Blip (Sep 16 2009) |
![]() |
| log in |
Message boards : Technical News : Blip (Sep 16 2009)
1 · 2 · 3 · Next
| Author | Message |
|---|---|
|
Hello again. Sorry about the lack of information lately. I was out sick a large chunk of last week. | |
| ID: 933842 · | |
Hello again. Welcome back. From some of the postings, looks like you were missed. Hope you're recharged! ... was a donated system of unknown quality. Several hours of fighting with it yesterday makes me believe mindy may be a dud (processor errors during boot, etc.). Donated kit is only as free as the time it takes to get it running and to keep it running... And also the cost of a system soak test! ... NTPCkr data will start making slightly more sense (the top candidates are all pretty much junk right now). Until then, I will only upload the top ten list by hand every so often. The examples so far have been a good test run. Here's hoping for the clean and automated updates... Happy crunchin', Martin ____________ Mandriva Linux A user friendly OS! See new freedom Mageia2 The Future is what We make IT (GPLv3) | |
| ID: 933857 · | |
|
Welcome back Matt, hope you are feeling better. | |
| ID: 933908 · | |
|
Pleased to hear from you again Matt | |
| ID: 933940 · | |
... Hmm, the Fall 2009 observation schedule for the A2010 ALFALFA project notes: ALFA will be taken down from Sep 14 until about Nov 5, 2009 for servicing. The aoschedule pages seem to agree except it might get back into operation at the end of October. How's the software blanking coming along? Participants would be more comfortable if you could fill at least part of the gap with data recalled from NERSC HPSS. Joe | |
| ID: 933956 · | |
|
Hi Matt, | |
| ID: 933972 · | |
|
Sorry, I didn't realize that the switch hadn't been implemented. I'll figure out how to do so. | |
| ID: 934085 · | |
Sorry, I didn't realize that the switch hadn't been implemented. I'll figure out how to do so. Any Progress in getting the Astropulse_v505 switch implemented? Claggy | |
| ID: 938403 · | |
|
Hello Eric, | |
| ID: 939338 · | |
Hello Eric, There have been fundamental changes in how the Astropulse application processes and reports results. Those make it impossible to validate the results between the newer and older apps. It's the same kind of difference as occurred between "setiathome" and "setiathome_enhanced". Joe | |
| ID: 939482 · | |
|
Question: | |
| ID: 939866 · | |
Question: Astropulse v505 has been active for ~4 months. Although there are probably very few AP v5 tasks not yet resolved, given 1 month deadlines I'd be surprised if there are none. The obvious "fix" to the selection switches is to have them depend on which applications are not deprecated, like the Applications page. "Astropulse" was deprecated months ago so it disappeared from that page, and that also removed it from the applications for which the Feeder will pass tasks to the Scheduler. Similarly when "Astropulse v505" was added it appeared on that page and became eligible for work. Joe | |
| ID: 939971 · | |
|
I'm mystified why people are making this so complicated. My understanding is that volunteers want the choice of whether to run Enhanced only, Astropulse only, or both, because Astropulse take a lot longer to run than Enhanced. | |
| ID: 940090 · | |
I'm mystified why people are making this so complicated. My understanding is that volunteers want the choice of whether to run Enhanced only, Astropulse only, or both, because Astropulse take a lot longer to run than Enhanced. You are right, that it should be like you discribe it, but reality turned out to be different. If I get this right, Astropulse v5 and V505 are not different versions of Astropulse, but separate aplications, and therefor they need their own separate switch? If it is possible I'd go with Norwich Gadfly and prefer a single AP masterswitch. But maybe the lack of a v505-switch was on purpose, so you have to set the "use other applications"-switch to get AP-units. Set up like this you'll get MB-units when AP runs out of work, wether you have set the MB-switch or not. This would hinder people from building up a 10 day cache full of AP's and not process any MB's at all, and by that futher increase the AP-demand and slow down the MB-processing. (yes, this sounds a little like a cospiracy theory :-)) ____________ | |
| ID: 940134 · | |
I'm mystified why people are making this so complicated. My understanding is that volunteers want the choice of whether to run Enhanced only, Astropulse only, or both, because Astropulse take a lot longer to run than Enhanced. Andreas has it right. The problem is that the work unit formats are different. BOINC needs to be able to tell the difference between 5.0 and 5.05 to send the right work, and to validate it properly. This was also quite important when Astropulse started because credits weren't all that constant, and some didn't like the crunch time. I'm not sure it is as important. ... and I think it's something that someone will do, but things like radar blanking (so we don't run completely out of work) are more important right now. ____________ | |
| ID: 940236 · | |
How will the user setting options to run 5.0 only, 5.05 only, or both, help BOINC decide which to run for a specific work unit ? Surely there must be an indicator in the work unit to specify which it is ? ____________ | |
| ID: 940403 · | |
The indication of which application to use is in the sched_reply_setiathome.berkeley.edu.xml file, there's no difference in the WUs for AP v5 and AP v505 but there are multiple differences in what the two different applications do. What was reported as positive DM changed to negative and vice versa, the builtin noise removal code was revised so it causes fewer false positives, the peak bin for repetitive pulses is reported differently, and some critical parameters were changed from single floats to double floats. If a task was originally sent to hosts running Astropulse v5 then any reissue must also be processed by Astropulse v5 to allow validation. All newly split work is assigned to be done by Astropulse v505. The preference settings allow users to choose the kind of work they get. For the transition from Astropulse to Astropulse v5 some users chose to concentrate on helping clean up the old work, some chose the newer, most chose to take whatever came. An edit of the boinc\html\user\project_specific_prefs.inc file is needed to provide similar choice for the current situation. Joe | |
| ID: 940959 · | |
My preferences are SETI@home Enhanced: yes Astropulse: yes Astropulse v5: yes So, what software version will process Astropulse units I receive ? ____________ | |
| ID: 941027 · | |
My preferences are If you have set "If no work for selected applications is available, accept work from other applications?" to "no," you won't get any. Gruß, Gundolf [edit]If you wanna know why, ask in the "Questions and answers" forum or in "Number crunching."[/edit] | |
| ID: 941042 · | |
My preferences are My apologies for not posting all the options. They are Run only the selected applications SETI@home Enhanced: yes Astropulse: yes Astropulse v5: yes If no work for selected applications is available, accept work from other applications? yes ____________ | |
| ID: 941048 · | |
Message boards : Technical News : Blip (Sep 16 2009)
| Copyright © 2013 University of California |