Message boards : Number crunching : Initial Replication
Message board moderation

To post messages, you must log in.

1 · 2 · 3 · 4 . . . 7 · Next

AuthorMessage
Betting Slip

Send message
Joined: 17 Sep 04
Posts: 41
Credit: 27,497
RAC: 0
Message 17557 - Posted: 26 Jul 2007, 8:58:48 UTC

Why waste computing time? If you only need a quorum of 3 why an initial replication of 5???
ID: 17557 · Report as offensive     Reply Quote
daemion

Send message
Joined: 4 Dec 06
Posts: 33
Credit: 75,491
RAC: 0
Message 17561 - Posted: 26 Jul 2007, 14:33:31 UTC

and now people are complaining that there are too MANY work units. there is just no pleasing BOINC users.....
ID: 17561 · Report as offensive     Reply Quote
Profile Neasan
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 30 Nov 06
Posts: 234
Credit: 11,078
RAC: 0
Message 17563 - Posted: 26 Jul 2007, 14:58:25 UTC

Me?
I'm laughing!
It's all you can do :-D
ID: 17563 · Report as offensive     Reply Quote
Betting Slip

Send message
Joined: 17 Sep 04
Posts: 41
Credit: 27,497
RAC: 0
Message 17572 - Posted: 27 Jul 2007, 9:39:53 UTC - in response to Message 17561.  

and now people are complaining that there are too MANY work units. there is just no pleasing BOINC users.....


Nothing to do with how many WU's there are available but everything to do with how many times we crunch those that are available.

There are multiple projects and they all need resources without any one project wasting them.
ID: 17572 · Report as offensive     Reply Quote
Betting Slip

Send message
Joined: 17 Sep 04
Posts: 41
Credit: 27,497
RAC: 0
Message 17573 - Posted: 27 Jul 2007, 9:44:08 UTC - in response to Message 17563.  

Me?
I'm laughing!
It's all you can do :-D



Could it be that you can't understand the arguement and you are laughing from embarrassment?
ID: 17573 · Report as offensive     Reply Quote
Betting Slip

Send message
Joined: 17 Sep 04
Posts: 41
Credit: 27,497
RAC: 0
Message 17574 - Posted: 27 Jul 2007, 10:13:51 UTC - in response to Message 17561.  

and now people are complaining that there are too MANY work units. there is just no pleasing BOINC users.....


If you read and understood the originl post you would of course realise this was not a complaint about the quantity of WU's.

TIP, Don't try to score cheap points of me or anyone else!
ID: 17574 · Report as offensive     Reply Quote
Profile Gary Roberts

Send message
Joined: 22 Jul 05
Posts: 72
Credit: 3,962,626
RAC: 0
Message 17575 - Posted: 27 Jul 2007, 10:18:28 UTC - in response to Message 17561.  

and now people are complaining that there are too MANY work units. there is just no pleasing BOINC users.....



Did you actually understand the message you were responding to? The OP didn't say anything about work units. His comment was about initial replication (IR).

I think I saw Neasan mention somewhere that there were about 20,000 work units in the last series. Whether the IR is 1, 3, 5, or 101, there are still 20,000 WUs. So the OP wasn't complaining about too many WUs as you assert.

The OP has actually raised an issue (admittedly debated many times in the past) that is probably worthy of further debate due to changing circumstances. As I understand it, a couple of years ago the scientists wanted quick answers so they could get on with the design, with each stage depending on the answers to the previous simulations. I would imagine that things are rather different now as the time for commissioning should be fast approaching.

To get quick answers back then, the scientists opted to be rather profligate with crunching resources. Do they really need to be that way now? With the current push to limit greenhouse emissions, I would have thought that now is the time to cut back on unnecessary crunching. With an IR=5, there are 100,000 results in the last series. If IR=3 there would only be 60,000 and a lot of electricity gets saved.

Sure, more will be sent out to cover any that fail to return by the deadline or fail to validate, but that happens anyway. To cut down on the total time to get all results back, just reduce the deadline from 7 days to 4 days so that a 4th result gets sent out more quickly when needed. Recent versions of BOINC seen to be good at not sending work where there is a risk of it not completing in time so a 4 day deadline would restrict those people who were deliberately trying to fill huge caches. Neasan mentioned at one stage that he was giving some thought to ways to stop people being greedy by setting large caches. A nice short deadline would certainly add some encouragement to limiting the number of results per CPU that people attempt to get.


Cheers,
Gary.
ID: 17575 · Report as offensive     Reply Quote
Bob Guy

Send message
Joined: 28 Sep 05
Posts: 21
Credit: 11,715
RAC: 0
Message 17583 - Posted: 28 Jul 2007, 7:00:37 UTC

I mean for this comment to be somewhat comical, so don't take it to be an attack on anyone's political or ecological beliefs.

My comment:

One of the largest single sources of greenhouse emmissions is cows, and they don't take kindly to corks!
ID: 17583 · Report as offensive     Reply Quote
Mike Dunn

Send message
Joined: 15 Jul 05
Posts: 9
Credit: 770,253
RAC: 0
Message 17584 - Posted: 28 Jul 2007, 7:40:58 UTC - in response to Message 17575.  
Last modified: 28 Jul 2007, 7:41:35 UTC

If IR=3 there would only be 60,000 and a lot of electricity gets saved.

I doubt that very much. In an ideal world, or one where the LHC cruncher JUST does LHC, yes you're right.

But how many ONLY do this project ? Anyone ?

So in reality - we crunch 60,000 WUs and then swap back to other projects that have WUs on a regular basis - no electricity saved there ......

