Message boards :
Number crunching :
What does SETI need? BOINC?
Message board moderation
Author | Message |
---|---|
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
As I had mentioned in another thread I have going here, I have been away from actively participating in SETI for a few years now, and the last thing I remember being a big deal was the need for some more hardware (servers if I recall, going from a scrum of hand me downs to some actual decent hardware worthy of a server closet, via a donation in memory of someones parents. Which was quite cool, btw. I have been doing a little reading around on the threads, but was wondering if someone could take the time to highlight what the current status is of SETI, what it needs (hardware, funding, programmers, more data to be processed?). Also, the status of BOINC, as I read a thread about funding being cut off, and possibly being restored sometime around now if I remember correctly. And how that effects SETI. It appears that things seem to be plugging along fairly ok right now best as I can gather scanning thru some threads, not going perfectly of course, but not collapsing either. If someone could fill in the blanks of recent history, as well as the current and possibly expected in the near future situation, I'd greatly appreciate it. Doesn't have to be a book, but a clear, concise and somewhat detailed would be great. |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30702 Credit: 53,134,872 RAC: 32 |
As I had mentioned in another thread I have going here, I have been away from actively participating in SETI for a few years now, and the last thing I remember being a big deal was the need for some more hardware (servers if I recall, going from a scrum of hand me downs to some actual decent hardware worthy of a server closet, via a donation in memory of someones parents. Which was quite cool, btw. BOINC funding is gone and it doesn't look like it is coming back. So updates to it are strictly volunteer. Fortunately it doesn't need much updates in the short term. Long term as O/S updates happen and new hardware is released, it will need programming. SETI is always short of cash. It appears as if it has a new data source. It looks like the software to crunch that new source it written and tested. There is a need for programming, hardware, and the like to get the near time persistency checker operating 24/7. Obviously some budget to operate the existing data servers and man power for that. On the data source front, the Arecibo radio telescope is in need of funding. Government funding. Write your Congress people and tell them to fund it. Also on the data source front Berkeley Seti might be able to use cash to build more receivers to site at radio astronomy sites worldwide as the sites allow. Some political support in those home countries for Seti might also be of assistance. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
Einstein@home is getting data from both Arecibo and Parkes radiotelescopes. I don't know on which terms. Parkes is in danger of closure due to the need for CSIRO to fund the Square Kilometer Array in Australia and South Africa. I just learned from a "Nature" article that the headquarters of the SKA are at Jodrell Bank. UK. I had read also in "Nature" that Padova was a possible site, in a palace near the University of Galileo. Then the UK protested and got the site. Tullio |
rob smith Send message Joined: 7 Mar 03 Posts: 22237 Credit: 416,307,556 RAC: 380 |
One forgets that SETI is about to get data from the GBT, which , unlike Arecibio is a fully steerable dish, so the sky coverage will be greater. On Beta there have been a few work units sent out as a test of the splitters to split and v8 to process then data. It will be interesting to see if there are any plans to get data from either the SKA or Parkes, as either would dramatically increase the sky coverage below the Equator. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
So, as always, money is the main need, both directly and indirectly. Are there any pressing needs like the server crisis I remembered from back then? You mentioned some hardware, is there any specifics of what is needed currently to make the project operate more smoothly? What exactly is the near time persistency checker, and why isn't it operating 24/7? |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
So, as always, money is the main need, both directly and indirectly. Are there any pressing needs like the server crisis I remembered from back then? You mentioned some hardware, is there any specifics of what is needed currently to make the project operate more smoothly? What exactly is the near time persistency checker, and why isn't it operating 24/7? The server infrastructure can't handle running ntpckr & the project at the same time. It's the 10lbs of stuff in a 5lb box problem. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
rob smith Send message Joined: 7 Mar 03 Posts: 22237 Credit: 416,307,556 RAC: 380 |
What exactly is the near time persistency checker Its an application that is meant to search through the results database looking for matches, and do this in "near real time". Due to the size of the database this search is not possible in real time on the current hardware. Bob Smith Member of Seti PIPPS (Pluto is a Planet Protest Society) Somewhere in the (un)known Universe? |
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
Ok, thanks. What type of hardware would be needed to be able to run either/both of these processes in real time? |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14654 Credit: 200,643,578 RAC: 874 |
