SETI@Home will now cache 150 CPU work units and 150 per installed GPU

Message boards : Number crunching : SETI@Home will now cache 150 CPU work units and 150 per installed GPU
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3 · 4 · Next

AuthorMessage
Profile Jimbocous Project Donor
Volunteer tester
Avatar

Send message
Joined: 1 Apr 13
Posts: 1856
Credit: 268,616,081
RAC: 1,349
United States
Message 2022319 - Posted: 8 Dec 2019, 5:45:44 UTC

Given the new limits I backed myself down to 0.25 day + 0.25 additional.
Looks like it will probably settle the Linux boxes at about 150/2150 (CPU/GPU) and 130/1500 and 75/250 for the Win box.
Really needed to back that off, as with 2000 and 3000+ respectively on the linux at 0.50 day + 0.50 additional, boinctasks was starting to see timeouts dealing with the sheer number of tasks.
ID: 2022319 · Report as offensive     Reply Quote
OKA

Send message
Joined: 18 Jun 01
Posts: 17
Credit: 34,397,115
RAC: 131
Norway
Message 2022375 - Posted: 8 Dec 2019, 17:54:00 UTC
Last modified: 8 Dec 2019, 17:54:44 UTC

I suddenly got 900 wu i progress. No metioning of it here. Only me? No ghosts as far as I can tell.
ID: 2022375 · Report as offensive     Reply Quote
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 2022377 - Posted: 8 Dec 2019, 17:57:23 UTC - in response to Message 2022375.  

This was discussed in panic mode thread. Looks like 200 work units for your chip and 400 work units for each GPU you have.
ID: 2022377 · Report as offensive     Reply Quote
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 2022461 - Posted: 9 Dec 2019, 8:38:53 UTC

A definitive statement has been posted on GitHub:

Eric Korpela wrote:
I was the one who increased the limits. I based it on database I/O rates
which, it seemed could handle the change. Long term the rates don't change
but the DB "in process" lookups take longer. If it becomes an issue I'll
drop them.
Tasks 'out in the field' (not sure if that exactly matches 'in process') is still rising slowly, and has reached 6.93 million: it will be interesting to see how that changes as the new working week starts around the world.
ID: 2022461 · Report as offensive     Reply Quote
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 2022473 - Posted: 9 Dec 2019, 13:31:55 UTC

'Results in the field' has now passed 7 million.
ID: 2022473 · Report as offensive     Reply Quote
Profile Mr. Kevvy Crowdfunding Project Donor*Special Project $250 donor
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 3806
Credit: 1,114,826,392
RAC: 3,319
Canada
Message 2022475 - Posted: 9 Dec 2019, 13:47:51 UTC
Last modified: 10 Dec 2019, 1:09:32 UTC

Apologies for all the system e-mails you may have received from post moves into this thread...

For anyone running a "spoofed" client, you will of course now get 300x the number of "spoofed" GPUs which may be an overload so may want to reduce the "spoof" number appropriately, ie to one-third what it currently is. The hard limit of 10K platform tasks remains, so I was seeing just under 10,200 tasks (10K GPU + 200 CPU) on my largest host (which incidentally isn't overdoing it; this is about a day's worth.) All of the others I had to reduce appropriately as noted.
ID: 2022475 · Report as offensive     Reply Quote
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 2022492 - Posted: 9 Dec 2019, 16:46:01 UTC
Last modified: 9 Dec 2019, 16:47:47 UTC

I UL a new spoofed build with a 30000 WU Buffer.
Remember use such big cachê only If you realy have a Very fast system.
And If you use think about use te mutex to make the lag small as posible.
ID: 2022492 · Report as offensive     Reply Quote
Oddbjornik Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 220
Credit: 349,610,548
RAC: 1,728
Norway
Message 2022539 - Posted: 9 Dec 2019, 22:07:15 UTC

Looks like the GPU limit has been lowered again. I'm getting this:
09.12.2019 23.05.42 | SETI@home | Reporting 14 completed tasks
09.12.2019 23.05.42 | SETI@home | Requesting new tasks for CPU and NVIDIA GPU
09.12.2019 23.05.42 | SETI@home | [sched_op] CPU work request: 2123677.81 seconds; 0.00 devices
09.12.2019 23.05.42 | SETI@home | [sched_op] NVIDIA GPU work request: 941881.19 seconds; 0.00 devices
09.12.2019 23.05.46 | SETI@home | Scheduler request completed: got 0 new tasks
09.12.2019 23.05.46 | SETI@home | [sched_op] Server version 709
09.12.2019 23.05.46 | SETI@home | No tasks sent
09.12.2019 23.05.46 | SETI@home | No tasks are available for AstroPulse v7
09.12.2019 23.05.46 | SETI@home | No tasks are available for SETI@home v8
09.12.2019 23.05.46 | SETI@home | Tasks for AMD/ATI GPU are available, but your preferences are set to not accept them
09.12.2019 23.05.46 | SETI@home | Tasks for Intel GPU are available, but your preferences are set to not accept them
09.12.2019 23.05.46 | SETI@home | This computer has reached a limit on tasks in progress
09.12.2019 23.05.46 | SETI@home | Project requested delay of 303 seconds

ID: 2022539 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 2022542 - Posted: 9 Dec 2019, 22:59:33 UTC

Think you are correct. I am not getting a 1:1 replacement for work turned in. Any tasks I get are just replacing cpu tasks turned in. The gpu task caches are falling.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 2022542 · Report as offensive     Reply Quote
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 2022544 - Posted: 9 Dec 2019, 23:08:28 UTC

And RTS has spiked up.
ID: 2022544 · Report as offensive     Reply Quote
Profile Tom M
Volunteer tester

Send message
Joined: 28 Nov 02
Posts: 5126
Credit: 276,046,078
RAC: 462
Message 2022556 - Posted: 10 Dec 2019, 0:17:06 UTC - in response to Message 2022475.  

