Upwards and Onwards (May 28 2009)


log in

Advanced search

Message boards : Technical News : Upwards and Onwards (May 28 2009)

1 · 2 · 3 · Next
Author Message
Profile Matt Lebofsky
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar
Send message
Joined: 1 Mar 99
Posts: 1389
Credit: 74,079
RAC: 0
United States
Message 900643 - Posted: 28 May 2009, 20:37:47 UTC
Last modified: 28 May 2009, 20:37:57 UTC

Question: so what's up with the near time persistency checker (NTPCkr)? If the live web streaming were working last Thursday you would have seen the tail end of my and Jeff's talk where Jeff went into a little details about the current status of things. Basically, we have some screws to tighten here and there, but the general thing is working. We're up against some database throughput issues which we hope to fix sooner than later, plus we are still tweaking the scoring algorithms. We hope to have a public page available soon where you can peer into the progress of things. Until then, here's version 0.0.1 of the NTPCkr FAQ.

It's becoming clearer that we need to adjust the weight of our applications so that we send out more SETI@home/multibeam workunits. We have things effectively set such that Astropulse work gets sent out as soon as it becomes available. This was partly to expedite getting as many Astropulse results back as possible (in the interest of getting that science done) but this is getting less and less possible given our resources and current participant demands. Things on this front may shift in the near future.

We've been near our bandwidth limit for the past day since unclogging the mysql database, providing more data for Astropulse to split, and our active user base going up about 15% over the past couple of weeks. This may account for recent upload/download difficulty. It looks like it's getting better, as least for the moment.

- Matt
____________
-- BOINC/SETI@home network/web/science/development person
-- "Any idiot can have a good idea. What is hard is to do it." - Jeanne-Claude

Profile perryjay
Volunteer tester
Avatar
Send message
Joined: 20 Aug 02
Posts: 3377
Credit: 15,916,225
RAC: 11,736
United States
Message 900654 - Posted: 28 May 2009, 21:09:16 UTC - in response to Message 900643.
Last modified: 28 May 2009, 21:10:56 UTC

My little Celeron has been getting a nice mix of 6.03s and 6.08s with just enough APs to keep things interesting here lately. Sure would be nice if it keeps it up.

Also good to see the replica database finally catching up and staying current. It got tiring trying to explain to the new guys why they weren't seeing any credit for their work. That extra 15% we've picked up must be the ones that have been asking. :)

Have you given any thought to how to keep new people from getting APs for their first work? It seems like we lose a bunch of them when they see the time to completion on those for the first time. I've had a few wingmen that don't even come in to ask, they just disappear. At least those that do visit the site get answers they can live with and stick with it.

I forgot to say thank you. We all know how hard you guys work behind the scenes to keep us going.
____________


PROUD MEMBER OF Team Starfire World BOINC

Profile Johnney Guinness
Volunteer tester
Avatar
Send message
Joined: 11 Sep 06
Posts: 3092
Credit: 2,607,618
RAC: 2,345
Ireland
Message 900661 - Posted: 28 May 2009, 21:24:15 UTC
Last modified: 28 May 2009, 21:29:48 UTC

The NitPicker FAQ looks good, but its a bit like showing a child your bar of chocolate and saying you can eat some of this chocolate SOON!, just not right now!

Give us a sneak preview Matt, I'm dying to get a taste of the real thing :)

The live web streaming was NOT working last Thursday! We could not see it and its not avalable for download.

John.
____________

Richard HaselgroveProject donor
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8631
Credit: 51,490,218
RAC: 48,405
United Kingdom
Message 900891 - Posted: 29 May 2009, 9:07:05 UTC - in response to Message 900875.

I must confess......
The kitties will be a bit upset if you back off on AP availability...they have barely been able to keep the caches full for the last week or so.

I was under the impression that AP, due to it's longer crunch times, was actually easier on the Seti servers than having a bunch of Cuda monsters pounding on the door for more work constantly....

Now you want to increase the availability of MB at the expense of AP work????

I am sorry.....I do not quite understand.

Mark, have you looked at the server status page recently?

There are 130 'tapes' listed there: Astropulse has 'done' 128 of them, so they're just sitting there, waiting to be split for MB.

That's something like six terabytes of raw data, being kept online simply because the AP and MB split/issue rates are out of balance.

Although raw storage hasn't been such a problem for the project since the Overland Storage donation, that contribution was 'only' ten terabytes: using over half of that just to keep the kitties supplied with their particular flavour of kibble seems just a touch excessive!

Seriously, wouldn't it be better for the project as a whole if both types of work were consistently available, and split at roughly the same speed?

Profile MarkJProject donor
Volunteer tester
Avatar
Send message
Joined: 17 Feb 08
Posts: 941
Credit: 25,098,568
RAC: 29,258
Australia
Message 900892 - Posted: 29 May 2009, 9:09:58 UTC - in response to Message 900654.