As to wasting the time - that's a personal opinion, and one I don't hold. I'm happy to have LHC WUs, myself.
ID: 17584 · Report as offensive     Reply Quote
Profile Gary Roberts

Send message
Joined: 22 Jul 05
Posts: 72
Credit: 3,962,626
RAC: 0
Message 17589 - Posted: 29 Jul 2007, 2:53:05 UTC - in response to Message 17583.  

.... cows, and they don't take kindly to corks!


Ahhh yes! We have a lot of talk about clean coal technology so maybe we need a clean and green cow technology ;).

But what do we do about flatulent humans ... :).



Cheers,
Gary.
ID: 17589 · Report as offensive     Reply Quote
Profile Gary Roberts

Send message
Joined: 22 Jul 05
Posts: 72
Credit: 3,962,626
RAC: 0
Message 17590 - Posted: 29 Jul 2007, 3:06:05 UTC - in response to Message 17584.  

... swap back to other projects that have WUs on a regular basis - no electricity saved there ......


You're absolutely right! More science gets done but no electricity is saved. It's just like designing more fuel efficient engines. People will probably just travel more miles and no fuel will likely be saved so we don't need those better engines either :).

As to wasting the time - that's a personal opinion, and one I don't hold. I'm happy to have LHC WUs, myself.


I'm very happy to have meaningful work from LHC. It's just that I'd like to see some other science project have the benefit of the time spent on 40,000 redundant LHC results. Just a personal opinion ... :).


Cheers,
Gary.
ID: 17590 · Report as offensive     Reply Quote
Profile Alex

Send message
Joined: 2 Sep 04
Posts: 378
Credit: 10,765
RAC: 0
Message 17593 - Posted: 29 Jul 2007, 19:37:31 UTC - in response to Message 17557.  
Last modified: 29 Jul 2007, 20:07:21 UTC

Why waste computing time? If you only need a quorum of 3 why an initial replication of 5???



There's a mathematical reason that there's 5 sent out instead of 4.
20-25% of the crunching instances have errors. (I looked at the most recent work units that MY pc crunched, and there were 5 out of 27 client errors reported by all pc's crunching)
With that error rate, a replication of 4 ends up with the server having to resend out work units again after the deadline has passed for 20% of the work units.

For a 20% error rate:
1000 work units, will need to have 200 work units redone. 200 reworked units will fail again, resulting in another 40 to be redone, those 40 will have 8 that needs to be redone, and another 8 will be redone.
This results in having to wait 3 iterations for all results to come in.

If you have a replication of 5.
1000 work units will need to have 70 redone. Those 70 will need 10 redone, and those 10 may need 1 redone.

It's when you have a 25% error rate, that having a replication of 4 ends up in more work than an initial replicaton of 5.

Crunching for LHC means that the Intel PC's don't always agree with the AMD PC's, so you will always see a higher redo rate than Seti (which is just a fast fourier filter performed on data which is not sensitive to floating point errors in the the last precision digit)

Then again, some people's computers just suck: http://lhcathome.cern.ch/lhcathome/results.php?hostid=88058




I'm not the LHC Alex. Just a number cruncher like everyone else here.
ID: 17593 · Report as offensive     Reply Quote
KAMasud

Send message
Joined: 7 Oct 06
Posts: 114
Credit: 23,192
RAC: 0
Message 17628 - Posted: 30 Jul 2007, 13:19:43 UTC


:-)welcome back Dagorath. we were computing the adjustments of the magnets for the machine at CERN. i think the power required to warm it up will be far more then what we spend on crunching extra WU's. keep the safety factor in mind.
Regards
Masud.
ID: 17628 · Report as offensive     Reply Quote
Colin Porter

Send message
Joined: 14 Jul 05
Posts: 35
Credit: 71,636
RAC: 0
Message 17639 - Posted: 30 Jul 2007, 21:58:12 UTC
Last modified: 30 Jul 2007, 22:01:27 UTC

As an example I give you a wu from Einstein

Einstein unit

basically the wu was initially issued on 8th June but did not reach quorum until 27th July !! enought said?

The project admins/scientist are the only people who have the whole picture, we who contribute our resourses can only trust that they are choosing the correct
parameters. Thats not to say we cannot suggest what we think is better. Einstein can afford this kind of thing occasionally, LHC can't.
ID: 17639 · Report as offensive     Reply Quote
Profile Alex

Send message
Joined: 2 Sep 04
Posts: 378
Credit: 10,765
RAC: 0
Message 17641 - Posted: 31 Jul 2007, 2:36:02 UTC - in response to Message 17627.  

I don't care what you think Dagorath.



I'm not the LHC Alex. Just a number cruncher like everyone else here.
ID: 17641 · Report as offensive     Reply Quote
Moebius2000

Send message
Joined: 28 Jan 07
Posts: 1
Credit: 86,828
RAC: 0
Message 17643 - Posted: 31 Jul 2007, 8:03:06 UTC
Last modified: 31 Jul 2007, 8:03:49 UTC

Is it really necessary to write like this?
Please have a look on the rules next to the textfield you use to enter your posts.
I think we ALL are capable to dicuss without attacking each others.

So, back on topic:
I think it is good that this topic was raised and I'm sure that the project admins will tell us their opinion and maybe their reasons why they use 5 IR, when they have time for it. Until then, I think we don't need to attack each other here as I said before.

So long, have a nice day.

Moebius2000
ID: 17643 · Report as offensive     Reply Quote
1 · 2 · 3 · 4 . . . 7 · Next

Message boards : Number crunching : Initial Replication


©2024 CERN