Apologies for all the system e-mails you may have received from post moves into this thread...


+1

No apology needed. You were doing what a moderator does. Bringing order out of a chaotic new topic.

THANK YOU!

Tom
A proud member of the OFA (Old Farts Association).
ID: 2022556 · Report as offensive     Reply Quote
Profile Mr. Kevvy Crowdfunding Project Donor*Special Project $250 donor
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 3806
Credit: 1,114,826,392
RAC: 3,319
Canada
Message 2022567 - Posted: 10 Dec 2019, 0:33:00 UTC - in response to Message 2022556.  
Last modified: 10 Dec 2019, 1:13:55 UTC

Thanks, Tom. :^)
Edit: Also thanks for the confirmation, Dr. Z. Title and my post updated.
ID: 2022567 · Report as offensive     Reply Quote
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 2022572 - Posted: 10 Dec 2019, 0:49:04 UTC - in response to Message 2022567.  

Yup, looks like they lowered it to 300 per GPU
ID: 2022572 · Report as offensive     Reply Quote
Richard Haselgrove Project Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 14679
Credit: 200,643,578
RAC: 874
United Kingdom
Message 2022654 - Posted: 10 Dec 2019, 22:13:31 UTC

More confirmation via GitHub:

Eric Korpela wrote:
I've dropped the GPU limit to 300 to keep us below 7M. We'll run it there
for a while.
ID: 2022654 · Report as offensive     Reply Quote
Profile MagicEye
Volunteer tester
Avatar

Send message
Joined: 19 Sep 99
Posts: 70
Credit: 40,327,877
RAC: 75
Germany
Message 2022816 - Posted: 11 Dec 2019, 22:50:44 UTC - in response to Message 2022654.  

More confirmation via GitHub:

Eric Korpela wrote:
I've dropped the GPU limit to 300 to keep us below 7M. We'll run it there
for a while.


What kind of "7M" does he mean?
ID: 2022816 · Report as offensive     Reply Quote
Ian&Steve C.
Avatar

Send message
Joined: 28 Sep 99
Posts: 4267
Credit: 1,282,604,591
RAC: 6,640
United States
Message 2022820 - Posted: 11 Dec 2019, 22:54:16 UTC - in response to Message 2022816.  

7M = 7 million = 7,000,000.

Results out in the field.
Seti@Home classic workunits: 29,492 CPU time: 134,419 hours

ID: 2022820 · Report as offensive     Reply Quote
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 36820
Credit: 261,360,520
RAC: 489
Australia
Message 2022821 - Posted: 11 Dec 2019, 22:56:54 UTC - in response to Message 2022816.  

More confirmation via GitHub:

Eric Korpela wrote:
I've dropped the GPU limit to 300 to keep us below 7M. We'll run it there
for a while.


What kind of "7M" does he mean?
What he means is the "results out in the field" and 7M means 7 million. ;-)

Cheers.
ID: 2022821 · Report as offensive     Reply Quote
Speedy
Volunteer tester
Avatar

Send message
Joined: 26 Jun 04
Posts: 1643
Credit: 12,921,799
RAC: 89
New Zealand
Message 2023143 - Posted: 14 Dec 2019, 21:14:12 UTC - in response to Message 2022654.  
Last modified: 14 Dec 2019, 21:15:57 UTC

Eric Korpela wrote:
I've dropped the GPU limit to 300 to keep us below 7M. We'll run it there
for a while.

I wonder if Eric has increased the limits again because we are currently running at 7.27 million.
ID: 2023143 · Report as offensive     Reply Quote
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 36820
Credit: 261,360,520
RAC: 489
Australia
Message 2023144 - Posted: 14 Dec 2019, 21:17:22 UTC

I wonder if Eric has increased the limits again because we are currently running at 7.27 million.
No changes, it's still 200 for CPU and 300 for GPU's.

Cheers.
ID: 2023144 · Report as offensive     Reply Quote
Profile Keith Myers Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 29 Apr 01
Posts: 13164
Credit: 1,160,866,277
RAC: 1,873
United States
Message 2023147 - Posted: 14 Dec 2019, 21:27:06 UTC - in response to Message 2023144.  

The infrequent crunchers or ones that only contact the project every ten days are getting filled up at the new cache levels.
Seti@Home classic workunits:20,676 CPU time:74,226 hours

A proud member of the OFA (Old Farts Association)
ID: 2023147 · Report as offensive     Reply Quote
Previous · 1 · 2 · 3 · 4 · Next

Message boards : Number crunching : SETI@Home will now cache 150 CPU work units and 150 per installed GPU


 
©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.