AP WUs?

Message boards : Number crunching : AP WUs?
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Bill Collins

Send message
Joined: 5 Nov 05
Posts: 25
Credit: 57,544,918
RAC: 0
United States
Message 1509077 - Posted: 27 Apr 2014, 3:11:32 UTC

Are there any more AP WUs coming out of the hopper soon? It's feast or famine.

- Bill
-Bill
ID: 1509077 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1509097 - Posted: 27 Apr 2014, 6:10:33 UTC

Once the current batch of tapes are done being split into MB tasks there will be.
See: http://setiathome.berkeley.edu/sah_status.html
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1509097 · Report as offensive
Thomas
Volunteer tester

Send message
Joined: 9 Dec 11
Posts: 1499
Credit: 1,345,576
RAC: 0
France
Message 1509118 - Posted: 27 Apr 2014, 7:33:50 UTC

You can read this, Bill, posted by Rob Smith >> AP Availability
ID: 1509118 · Report as offensive
Sleepy
Volunteer tester
Avatar

Send message
Joined: 21 May 99
Posts: 219
Credit: 98,947,784
RAC: 28,360
Italy
Message 1509533 - Posted: 28 Apr 2014, 16:06:04 UTC - in response to Message 1509118.  

Should be sooner than later now, anyway, judging by the server status.
MBs are drying out on the server and soon new tapes will have to be loaded on the system.

Better not too many rather than 500+ channels to avoid long periods without APs.
The more powerful your system, the lower number of tapes to be loaded at any time you would prefer (however neglecting possible problems of network congestion and AP WU production during the reloading periods by the clients).

From my observations, 300-350 channels at a time would best suit midrange crunchers as myself.
Bigger crunchers would prefer maybe 150 at a time (but I imagine this would require more work by the people at the lab).
Slower crunchers can do well with 500+ at a time.

Sleepy
ID: 1509533 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11367
Credit: 29,581,041
RAC: 66
United States
Message 1509607 - Posted: 28 Apr 2014, 17:45:24 UTC

And now we are in a major famine.
ID: 1509607 · Report as offensive
Bill Collins

Send message
Joined: 5 Nov 05
Posts: 25
Credit: 57,544,918
RAC: 0
United States
Message 1509734 - Posted: 29 Apr 2014, 3:05:53 UTC - in response to Message 1509118.  

Thanks Thomas and Hal. I thought it would be something like that. I've been trying to maximize my SETI productivity by gearing up with a new GTX670 and some other stuff in the quest to find ET. So I'm very much aware of the loss of AP units as they are direct indicators of what my machines can do when using the Lunatics optimized crunching applications. In trying to step up one of my machines I upgraded the power supply, but it didn't have a molex to ATX connector on it so I'm down a small cruncher until later this week when the connector comes in the mail.

Maybe by then we'll have some more APs to crunch.

- Bill
-Bill
ID: 1509734 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1509742 - Posted: 29 Apr 2014, 3:56:11 UTC

With only a few observations it looks like we burn through enough MB tasks for 20-30 channels to be processes a day. Right now there are around 160 more channels to do. So if no new data was loaded then we still have about a weeks worth of data before the tape run out. They may load new data to be split before then, but they may not.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1509742 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1509790 - Posted: 29 Apr 2014, 6:35:28 UTC

I guess this is probably about the only good thing about using CPU only to crunch APs... a 10-day cache is ~60 APs for me. I've only run dry I think twice since building this machine two years + 3 months ago. I still have about 6 days of cache left. :p
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1509790 · Report as offensive
Profile Gonad the Destroyer®©™
Avatar

Send message
Joined: 6 Aug 99
Posts: 204
Credit: 12,463,705
RAC: 0
United States
Message 1509921 - Posted: 29 Apr 2014, 13:00:21 UTC

It would be nice if selecting a 10 day cache, you would actually get a 10 day cache....
ID: 1509921 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1510044 - Posted: 29 Apr 2014, 21:43:30 UTC - in response to Message 1509921.  

It would be nice if selecting a 10 day cache, you would actually get a 10 day cache....

