SETI Hardware Fundraisers (With Big News Inside)

Message boards : Number crunching : SETI Hardware Fundraisers (With Big News Inside)
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · 5 · Next

AuthorMessage
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1296498 - Posted: 18 Oct 2012, 11:30:23 UTC - in response to Message 1293645.  

We recently received a planning list of projects from SETI. This will keep us busy for likely 6-9 months.

Now all we're waiting on are our beginning specs and we'll get rolling on new projects.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1296498 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1297996 - Posted: 22 Oct 2012, 22:38:08 UTC - in response to Message 1296498.  

As a small update for folks waiting on AP tasks:

Marvin's AP array ran out of disk space. Jeff and Eric have swapped out Marvin's 750G drives for some of the donated GPUUG 2TB drives (our spares pool) to fix the issue.

Not sure when we'll see more AP tasks, but at least we know where the bork was.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1297996 · Report as offensive
Profile Ronald R CODNEY
Avatar

Send message
Joined: 19 Nov 11
Posts: 87
Credit: 420,920
RAC: 0
United States
Message 1297998 - Posted: 22 Oct 2012, 22:48:36 UTC

Thanks Brad for the heads up on the AP issue. Lookin' forward to the reemergence.
ID: 1297998 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1299158 - Posted: 26 Oct 2012, 23:32:00 UTC - in response to Message 1297998.  
Last modified: 26 Oct 2012, 23:32:42 UTC

Quick bit of news:

1. A check for $2000 from our SETI Direct donations was delivered to S&H on Thursday. I'm always proud to be able to cut such nice checks to the organization thanks to our donors.

2. We had a member donate today a 3TB hard drive to the project. Newegg was running a special so instead of our listed price, the drives are on special for $119. As a result I used some of our general funds to purchase 2 instead of the one paid for drive which will give SETI QTY 2, 3TB drives with which to use for AP Reob data.

You can track this purchase by visiting http://www.gpuug.org/catalog

That's all I have for now. We're still waiting on specs from the lab before starting our projects.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1299158 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1299418 - Posted: 27 Oct 2012, 15:57:01 UTC - in response to Message 1299158.  

And another 2 3TB drives donated by A.J. Ordered today, should ship all 4 drives out on Monday.

I'll update http://www.gpuug.org/purchases with tracking once I receive it from Newegg.

Our donors are awesome.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1299418 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1300566 - Posted: 30 Oct 2012, 20:50:13 UTC - in response to Message 1299418.  

Project Update:

Marvin was fitted with a new RAID 10 using 8 of our donated 2TB Seagate drives.

According to Eric these drives were 'really clean, no reallocated sectors on any of them.' This makes us think Seagate has gotten past its first of the year QC issues and is now shipping proper hardware again.

This also means AP should be back up and running like normal.

Lastly our 4 donated 3TB drives should hit the Lab tomorrow.

Thanks everyone.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1300566 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1303724 - Posted: 8 Nov 2012, 21:14:49 UTC - in response to Message 1300566.  

New big project coming as soon as we tie up things on the GPUUG end. More in the coming days, stay tuned!


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1303724 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1303897 - Posted: 9 Nov 2012, 7:20:02 UTC - in response to Message 1303724.  

New big project coming as soon as we tie up things on the GPUUG end. More in the coming days, stay tuned!

I can only hope it sorts out the Scheduler issue. Then of course we'll need to get he bandwidth problem sorted. Then of course we'll find another bottleneck...
Grant
Darwin NT
ID: 1303897 · Report as offensive
Profile James Sotherden
Avatar

Send message
Joined: 16 May 99
Posts: 10436
Credit: 110,373,059
RAC: 54
United States
Message 1304086 - Posted: 9 Nov 2012, 17:54:11 UTC - in response to Message 1303897.  

New big project coming as soon as we tie up things on the GPUUG end. More in the coming days, stay tuned!

I can only hope it sorts out the Scheduler issue. Then of course we'll need to get he bandwidth problem sorted. Then of course we'll find another bottleneck...


