Whoohoo! Validator on atlast...


log in

Advanced search

Message boards : Number crunching : Whoohoo! Validator on atlast...

Author Message
Prototype
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 67
Credit: 414,025
RAC: 0
United Kingdom
Message 1503 - Posted: 24 Jun 2004, 23:38:19 UTC

... S@H credits starting to roll (slowly) for now.

:)


machx0r
Send message
Joined: 3 Apr 99
Posts: 21
Credit: 151,017
RAC: 0
United States
Message 1517 - Posted: 24 Jun 2004, 23:47:58 UTC - in response to Message 1503.

Someone correct me if I'm wrong but to get credit I think three people need to process the same WU and come up with similiar results. I don't think there is a process on the server doing this.

freespace2dotcom
Send message
Joined: 18 Feb 04
Posts: 3
Credit: 139,190
RAC: 0
United States
Message 1529 - Posted: 24 Jun 2004, 23:56:56 UTC - in response to Message 1517.

Well, I've sent in a few results, and they seem to be waiting for validation. Everything seems to be working. Just have to wait for those slowpokes to catch up. ;)

> Someone correct me if I'm wrong but to get credit I think three people need to
> process the same WU and come up with similiar results. I don't think there is
> a process on the server doing this.

Profile Paul D. Buck
Volunteer tester
Send message
Joined: 19 Jul 00
Posts: 3898
Credit: 1,158,042
RAC: 0
United States
Message 1816 - Posted: 25 Jun 2004, 11:04:25 UTC

Yes, three people have to process the WU, get "similar" results from processing it (number of spikes, etc...) and then if they are all close enough, now errors in processing, a quorum of results is created and then credit will be granted.

Also, on timeing there are two uploads of the WU, and I can never keep them straight. I believe the first one (which is automatic after) is a very small amount of data, at the next communication between the BOINC Software and the server the actual data generated is uploaded. If I have it wrong one of the Sir Johns will straighten me out ...
<p>


<p>

Profile Keck_Komputers
Volunteer tester
Avatar
Send message
Joined: 4 Jul 99
Posts: 1575
Credit: 1,616,069
RAC: 750
United States
Message 2054 - Posted: 25 Jun 2004, 19:04:33 UTC - in response to Message 1816.

> Yes, three people have to process the WU, get "similar" results from
> processing it (number of spikes, etc...) and then if they are all close
> enough, now errors in processing, a quorum of results is created and then
> credit will be granted.
>
> Also, on timeing there are two uploads of the WU, and I can never keep them
> straight. I believe the first one (which is automatic after) is a very small
> amount of data, at the next communication between the BOINC Software and the
> server the actual data generated is uploaded. If I have it wrong one of the
> Sir Johns will straighten me out ...
> <p>
The data is uploaded automatically at the end of the workunit. The small descriptive report is sent later.

John Keck
testing BOINC since 2002/12/08

Profile ror
Send message
Joined: 3 Jun 04
Posts: 28
Credit: 3,020
RAC: 0
United Kingdom
Message 2058 - Posted: 25 Jun 2004, 19:11:15 UTC - in response to Message 1517.

> Someone correct me if I'm wrong but to get credit I think three people need to
> process the same WU and come up with similiar results. I don't think there is
> a process on the server doing this.
>
>

There is also a validator that has to check the 3 results are the same, hand out credit etc, although it's been online for a while now.

p.s. sorry if this is a double-post, I got disconnected whilst sending it the first time.

Message boards : Number crunching : Whoohoo! Validator on atlast...

Copyright © 2014 University of California