Optimized apps: Reliable and accurate?

Questions and Answers : Getting started : Optimized apps: Reliable and accurate?

To post messages, you must log in.

AuthorMessage
Steve Bergman
Volunteer tester

Send message
Joined: 20 May 07
Posts: 48
Credit: 292,679
RAC: 0
United States
Message 1037649 - Posted: 1 Oct 2010, 2:09:32 UTC

I'm considering running optimized versions of the apps. Q6600, 64 bit, Linux. I don't care about credits, and competing, and teams, and all that sort of thing. But I do care very much about the quality of the science work. How much assurance do I have that down the road, I won't find out about some subtle SSE3 bug or some such that renders my results useless, or of reduced science value? Or worse yet, *not* find out about one that exists but nobody noticed. I'd hate to miss the alien "Wormhole Construction HowTo" due to a silly app bug.

-Steve
ID: 1037649 · Report as offensive
John McLeod VII
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 15 Jul 99
Posts: 24806
Credit: 754,585
RAC: 65
United States
Message 1037663 - Posted: 1 Oct 2010, 2:40:24 UTC - in response to Message 1037649.  

I'm considering running optimized versions of the apps. Q6600, 64 bit, Linux. I don't care about credits, and competing, and teams, and all that sort of thing. But I do care very much about the quality of the science work. How much assurance do I have that down the road, I won't find out about some subtle SSE3 bug or some such that renders my results useless, or of reduced science value? Or worse yet, *not* find out about one that exists but nobody noticed. I'd hate to miss the alien "Wormhole Construction HowTo" due to a silly app bug.

-Steve

SETI does not use homogeneous redundancy. That means that the tasks you do on your computer get compared against tasks done on completely different computers. If disparate computers match, neither of them has a subtle bug in the underlying hardware. If yours is done on SSE3 your wingmates is done on a GPU, and they both match, there is nothing in common in the hardware.


BOINC WIKI
ID: 1037663 · Report as offensive
Profile BilBg
Volunteer tester
Avatar

Send message
Joined: 27 May 07
Posts: 3654
Credit: 8,596,598
RAC: 1,132
Bulgaria
Message 1037664 - Posted: 1 Oct 2010, 2:40:26 UTC - in response to Message 1037649.  


No one can assure that any program is 100% free of bugs
so the corporations sell software which may cost 1000s of $
and at the same time they push you to [Agree] that you get the program "As Is, with all faults"!

So far the optimized apps (released to public - not beta)
proved to be accurate on results (they validate OK for years against "stock" results).





- ALF - "Find out what you don't do well ..... then don't do it!" :)
ID: 1037664 · Report as offensive
Steve Bergman
Volunteer tester

Send message
Joined: 20 May 07
Posts: 48
Credit: 292,679
RAC: 0
United States
Message 1037684 - Posted: 1 Oct 2010, 3:35:02 UTC - in response to Message 1037664.  

Thanks for the replies. That sounds like a reasonable assurance. I've installed the latest AK Linux X86_64 SSE3 apps, and multibeam looks to be running 2-1/2 to 3 times faster.

-Steve
ID: 1037684 · Report as offensive

Questions and Answers : Getting started : Optimized apps: Reliable and accurate?


 
©2016 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.