My little Celeron has been getting a nice mix of 6.03s and 6.08s with just enough APs to keep things interesting here lately. Sure would be nice if it keeps it up.


I did suggest to Eric (and I believe Richard H has as well) that all the VLAR work units get sent only as 6.03 (ie in response to a CPU request only). Not only does it give the cpu's some work, it also provides a work around for the amount of time they take a gpu to process.

Unfortunately all my cpus regularly have to resort to other projects to keep themselves busy.

Thanks guys for all the hard work.
____________
BOINC blog

Sten-Arne
Volunteer tester
Send message
Joined: 1 Nov 08
Posts: 3514
Credit: 20,699,508
RAC: 23,060
Sweden
Message 900900 - Posted: 29 May 2009, 9:39:00 UTC - in response to Message 900891.

Mark, have you looked at the server status page recently?

There are 130 'tapes' listed there: Astropulse has 'done' 128 of them, so they're just sitting there, waiting to be split for MB.

That's something like six terabytes of raw data, being kept online simply because the AP and MB split/issue rates are out of balance.

Although raw storage hasn't been such a problem for the project since the Overland Storage donation, that contribution was 'only' ten terabytes: using over half of that just to keep the kitties supplied with their particular flavour of kibble seems just a touch excessive!

Seriously, wouldn't it be better for the project as a whole if both types of work were consistently available, and split at roughly the same speed?


Looks to me that instead of complaining about lack of AP work. The solution would be for AP loving people, to disable AP in their preferences for a while. In that way they would help greatly to reduce the huge amount of MB's available, and also help themselves to get more AP's much faster.

The huge amount of MB WU's seems to indicate that too many participants have instead disabled MB WU's in their preferences, and thereby helped create the problems they now complain about.

Ok I know AP pays better, but hey this is about science in the first place, and "credits" in the second place.

Sten-Arne

EPG
Send message
Joined: 3 Apr 99
Posts: 110
Credit: 10,405,863
RAC: 0
Hungary
Message 900908 - Posted: 29 May 2009, 11:05:32 UTC - in response to Message 900900.


...The solution would be for AP loving people, to disable AP in their preferences for a while. ...
Sten-Arne


They don't have to disable AP or enable MB, just allow other work if there is no AP and put the MB part into app_info(if they using opp. app.). That way they will only ask AP, get it if there is any in the queue, and in the "worst case" they get MB. And no need for manual switching all the time.
____________

Ianab
Volunteer tester
Send message
Joined: 11 Jun 08
Posts: 673
Credit: 12,623,886
RAC: 5,946
New Zealand
Message 900910 - Posted: 29 May 2009, 11:21:23 UTC - in response to Message 900900.

I'm in the "Wanting More AP" camp....

I would suggest that the balance over the last month has been to give out AP to anyone requesting work, wether they want it, or can even process it in time.

I like AP because it pays better, and the PCs that I have selected to run it...well they can do it.

A bit of priority to the folks that WANT AP and give the MB to the folks running standard seti, probably keep more folks happy?

Ian

WinterKnight
Volunteer tester
Send message
Joined: 18 May 99
Posts: 8683
Credit: 24,957,972
RAC: 29,134
United Kingdom
Message 900938 - Posted: 29 May 2009, 13:47:07 UTC

As in the few AP tasks I have there are several _6's, I do think, even though I agree we should be doing both applications, that the time has come for AP to be user selected only.
Sending out 7 copies at 8 MByte each cannot be helping with the bandwidth or database problems, and being user selected only should almost eliminate this problem.
It should also help balance the AP/MB splitting as all computers are capable of processing MB tasks but some are marginal at processing AP tasks within the deadline.

Profile Virtual Boss*
Volunteer tester
Avatar
Send message
Joined: 4 May 08
Posts: 417
Credit: 6,198,796
RAC: 468
Australia
Message 900941 - Posted: 29 May 2009, 13:52:31 UTC - in response to Message 900938.

As in the few AP tasks I have there are several _6's, I do think, even though I agree we should be doing both applications, that the time has come for AP to be user selected only.
Sending out 7 copies at 8 MByte each cannot be helping with the bandwidth or database problems, and being user selected only should almost eliminate this problem.
It should also help balance the AP/MB splitting as all computers are capable of processing MB tasks but some are marginal at processing AP tasks within the deadline.


I'll second that.

Dena Wiltsie
Send message
Joined: 19 Apr 01
Posts: 1174
Credit: 557,760
RAC: 327
United States
Message 900943 - Posted: 29 May 2009, 13:55:57 UTC

I on the other hand am unable to process AP because it takes a special application on Apple products. When will a standard application become available so I can join the fun? ;)
____________

Profile Bob Mahoney Design
Avatar
Send message
Joined: 4 Apr 04
Posts: 178
Credit: 9,205,632
RAC: 0
United States
Message 900973 - Posted: 29 May 2009, 15:34:04 UTC - in response to Message 900938.

