which then
would significantly increase the chance of a given user giving up on yet
another transfer.
In another field this is called a chain reaction, the end result is ugly
and it generates a lot of hot air.
Terje
--
- <Terje.Mathisen@hda.hydro.com>
"almost all programming can be viewed as an exercise in caching" |
|
| Back to top |
|
|
Edward Wolfgram
Guest
|
| Posted:
Tue Dec 20, 2005 1:15 am Post subject:
Re: The Emperor's new clothes |
|
|
Scott Moore wrote:
| Quote: | Joe Seigh wrote On 11/21/05 08:41,:
So these processor manufacturers all have these
nice new multi-core cpu's but apart from market
hyperbole (these cpu's will save the environment, etc...)
I don't see them actually doing anything to exploit their
potential. By "them", I mean them not us. We of course
know to do. But what's going on to get all the applications
to start exploiting this? The magic parallelization fairy?
It will allow you to compile without your yahoo music service
skipping, or allow you to run virus checks without slowing your
system to a crawl, etc.
Yes, certainly applications are way behind. I notice my web
browser halts all of its open windows when it is waiting to
get to just one site.
I personally see that as a tooling issue, since many programmers
who have tried multithreaded applications realize it mostly
generates massive and difficult to understand bugs.
I am personally for supporting it at the language level (and I
do).
|
The problems you are describing have nothing to do with parallelization
and everything to do with intelligent OS scheduling and prioritization.
Multiuser operating systems deal with these issues everyday, as so
designers of applications on multiuser systems.
There is nothing magic about multi-tasking systems. They are not hard to
code. If you stop buying bad software, companies will start writing good
software. :)
Edward Wolfgram |
|
| Back to top |
|
|
Jan Vorbrüggen
Guest
|
| Posted:
Tue Dec 20, 2005 9:15 am Post subject:
Re: The Emperor's new clothes |
|
|
Scott Moore wrote:
| Quote: | Yes, certainly applications are way behind. I notice my web
browser halts all of its open windows when it is waiting to
get to just one site.
|
My browser only does that when it's in some parts of the TCP/IP
stack (e.g., name resolution). So let's tell those guys in Red-
mond to do their jib properly, shall we? Good luck to all of us.
Jan |
|
| Back to top |
|
|
|
|
|
|
FAQ
Memberlist
Register
Profile
Log in to check your private messages
Log in