Engadget article on SETI

Message boards : News : Engadget article on SETI
Message board moderation

To post messages, you must log in.

AuthorMessage
Eric KorpelaProject Donor
Volunteer moderator
Project administrator
Project developer
Project scientist
Avatar

Send message
Joined: 3 Apr 99
Posts: 1267
Credit: 20,763,130
RAC: 10,695
United States
Message 1869987 - Posted: 28 May 2017, 18:10:59 UTC

Engadget has published an article on SETI which includes discussion of Breakthrough Listen and SETI@home.
@SETIEric

ID: 1869987 · Report as offensive
Filipe

Send message
Joined: 12 Aug 00
Posts: 181
Credit: 8,461,655
RAC: 277
Portugal
Message 1870003 - Posted: 28 May 2017, 21:12:29 UTC

However, even with the million-odd CPU cores at SETI@Home's disposal, analyzing all that data is still slow going. Its volunteers only account for around 2 percent of the Breakthrough Listen Initiative's analytical power.



Seti@home represents only 2% of Breakthrough Listen Initiative's compunting power?

Is it for real? Only 2%?
ID: 1870003 · Report as offensive
Richard HaselgroveProject Donor
Volunteer tester

Send message
Joined: 4 Jul 99
Posts: 11583
Credit: 107,657,473
RAC: 60,823
United Kingdom
Message 1870008 - Posted: 28 May 2017, 22:20:22 UTC - in response to Message 1870003.  

However, even with the million-odd CPU cores at SETI@Home's disposal, analyzing all that data is still slow going. Its volunteers only account for around 2 percent of the Breakthrough Listen Initiative's analytical power.
Seti@home represents only 2% of Breakthrough Listen Initiative's computing power?

Is it for real? Only 2%?
2% of activity, or 2% of need?

When I first read the Engadget article, I assumed only 2% of the data was being processed. But reading it again, I'm inclined to agree with you - somewhere else is supplying ~50 times the power of SETI@Home. It would be interesting to know where that is, and whether their results will feed into something like the Nebula post-processing stream.

I'm disappointed that Breakthrough Listen itself still seems to have stated nothing more explicit than http://breakthroughinitiatives.org/OpenData/1:

OPEN DATA

Fundamental to Breakthrough Listen is transparency of data for everyone.

Each telescope will generate vast amounts of data. All of it will be open to the public. This is likely to produce the largest amount of scientific data ever made available to the public.

Breakthrough Initiatives are also developing powerful software for analyzing this data. All this software will be open source. The hardware and software used in the LISTEN project will be compatible with other telescopes around the world, enabling them to join the search for intelligent life. And scientists and members of the public can develop their own apps to sift the data for phenomena of interest.

We will also be joining forces University of California, Berkeley and their famous SETI@home program to bring the power of distributed computing to bear on the data. SETI@home is a platform of 9 million computers around the globe – collectively forming one of the world’s biggest supercomputers.
Both statements are vague, and still written in the future tense - although SETI@Home has been working on the data for over a year now.
ID: 1870008 · Report as offensive
Profile Chris SCrowdfunding Project Donor*
Volunteer tester
Avatar

Send message
Joined: 19 Nov 00
Posts: 40289
Credit: 36,143,532
RAC: 49,766
United Kingdom
Message 1870073 - Posted: 29 May 2017, 9:30:50 UTC
Last modified: 29 May 2017, 10:09:52 UTC

Thanks for posting this Eric.

Three points stood out to me.

The program currently only has around 150,000 volunteers (down from a peak of 1.5 million users) and "we're getting back into our problem again in that the telescope can generate far more data than we can analyze with the best sensitivity," said Dr. Eric Korpela, head of the SETI@Home project.
People walked because we sat on 17 years of data and couldn't practically do anything with it. It took Nebula to hopefully sort that out.

And the fact that many people have ditched their desktops for mobile devices is not helping either. "It is an issue that we worry about," Korpela admitted. "We do have an app for Android. The processors that are in a typical phone right now are not comparable with what are in most desktops, but they're certainly better than a processor from 1999." The app is currently running on 22,000 volunteer mobile devices, or around 15 percent of the total base. However, these devices are only contributing 2.3 TFLOP/s of processing, 0.5 percent of the program's total computational power.
That could explain why they changed the colour scheme to favour mobile devices at the expense of Desktop users.

But given that there are a couple billion Android devices out there," Korpela mused, "there are another 200 petaFLOP/s out there that we haven't tapped yet."
The Demographics of mobile device users suggest to me that the groups concerned are maybe unlikely to be interested in ET and SETI. Imagine the ultimate of 2 billion users trying to upload and download data, we cant cope with 150,000 without shutting down one day a week.

Lastly, no mention of the sheer computing power of medium and top end GPU devices., which get updated every few months these days with even more power. We know the world is turning to portable devices because of 21C lifestyles, but trying to get 2 billion android owners to crunch Seti, will maybe take more effort, than to get better GPU apps like Petri's, and Lunatics, and increase our measly 150,000 user base to be larger and more productive.

Postscript - "SETI@home is a platform of 9 million computers around the globe"

Really? OK, that is an average of 60 each then. When will I get sent my other 55?
ID: 1870073 · Report as offensive
ryckeley

Send message
Joined: 25 Jul 15
Posts: 1
Credit: 7,627,728
RAC: 4,088
Message 1871670 - Posted: 8 Jun 2017, 2:04:38 UTC

We would do well to write code which can be executed on bitcoin mining equipment, ASICs in particular. There are millions of potential computers awaiting new purpose. I would donate some funds for that :)
ID: 1871670 · Report as offensive
Profile Mr. KevvyCrowdfunding Project Donor*
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 1771
Credit: 384,698,595
RAC: 422,344
Canada
Message 1871674 - Posted: 8 Jun 2017, 2:31:51 UTC - in response to Message 1871670.  
Last modified: 8 Jun 2017, 2:33:52 UTC

We would do well to write code which can be executed on bitcoin mining equipment, ASICs in particular. There are millions of potential computers awaiting new purpose. I would donate some funds for that :)


Being ASIC, the AS being "Application Specific", they are hardcoded for mining and missing features and functions that SETI@Home would require. I don't know the specifics of what, but hashing doesn't require FP math as far as I know, so these ASIC chips may be only able to do integer math (at least well), for example.
“Never doubt that a small group of thoughtful, committed citizens can change the world; indeed, it's the only thing that ever has.”
--- Margaret Mead

ID: 1871674 · Report as offensive
Profile Gordon Lowe
Volunteer moderator
Avatar

Send message
Joined: 5 Nov 00
Posts: 9894
Credit: 5,221,571
RAC: 2,417
United States
Message 1871835 - Posted: 8 Jun 2017, 23:50:37 UTC

Essentially, the Institute can point the ATA at a known exoplanet (or system like TRAPPIST-1), measure the star system's radio signature, then wait for the planet to transit behind the sun and measure the system's radio signature again. If there's a significant difference in the readings, that could mean there's something afoot on that exoplanet. The Institute wouldn't necessarily be able to understand what's encoded in the transmission but it would provide strong impetus for other astronomical research groups to follow up with their own observations. This is actually very close to what the Breakthrough Listen Initiative team is doing, albeit with a less sensitive piece of equipment.


This strategy actually makes sense to me.

I hope the ATA gets the scope of it's antenna grid expanded, and other telescopes come online soon, too.
The mind is a weird and mysterious place
ID: 1871835 · Report as offensive

Message boards : News : Engadget article on SETI


 
©2017 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.