As in the few AP tasks I have there are several _6's, I do think, even though I agree we should be doing both applications, that the time has come for AP to be user selected only.
Sending out 7 copies at 8 MByte each cannot be helping with the bandwidth or database problems, and being user selected only should almost eliminate this problem.
It should also help balance the AP/MB splitting as all computers are capable of processing MB tasks but some are marginal at processing AP tasks within the deadline.

I second that also.

I just completed a _8 AP task. It was rattling around the world since February.

Perhaps we do not need to go as strict as 'user-selected only', but some form of 'go/no-go for AP' based on host system power would be helpful. This would be SETI server based, not BOINC based. If that is not possible, 'user-selected only' makes sense.

Bob
____________
Opinion stated as fact? Who, me?

Profile perryjay
Volunteer tester
Avatar
Send message
Joined: 20 Aug 02
Posts: 3377
Credit: 15,916,225
RAC: 11,736
United States
Message 900980 - Posted: 29 May 2009, 15:43:23 UTC - in response to Message 900973.

User selected makes sense to me. I'm seeing way too many new people with zero credit/ zero RAC timing out with nothing but APs on their machines. I think it would be better to send only MBs to the new people until they get the hang of this game and then let them ask about or just try out the APs when they find them.
____________


PROUD MEMBER OF Team Starfire World BOINC

Profile James Sotherden
Avatar
Send message
Joined: 16 May 99
Posts: 8907
Credit: 35,867,600
RAC: 43,367
United States
Message 901008 - Posted: 29 May 2009, 16:30:37 UTC

I agree with Perry on this. when i ran AP on my old P4 to many times i got slow wingmen or they just aborted. It was bad enough to crunch for 8 days then wait for 2 more months to get it validated.
i remember my first AP i aborted it to. Then i started reading the forums and learned what was up.


____________

Old James

Andy Williams
Volunteer tester
Avatar
Send message
Joined: 11 May 01
Posts: 187
Credit: 112,464,820
RAC: 0
United States
Message 901038 - Posted: 29 May 2009, 17:07:09 UTC - in response to Message 900973.

As in the few AP tasks I have there are several _6's, I do think, even though I agree we should be doing both applications, that the time has come for AP to be user selected only.


I second that also.

I just completed a _8 AP task. It was rattling around the world since February.


I third, fourth and fifth that. My rule of thumb for a new, modern, multi-core machine is that it takes three months of steadily doing APs before they begin to validate as fast as they are produced.

It can't possibly be in the best interests of the project for so many newbies to be aborting APs. I suspect many of them never return.
____________
--
Classic 82353 WU / 400979 h

HarryM
Volunteer tester
Send message
Joined: 24 Jul 08
Posts: 68
Credit: 1,826,327
RAC: 0
United States
Message 901047 - Posted: 29 May 2009, 17:17:01 UTC
Last modified: 29 May 2009, 17:18:02 UTC

You can uncheck both "AP's" in the "Preferences" section. Be sure to also uncheck the "If no work for selected applications is available, accept work from other applications?" section also.

I would call this "User selectable".
____________

Profile perryjay
Volunteer tester
Avatar
Send message
Joined: 20 Aug 02
Posts: 3377
Credit: 15,916,225
RAC: 11,736
United States
Message 901051 - Posted: 29 May 2009, 17:22:05 UTC - in response to Message 901047.

That is user selectable but in this case we are talking about a new cruncher that receives APs for the first WUs they get from the project. The time to completion is usually way over-estimated and on seeing that the new guy runs for the hills leaving the work behind to time out.
____________


PROUD MEMBER OF Team Starfire World BOINC

Profile KWSN THE Holy Hand Grenade!
Volunteer tester
Avatar
Send message
Joined: 20 Dec 05
Posts: 1959
Credit: 10,458,828
RAC: 9,848
United States
Message 901055 - Posted: 29 May 2009, 17:30:05 UTC - in response to Message 901047.
Last modified: 29 May 2009, 17:34:02 UTC

You can uncheck both "AP's" in the "Preferences" section. Be sure to also uncheck the "If no work for selected applications is available, accept work from other applications?" section also.

I would call this "User selectable".


– but both are checked by default, so AP is enabled for a new user. (and probably the first thing they get!) What is being suggested, here, is that they be un-checked by default, so that a new user only gets multi-beam, until he/she "discovers" AP.
____________
.

HarryM
Volunteer tester
Send message
Joined: 24 Jul 08
Posts: 68
Credit: 1,826,327
RAC: 0
United States
Message 901057 - Posted: 29 May 2009, 17:33:00 UTC

I understand now.
____________

1 · 2 · 3 · Next

Message boards : Technical News : Upwards and Onwards (May 28 2009)

Copyright © 2014 University of California