I agree, It seems maybe we are throwing hardware at UC Berkley being stubborn about a bandwidth problem.

I will donate what I can, But id like to hear some news on whats going on to.
[/quote]

Old James
ID: 1304086 · Report as offensive
MarkJ Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 08
Posts: 1139
Credit: 80,854,192
RAC: 5
Australia
Message 1304263 - Posted: 10 Nov 2012, 2:22:34 UTC

There was mention in a previous post by Slavac about getting a hardware load balancer, so that may help things a bit.

My understanding was the project staff are trying to get the v7 app out the door. This would take longer to process and therefore reduce the work requests on the the server, at least until hardware and software improvements negate the extra processing time.
BOINC blog
ID: 1304263 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13720
Credit: 208,696,464
RAC: 304
Australia
Message 1304294 - Posted: 10 Nov 2012, 4:44:25 UTC - in response to Message 1304263.  

There was mention in a previous post by Slavac about getting a hardware load balancer, so that may help things a bit.

It might help with the network traffic to some extent, but the big issue for the last few weeks has been the Scheduler getting tied up in knots.

Grant
Darwin NT
ID: 1304294 · Report as offensive
MarkJ Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 08
Posts: 1139
Credit: 80,854,192
RAC: 5
Australia
Message 1304317 - Posted: 10 Nov 2012, 6:34:14 UTC - in response to Message 1304294.  

There was mention in a previous post by Slavac about getting a hardware load balancer, so that may help things a bit.

It might help with the network traffic to some extent, but the big issue for the last few weeks has been the Scheduler getting tied up in knots.


Agreed it won't help the scheduler, but should help the download server "balancing".

@Slavac, any news on the load balancer mentioned previously? Haven't seen a fund-raiser for it specifically.
BOINC blog
ID: 1304317 · Report as offensive
Filipe

Send message
Joined: 12 Aug 00
Posts: 218
Credit: 21,281,677
RAC: 20
Portugal
Message 1304457 - Posted: 10 Nov 2012, 13:38:14 UTC

My understanding was the project staff are trying to get the v7 app out the door. This would take longer to process and therefore reduce the work requests on the the server, at least until hardware and software improvements negate the extra processing time.


What is the new science in the v7 app that would make the wu take longer to process?
ID: 1304457 · Report as offensive
Profile Mike Special Project $75 donor
Volunteer tester
Avatar

Send message
Joined: 17 Feb 01
Posts: 34253
Credit: 79,922,639
RAC: 80
Germany
Message 1304675 - Posted: 10 Nov 2012, 22:48:18 UTC - in response to Message 1304457.  

My understanding was the project staff are trying to get the v7 app out the door. This would take longer to process and therefore reduce the work requests on the the server, at least until hardware and software improvements negate the extra processing time.


What is the new science in the v7 app that would make the wu take longer to process?


Its searching for another type of signal called autoCorrelation.



With each crime and every kindness we birth our future.
ID: 1304675 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11360
Credit: 29,581,041
RAC: 66
United States
Message 1304716 - Posted: 11 Nov 2012, 0:13:32 UTC - in response to Message 1304675.  

IIRC autocorrelation is a noise filtering technique, that should enable us to detect weaker signals hidden in the background.
ID: 1304716 · Report as offensive
Sakletare
Avatar

Send message
Joined: 18 May 99
Posts: 132
Credit: 23,423,829
RAC: 0
Sweden
Message 1305112 - Posted: 11 Nov 2012, 21:58:39 UTC - in response to Message 1304716.  

IIRC autocorrelation is a noise filtering technique, that should enable us to detect weaker signals hidden in the background.

That is correct. http://en.wikipedia.org/wiki/Autocorrelation
ID: 1305112 · Report as offensive
Filipe

Send message
Joined: 12 Aug 00
Posts: 218
Credit: 21,281,677
RAC: 20
Portugal
Message 1305134 - Posted: 11 Nov 2012, 23:03:05 UTC

IIRC autocorrelation is a noise filtering technique, that should enable us to detect weaker signals hidden in the background.


