more data on the way


log in

Advanced search

Message boards : News : more data on the way

Previous · 1 · 2 · 3 · 4 · Next
Author Message
Mike Dymond
Send message
Joined: 3 Apr 02
Posts: 3
Credit: 4,037,655
RAC: 20,827
United States
Message 1151677 - Posted: 13 Sep 2011, 3:37:48 UTC - in response to Message 1150111.

In an attempt to push some older unanalyzed files through the pipeline we encountered data that could not be successfully preprocessed or split. This has resulted in work distribution going to near zero. We are currently transferring newer files from off site storage and soon will be receiving a disk shipment from Arecibo. Once these data are on hand work distribution will pick up.



Are the data packets small enough to be accomplished within a month? I don't mind dedicating my main computer to analyzing only those packets to help catch up (credit or no). Just yell if there's anything i can do.

Profile Donald L. JohnsonProject donor
Avatar
Send message
Joined: 5 Aug 02
Posts: 6186
Credit: 698,508
RAC: 1,189
United States
Message 1151689 - Posted: 13 Sep 2011, 4:41:05 UTC - in response to Message 1151677.
Last modified: 13 Sep 2011, 4:41:38 UTC

In an attempt to push some older unanalyzed files through the pipeline we encountered data that could not be successfully preprocessed or split. This has resulted in work distribution going to near zero. We are currently transferring newer files from off site storage and soon will be receiving a disk shipment from Arecibo. Once these data are on hand work distribution will pick up.


Are the data packets small enough to be accomplished within a month? I don't mind dedicating my main computer to analyzing only those packets to help catch up (credit or no). Just yell if there's anything i can do.

Mike, these are not packets, they are whole data files (50 GB) like the ones you see on the splitter status table. As I understand the problem, the data files had issues such that they could not be prepped and/or split into Work Units by the S@H servers, Unless/until they find a way to fix those data files, there is nothing anyone outside the project staff can do with that data.
____________
Donald
Infernal Optimist / Submariner, retired

Profile Xen
Avatar
Send message
Joined: 22 Jul 00
Posts: 86
Credit: 2,085,157
RAC: 1,026
United States
Message 1151858 - Posted: 13 Sep 2011, 21:22:56 UTC

Thanks for the info Jeff!!!
____________
Nobody is nobody. Everyone has something to offer

Mike Dymond
Send message
Joined: 3 Apr 02
Posts: 3
Credit: 4,037,655
RAC: 20,827
United States
Message 1151985 - Posted: 14 Sep 2011, 3:49:44 UTC - in response to Message 1151858.

Well, it was worth a shot...

CincyRick
Avatar
Send message
Joined: 28 Dec 08
Posts: 2
Credit: 160,336
RAC: 0
United States
Message 1152056 - Posted: 14 Sep 2011, 9:37:24 UTC

I have two computers and about four or five files to upload and the program keeps backing off from the upload. when I did get new data on one of my computers I had to stop getting tasks because I recieved 30 or more files for crunching, working through those now and will continue, but the files are not uploading as of 8 pm eastern time on the 13 Sept.

Profile Mad Fritz
Avatar
Send message
Joined: 20 Jul 01
Posts: 87
Credit: 11,334,904
RAC: 0
Switzerland
Message 1152064 - Posted: 14 Sep 2011, 10:40:16 UTC - in response to Message 1152056.

I have two computers and about four or five files to upload and the program keeps backing off from the upload. when I did get new data on one of my computers I had to stop getting tasks because I recieved 30 or more files for crunching, working through those now and will continue, but the files are not uploading as of 8 pm eastern time on the 13 Sept.


http://setiathome.berkeley.edu/sah_status.html

Upload-server under maintenance.

BTW, I suggest to use

http://setiathome.berkeley.edu/forum_forum.php?id=10

for questions & complaints...

BR

MF


____________

Aurora Borealis
Volunteer tester
Avatar
Send message
Joined: 14 Jan 01
Posts: 2975
Credit: 5,028,233
RAC: 1,302
Canada
Message 1152065 - Posted: 14 Sep 2011, 10:44:25 UTC - in response to Message 1152056.
Last modified: 14 Sep 2011, 10:46:55 UTC

Please visit the Number Crunching forum to discuss user problems. The News forum is for updates from the project scientist.

Profile Douglas WONG
Send message
Joined: 6 Apr 03
Posts: 36
Credit: 1,493,275
RAC: 0
Hong Kong
Message 1158727 - Posted: 4 Oct 2011, 4:22:47 UTC - in response to Message 1152065.

lovely that it is back!!!!
____________

Jeff
Send message
Joined: 18 May 99
Posts: 1
Credit: 1,812,471
RAC: 1,385
United Kingdom
Message 1159004 - Posted: 5 Oct 2011, 10:46:59 UTC - in response to Message 1158727.

Are you sure? Data stopped arriving yesterday. Any reason for this?

Gerald King
Send message
Joined: 3 Apr 99
Posts: 14
Credit: 9,052,060
RAC: 6,966
United States
Message 1159051 - Posted: 5 Oct 2011, 13:45:42 UTC - in response to Message 1159004.

I've gotten ALOT of CUDA downloads and have some waiting to download, but I haven't gotten any normal workloads in quite some time.
____________

Profile coyote
Avatar
Send message
Joined: 29 Apr 99
Posts: 8
Credit: 44,541,516
RAC: 1,699
United States
Message 1159153 - Posted: 5 Oct 2011, 18:05:36 UTC - in response to Message 1159051.