Well BOINC will fill a 10 day cache if you have enough other projects that do not have task limits. As I run almost only CPU & have a lot of old hardware the 100 task limit is nearly a 10 day cache on most of my machines. Several of my machines that I allow AP work still have 20-30 AP to do. While my fastest only has 9 left. So the RAC on that system is a submarine... "Prepare to dive!" lol
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1510044 · Report as offensive
Profile ivan
Volunteer tester
Avatar

Send message
Joined: 5 Mar 01
Posts: 783
Credit: 348,560,338
RAC: 223
United Kingdom
Message 1510057 - Posted: 29 Apr 2014, 22:12:46 UTC

I think they're starting up again...
ID: 1510057 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11367
Credit: 29,581,041
RAC: 66
United States
Message 1510059 - Posted: 29 Apr 2014, 22:20:53 UTC - in response to Message 1510057.  

Yes, and I just got 5 on 1 machine.
ID: 1510059 · Report as offensive
Profile ivan
Volunteer tester
Avatar

Send message
Joined: 5 Mar 01
Posts: 783
Credit: 348,560,338
RAC: 223
United Kingdom
Message 1510065 - Posted: 29 Apr 2014, 22:28:24 UTC - in response to Message 1510059.  

Yes, and I just got 5 on 1 machine.

Yeah, I hadn't scrolled down far enough on the status page to see that a new "tape" had been mounted just before the hour; I was going on having just received two WUs, and a nascent uptick on the Cricket graph. I've now got four crunching on my two GPUs (along with 7 CPU jobs) so I've got 100 mins or so to get enough more to continue crunching when these are done. :-)
ID: 1510065 · Report as offensive
Cosmic_Ocean
Avatar

Send message
Joined: 23 Dec 00
Posts: 3027
Credit: 13,516,867
RAC: 13
United States
Message 1510157 - Posted: 30 Apr 2014, 4:48:40 UTC
Last modified: 30 Apr 2014, 4:49:23 UTC

Yeah. Full 10-day cache is ~60 WUs for me. I have tried once or twice before to bring my max CPUs up to 100% (normally at 50%) to stockpile a few more days worth (and then drop back down to 50%). In fact.. I did that before the move to to co-lo. I was sitting on 19 days of AP cache and the downtime only ended up being like..two days, maybe 3. That was not the expected result by basically anyone.

I haven't done that since though. Seems kind of like cheating/hoarding even though it really isn't.

Anyway.. I acquired 38 APs to top the 10-day cache off again. Filled to the brim.
Linux laptop:
record uptime: 1511d 20h 19m (ended due to the power brick giving-up)
ID: 1510157 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11367
Credit: 29,581,041
RAC: 66
United States
Message 1510528 - Posted: 1 May 2014, 0:25:40 UTC - in response to Message 1510157.  

Filled to the brim.

That creates a good feeling.
ID: 1510528 · Report as offensive
Profile Gonad the Destroyer®©™
Avatar

Send message
Joined: 6 Aug 99
Posts: 204
Credit: 12,463,705
RAC: 0
United States
Message 1511766 - Posted: 3 May 2014, 13:52:33 UTC

Aaaaaaand, there gone.....:(
ID: 1511766 · Report as offensive
Bill Collins

Send message
Joined: 5 Nov 05
Posts: 25
Credit: 57,544,918
RAC: 0
United States
Message 1512635 - Posted: 5 May 2014, 23:34:19 UTC - in response to Message 1511766.  

But they're back again!

-Bill
-Bill
ID: 1512635 · Report as offensive
David S
Volunteer tester
Avatar

Send message
Joined: 4 Oct 99
Posts: 18352
Credit: 27,761,924
RAC: 12
United States
Message 1513190 - Posted: 7 May 2014, 13:54:05 UTC - in response to Message 1512635.  

But they're back again!

-Bill

And they're gone again.

Splitting is finished, so when the current pile of them runs dry, that's it until 390 channels of MB finish.
David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

ID: 1513190 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1514873 - Posted: 12 May 2014, 2:24:58 UTC - in response to Message 1514864.  

Anybody else experiencing longer than usual crunch times on this batch of APs due possibly to high blanking?

Bad timing for me because I'm trying to determine if a blown power supply (SST-ST1500!) hurt a couple of my GTX780ti's.

All of the APs I have looked at so far are all 0.00 blanking. However I didn't really check out the pulse count on them
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1514873 · Report as offensive
1 · 2 · Next

Message boards : Number crunching : AP WUs?


 
©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.