Nice! How much time it will increase the processing time per work unit?
ID: 1305134 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1305319 - Posted: 12 Nov 2012, 10:09:07 UTC - in response to Message 1304317.  
Last modified: 12 Nov 2012, 10:09:31 UTC

There was mention in a previous post by Slavac about getting a hardware load balancer, so that may help things a bit.

It might help with the network traffic to some extent, but the big issue for the last few weeks has been the Scheduler getting tied up in knots.


Agreed it won't help the scheduler, but should help the download server "balancing".

@Slavac, any news on the load balancer mentioned previously? Haven't seen a fund-raiser for it specifically.


Right now the plan is to wait on the load balancer after much discussion with the lab. The current plan (in order of likely priority which is subject to change)

1. Compute Node for the GBT. This will be a part of the high speed spectrometer.
1. A new upload server and download server. (Tossup as to priority with above. We're still chewing this one over with the lab)
2. Media converters for the lab. These will necessarily be funded, installed and tested before we reevaluate the load balancer. The hopes here are that the converters increase the much needed internal bandwidth of the project.

My plan right now is to attempt to get the balancer donated by the manuf. vs fundraising but again this is dependent on sorting out the media converters above. We'll likely slot the converters into priority 1 status once the specs from SSL come in.

______

As far as news on what's wrong with the project currently I'm sadly in the dark.

Also I should note we had another 3TB transport drive donated today which I'll get ordered this week.

Lastly I hope to start a new project this week, likely one of the above large ones. I'll follow that up with a new thread with the relevant data, specs and so on.

Thanks all.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1305319 · Report as offensive
Profile Slavac
Volunteer tester
Avatar

Send message
Joined: 27 Apr 11
Posts: 1932
Credit: 17,952,639
RAC: 0
United States
Message 1305321 - Posted: 12 Nov 2012, 10:13:46 UTC - in response to Message 1304086.  

New big project coming as soon as we tie up things on the GPUUG end. More in the coming days, stay tuned!

I can only hope it sorts out the Scheduler issue. Then of course we'll need to get he bandwidth problem sorted. Then of course we'll find another bottleneck...


I agree, It seems maybe we are throwing hardware at UC Berkley being stubborn about a bandwidth problem.

I will donate what I can, But id like to hear some news on whats going on to.


Bandwidth is something we're trying to work on (GPUUG - Lab - UCB) though the process is quite Sysyphean. I wish I could throw donor money at the problem and solve it but sadly that's out of my hands unless we do something creative (and expensive).

I'll ask Eric and or Jeff to let us know what's up in a tech news post. Keep in mind that Dan's in China working on a very sexy SETI project and Matt's entertaining the world, so the lab's quite understaffed and slammed with these issues we've been fighting.


Executive Director GPU Users Group Inc. -
brad@gpuug.org
ID: 1305321 · Report as offensive
Sakletare
Avatar

Send message
Joined: 18 May 99
Posts: 132
Credit: 23,423,829
RAC: 0
Sweden
Message 1305336 - Posted: 12 Nov 2012, 11:53:16 UTC - in response to Message 1305321.  

Bandwidth is something we're trying to work on (GPUUG - Lab - UCB) though the process is quite Sysyphean. I wish I could throw donor money at the problem and solve it but sadly that's out of my hands unless we do something creative (and expensive).

One way would be to get something like an ISP to donate bandwidth. We can't get that bandwidth to the lab, but we could place a proxy/workunit storage at the ISP.

I'm not sure the BOINC back-end could handle that. But if it could we could essentially halve the labs bandwidth requirement since we would only have to transfer an individual workunit once.
ID: 1305336 · Report as offensive
Previous · 1 · 2 · 3 · 4 · 5 · Next

Message boards : Number crunching : SETI Hardware Fundraisers (With Big News Inside)


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.