Ok, thanks. What type of hardware would be needed to be able to run either/both of these processes in real time? You might find it helpful to read through some of the history of the work that's been put into, in particular, the Near Time Persistency Checker. The quickest way is to open http://setiathome.berkeley.edu/tech_news.php and search for the abbreviation 'ntpckr'. I particularly liked this paragraph from 2011: This clearly points out oscar's inability to handle the crazy random i/o's we desire, though to be fair oscar is indeed operating better in its current state than the old science database. There's still MANY knobs to turn in informix-land before we need to add more disk spindles. For example, we still haven't given all the memory available in the system over to informix. The tweak we made today added an additional 20GB to the buffers. Note that it takes about a week to fill these buffers, so we won't notice any improvement, if any, until then. Oscar had been purchased in Fall 2010, and the specifications are here. Keynotes: RAM: 96GB (12 x 8GB) Operating at 1066MHz Max (DDR3-1333 ECC Registered DIMMs) Drive Set: 12 x 1TB Seagate Constellation ES (6Gb/s, 7.2K RPM, 16MB Cache) 3.5" SAS Since then, the science database has been switched over to PaddyM (with 132 GB RAM and 50% more CPU cores), but I hope that gives you some sort of a starting point. |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30702 Credit: 53,134,872 RAC: 32 |
Latest ntpckr news is here http://setiathome.berkeley.edu/forum_thread.php?id=78882 in the video. The data and I/O requirements for their first attempt became obviously too much for "a box" of real hardware to accomplish. They are now looking into a cloud solution. If it works then perhaps they might consider building their own cloud server farm, but that is a computer science project, not an astronomy science project. |
jason_gee Send message Joined: 24 Nov 06 Posts: 7489 Credit: 91,093,184 RAC: 0 |
Samsung's production of 15TB SaS SSDs could be a thing (just sayin'). Anyone have contacts there ? "Living by the wisdom of computer science doesn't sound so bad after all. And unlike most advice, it's backed up by proofs." -- Algorithms to live by: The computer science of human decisions. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
I've used two SSD on my Linux boxes, a 120 GB OCZ Vertex 4 and a 256 GB Samsung 840 and they both failed after a while. So I reverted to hard disks. Tullio |
HAL9000 Send message Joined: 11 Sep 99 Posts: 6534 Credit: 196,805,888 RAC: 57 |
I've used two SSD on my Linux boxes, a 120 GB OCZ Vertex 4 and a 256 GB Samsung 840 and they both failed after a while. So I reverted to hard disks. OCZ SSDs were as notorious for failing as the IBM Deskstar drives. I think that is one of the key things that lead to OCZ going bankrupt and being bought by Toshiba. I haven't seen a similar widespread failure rate in Samsung models. However I don't feel they are worth the extra cost vs some other drives. Enterprise level SSDs like SETI@home would need tend to be a bit more reliable than consumer desktop models. They have also surpassed HDDs in the reliability metric. SETI@home classic workunits: 93,865 CPU time: 863,447 hours Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[ |
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
Such Timing! Just released to the world! https://news.samsung.com/global/samsung-now-introducing-worlds-largest-capacity-15-36tb-ssd-for-enterprise-storage-systems I saw 3 articles, checked out 2 of them, both said that Pricing Hasn't Been Announced. Which to me says, if you have to ask, you Can't afford it.. hehe *edit* and just how many of those behemoths would SETI need? That is a huge amount of data per drive, at least to us normal humans... |
Jord Send message Joined: 9 Jun 99 Posts: 15184 Credit: 4,362,181 RAC: 3 |
..that Pricing Hasn't Been Announced. Just saw one that thought it would go for upwards of $5,000.- which is nothing. |
AMDave Send message Joined: 9 Mar 01 Posts: 234 Credit: 11,671,730 RAC: 0 |
..that Pricing Hasn't Been Announced. I'd say >$2,000 per Terabyte (using Fixstars 13TB SSD for comparison). WD has some high capacity HDDs (6TB, 8TB, & 10TB) for much less $. These are Helium filled units which according to "Western Digital’s own figures put the power consumption ... at 23% less than its own conventional air-filled drives." |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30702 Credit: 53,134,872 RAC: 32 |
*edit* and just how many of those behemoths would SETI need? That is a huge amount of data per drive, at least to us normal humans... I don't think the science DB is in the exabyte class yet. :) |
Al Send message Joined: 3 Apr 99 Posts: 1682 Credit: 477,343,364 RAC: 482 |
..that Pricing Hasn't Been Announced. Say What? ~$30K per Drive? I don't know, but to me, if that is true, that's insane... |
Filipe Send message Joined: 12 Aug 00 Posts: 218 Credit: 21,281,677 RAC: 20 |
WD has some high capacity HDDs (6TB, 8TB, & 10TB) for much less $. HDD can't cope with the needed I/O for the "near time persistency checker" |
AMDave Send message Joined: 9 Mar 01 Posts: 234 Credit: 11,671,730 RAC: 0 |
..that Pricing Hasn't Been Announced. I should have been clearer. The 13TB=$1,000/TB (in 'normal' mode), then further into the article: "it also features a proprietary “high durability†mode. With this mode enabled, the durability and longevity of the drive is promised to be tripled but in return, the capacity is cut in half. Oh dear, “only†6.5TB remain on your $13,000 dollar, 13TB SSD—for those of you that aren’t keeping count, that is $2,000 per Terabyte." |
©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.