Maxed (Dec 16 2010)


log in

Advanced search

Message boards : Technical News : Maxed (Dec 16 2010)

Previous · 1 · 2 · 3 · 4 · 5 . . . 6 · Next
Author Message
Profile Gary CharpentierProject donor
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 12729
Credit: 7,264,830
RAC: 17,534
United States
Message 1057063 - Posted: 17 Dec 2010, 17:08:55 UTC - in response to Message 1057036.

You have it correct. What you don't quite get is the "result" is called that even before it is sent to a cruncher's computer. At that point the state of the result is "ready to send" When it comes back the state gets changed, completed, error, time out, etc. When the wingman's result comes back the validator is notified and changes the state(s) again or it creates another result to be sent if inconclusive.

____________

Profile Bill Walker
Avatar
Send message
Joined: 4 Sep 99
Posts: 3402
Credit: 2,149,993
RAC: 2,239
Canada
Message 1057065 - Posted: 17 Dec 2010, 17:10:11 UTC
Last modified: 17 Dec 2010, 17:10:46 UTC

In most English speaking countries a "result" is indeed what comes after something. At Berkeley a result in a duplicate of a work unit, and it is called a result before it is sent, while it is out in the field, and after it comes home.

Edit: you just beat me Gary.
____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057073 - Posted: 17 Dec 2010, 17:24:30 UTC - in response to Message 1056883.

So who do you want to alienate? A lot of small crunchers and let the large crunchers hog the bandwidth/time slots, or a few big crunchers by letting everybody have an equal chunk of bandwidth/time slots.

Edited (note to self, don't forget to proof read before posting)


I suppose you are talking to me? I didn't know I had implied alienating anyone. I think the time and bandwidth would be best served by equal bandwidth and time more than letting one 'class' dominate the connections.


Well, you or anyone else with a "fix" for a finite resource, bandwidth.
You had said" ...the number of users fighting over spots makes it more of less a dog fight for the first position that opens up." I agree.

I think this is going to be true at any percentage of bandwidth or practical amount of bandwidth and judging from previous posts any limit artificially imposed by the S@H staff is going to alienate somebody.

So, does the staff lean towards the big crunchers (more efficient use of band width/less overhead with fewer connections doing larger downloads) or the smaller crunchers (everybody gets a little bit but with less efficient use of
bandwidth), or just let the dogfight continue?

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057080 - Posted: 17 Dec 2010, 17:37:38 UTC - in response to Message 1056883.

"I think the time and bandwidth would be best served by equal bandwidth and time more than letting one 'class' dominate the connections." BeNt

This favors the "small" cruncher over the "large cruncher", but there are people that could say I'm on the little end of the large crunchers. We haven't defined which is which yet. Now, I consider myself a small cruncher, but any fix the staff comes up with is OK with me because the project is more important than the individual cruncher.

That said, logically I'd go for the large cruncher resolution; best return on resources for the project.

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057087 - Posted: 17 Dec 2010, 17:48:27 UTC - in response to Message 1056886.

equal pay for everyone!!!


Cool, I'll be waiting at the mailbox for the check. :)

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057094 - Posted: 17 Dec 2010, 17:57:52 UTC - in response to Message 1056966.

equal pay for everyone!!!


From each according to their ability, to each according to their need. Isn't that the American way?


That's cool too. Since I have little ability and a bunch of need {8+ "procedures at last count), I'll be waiting at the mailbox.

Send American Eagles or stem cells. Preferably both. :)

Profile Profi
Volunteer tester
Send message
Joined: 8 Dec 00
Posts: 19
Credit: 13,917,080
RAC: 1,965
Poland
Message 1057099 - Posted: 17 Dec 2010, 18:05:32 UTC - in response to Message 1057065.
Last modified: 17 Dec 2010, 18:09:24 UTC

In most English speaking countries a "result" is indeed what comes after something. At Berkeley a result in a duplicate of a work unit, and it is called a result before it is sent, while it is out in the field, and after it comes home.



[sarcasm=on]
So a "result" is a outcome of the work of splitter process - raw chunks of telescope data?? - not even assigned to particular machine/user ?? - ufff what a mind job...

OK - I need to consult some geographer and linguist specialist, because I thought that Berkeley is in the US, and US is a English speaking country (vast majority)
[sarcasm=off]

Thanks guys for so swift replies.