I can concur with the lack of downloads. My log is chock full of "Scheduler
request failed: Cound't connect to server". I'm patient, but I would like to
get some work units (CPU or GPU) downloaded soon. I've had my best machine
idle for the past 3 days, due to a lack of work units, or problems in accessing
the site (know it's been down due to technnical problems).

I also find it interesting that, sometimes, after the weekly downtime (or during
unexpected downtime) that one of my machines will get no work units, while most
all of the others download a full share. And, many times, one machine will get
a pant load of CPU work units and no GPU units, while the other servers get a
mix. I don't understand why this would be, and it truly is of no great import,
but I thought I'd mention, as that segues quite nicely to a suggestion I'd like
to offer.

Not sure it is possible, but it would sure be nice if a the BOINC manager had
a means of transferring work units from one machine to another. I've got 7
active machines, and such a function would be quite nice when one machine is
loaded up with GPU work, while another machine with GPU work has no work units.
Probably a better board for offering this, so I'll poke around.

Coyote

Wembley
Volunteer tester
Avatar
Send message
Joined: 16 Sep 09
Posts: 415
Credit: 888,257
RAC: 0
United States
Message 1159190 - Posted: 5 Oct 2011, 20:09:46 UTC - in response to Message 1159153.

Or better yet, a work unit proxy server similar to what is available for distributed.net

That way people with multiple systems can set up one server to upload/download work units and all their other systems get work from that one system.

Less workload on the project servers, and no people's crunching machines starve.

____________


Donate with your searches and online buys:
http://www.goodsearch.com/toolbar/university-of-california-setihome

garfield
Send message
Joined: 4 Jan 02
Posts: 40
Credit: 1,728,404
RAC: 10
Austria
Message 1159394 - Posted: 6 Oct 2011, 12:19:33 UTC

Hi all,

the lack of data brought up an interesting discussion with my friends.
There are two things I want to point out here:
The signal from our Pioneer and Voyager Vessels are ~60dB below noise-level. They use a form of noise-reduction to extract the data from the incoming signal. Would it make sense to to the same here?
The other point is: Are there other informations in the data, that could be extracted? As far as I understood it, Astropulse is one form of doing that, but is there anything more we could do?

This could fill up the 'dry zones'.

Alexander
____________

Profile coyote
Avatar
Send message
Joined: 29 Apr 99
Posts: 8
Credit: 44,541,516
RAC: 1,699
United States
Message 1159508 - Posted: 6 Oct 2011, 19:15:49 UTC

Does anyone have an update on the status of GPU jobs for S@H? I now have a good bit of CPU work running, but haven't seen any GPU (cuda) jobs.

Also, want to express my thanks to the Seti@home team for all the fine
work they have done.

Thanks

Coyote

Richard HaselgroveProject donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8497
Credit: 49,873,620
RAC: 51,050
United Kingdom
Message 1159535 - Posted: 6 Oct 2011, 20:29:27 UTC - in response to Message 1159508.

Does anyone have an update on the status of GPU jobs for S@H? I now have a good bit of CPU work running, but haven't seen any GPU (cuda) jobs.

Claggy discovered and reported a problem last night. There is a fix available now, and the plan is to deploy it sometime 'today' - but that's a pretty elastic term, depending on workload and whether any other problems need to to solved first. It doesn't seem to have arrived yet, but there's still time (remember that the project staff are on Pacific time).

Profile coyote
Avatar
Send message
Joined: 29 Apr 99
Posts: 8
Credit: 44,541,516
RAC: 1,699
United States
Message 1159590 - Posted: 6 Oct 2011, 23:36:08 UTC - in response to Message 1159535.

Does anyone have an update on the status of GPU jobs for S@H? I now have a good bit of CPU work running, but haven't seen any GPU (cuda) jobs.

Claggy discovered and reported a problem last night. There is a fix available now, and the plan is to deploy it sometime 'today' - but that's a pretty elastic term, depending on workload and whether any other problems need to to solved first. It doesn't seem to have arrived yet, but there's still time (remember that the project staff are on Pacific time).


Thanks for that info. I do understand completely, as I'm intimately aware
of what can be involved in supporting an application in a high traffic work
flow pattern. My day job involves an awful lot of that :)

Best regards,

Coyote

Profile Brian
Send message
Joined: 9 Jul 11
Posts: 2
Credit: 647,331
RAC: 0
United States
Message 1159863 - Posted: 7 Oct 2011, 18:09:13 UTC - in response to Message 1150111.

I was wondering. Thought my puter deoderant wasn't working :)
____________

Profile coyote
Avatar
Send message
Joined: 29 Apr 99
Posts: 8
Credit: 44,541,516
RAC: 1,699
United States
Message 1159924 - Posted: 7 Oct 2011, 20:42:02 UTC

Are there any further updates, today, on the availability of S@H cuda work units?

Thanks,

Coyote

Richard HaselgroveProject donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8497
Credit: 49,873,620
RAC: 51,050
United Kingdom
Message 1159926 - Posted: 7 Oct 2011, 20:49:24 UTC - in response to Message 1159924.

Are there any further updates, today, on the availability of S@H cuda work units?

Thanks,

Coyote

The server patch has been applied, and the system is capable of allocating work to CUDA cards.

But everything is still very congested: work won't be allocated every time you ask, and any work which is allocated may take a long time to download. But it's all possible.

Previous · 1 · 2 · 3 · 4 · Next

Message boards : News : more data on the way

Copyright © 2014 University of California