Cheers

Profi

p.s. I understand of course result state - a "ready to send" state of the "result" is not very intuitive (at least for me)- however it may be caused by DB data flow or sth like that...
____________

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057101 - Posted: 17 Dec 2010, 18:08:34 UTC - in response to Message 1056978.

... This will be enacted next week during the return of the 3-day weekly outage. It's unclear how regular these extended weekly outages will be - we'll figure that all out in the new year.
...
- Matt


Seriously?

That means nothing has changed for us, the Cruncher, and we are at the
same Point as before the big Outage? Overloaded internet connection, a
3-Day Outage and babysitting our Rigs...

I had hoped for more (because of our generous donation)… :-(

Helli


We could always wait until after all the new kit has been sorted before making such a judgment.

My understanding was that there is at least a new hard drive, gigabit switch, and another server to be installed. Then the ensuing tweaking for efficiency.

Lets give the staff until the beginning if February before we get testy.

Barring any other emergencies, of course.

Profile HelliProject donor
Volunteer tester
Avatar
Send message
Joined: 15 Dec 99
Posts: 704
Credit: 90,245,824
RAC: 78,553
Germany
Message 1057108 - Posted: 17 Dec 2010, 18:18:54 UTC - in response to Message 1057101.

No doubt, we are waiting - as always. ;-)

Let's hope we can go back to a "normal", outage less, Situation soon - maybe next Year? :D

Helli
____________
A loooong time ago: My first Credits

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057113 - Posted: 17 Dec 2010, 18:26:10 UTC - in response to Message 1057083.

The dogfight (kittyfight) continues.

I, as a big cruncher.....or a donator....or a memeber of the clan of the green star......gets no more consideration by the scheduler when I make contact.

Other than by the fact that I have 8 rigs running Seti against some that have only 1.

So, there is nothing to support the supposition that anybody is getting preference here. (rest snipped)


I hadn't seen anybody say that anybody was getting preferential treatment. I thought we were discussing ways to maximize the throughput on the maxed out bandwidth.


Profile Albireo380
Send message
Joined: 21 Mar 08
Posts: 118
Credit: 1,403,094
RAC: 621
United Kingdom
Message 1057120 - Posted: 17 Dec 2010, 18:35:01 UTC - in response to Message 1057098.

[/quote]

If you think I am bad now........I was virtually hell on wheels back then.[/quote]

Just thinking about "Hell on wheels" Sattler makes me smile, Mark. I am just glad I didn't have to deal with it.

Got any new Kitty images? My wife keeps asking where the Kittyman's latest batch are.

Like you, I am just pleased that we are back up and running, even if not fully out of the woods - Keep crunching

Tom

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057123 - Posted: 17 Dec 2010, 18:43:36 UTC - in response to Message 1057107.

(snippage)


I'll give them exactly until hell freezes over.
Barring any other emergencies, of course.


I used to have a T-shirt that said "I was there when Hell froze over,
and had forgotten my ice skates."

I think my wife threw it out with the "It's not a bald spot, it's a solar panel for a s** machine" one. She hates T-shirts with rotted out armpits, whereas I think I've just got them broken in.

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057126 - Posted: 17 Dec 2010, 18:48:43 UTC - in response to Message 1057108.

No doubt, we are waiting - as always. ;-)

Let's hope we can go back to a "normal", outage less, Situation soon - maybe next Year? :D

Helli


That would be pleasant. Maybe we could come up with a new catch phrase,
"Next year, in Berkeley."

bill
Send message
Joined: 16 Jun 99
Posts: 861
Credit: 23,975,287
RAC: 13,839
United States
Message 1057130 - Posted: 17 Dec 2010, 18:55:27 UTC - in response to Message 1057117.

(snippage)

No matter. The sentiment remains the same......
No pay for play....no preferences.
All are equal in the scheduler's eyes.

One request, one response. Same for all.



Oh, I wouldn't be against 2 gigabit pipes. One with general access and one you could subscribe to, the more you used it the more you paid, say a penny a gigabit of download, uploads free.

Previous · 1 · 2 · 3 · 4 · 5 . . . 6 · Next

Message boards : Technical News : Maxed (Dec 16 2010)

Copyright © 2014 University of California