Message boards :
Number crunching :
RED ALERT !!!! SHIELDS UP
Message board moderation
Author | Message |
---|---|
FIEDLER Send message Joined: 27 Aug 99 Posts: 3 Credit: 2,367,185 RAC: 0 |
A LOT OF SERVERS NOT FUNCTIONING WITH NO NEW TECH NEWS WERE GOING DOWN CAPTAIN feeder.i686 bruno Not Running feeder.i686 ptolemy Not Running file_deleter1 bruno Not Running file_deleter2 bruno Not Running file_deleter3 bruno Not Running file_deleter4 bruno Not Running file_deleter5 bruno Not Running file_deleter6 bruno Not Running db_purge.x86_64 thumper Not Running transitioner1 vader Not Running transitioner2 vader Not Running transitioner3 bruno Not Running transitioner4 bruno Not Running vote_monitor bruno Not Running transitioner5 vader Not Running transitioner6 vader Not Running sah_validate1 ptolemy Not Running sah_validate2 ptolemy Not Running sah_validate3 ptolemy Not Running sah_validate4 ptolemy Not Running sah_validate5 ptolemy Not Running sah_validate6 ptolemy Not Running fix_missing_results ptolemy Disabled sah_assimilator1 ptolemy Not Running sah_assimilator2 ptolemy Not Running sah_assimilator3 ptolemy Not Running sah_assimilator4 ptolemy Not Running mb_splitter7 lando Disabled mb_splitter8 lando Disabled mb_splitter9 bambi Disabled mb_splitter10 bambi Disabled mb_splitter11 bambi Disabled mb_splitter12 bambi Disabled mb_splitter13 bambi Disabled mb_splitter14 bambi Disabled HEHEHEHHE |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
There is a note (posted on Sunday) by Matt -- seems they are doing 'extreme debugging'...: 23 Dec 2007 19:05:25 UTC Quick note: We never really did recover from the science database issues from a couple days ago due to DOS'ing ourselves with fast workunits. Whatever. We chose to let things naturally pass through the system. Kinda like kidney stones. Meanwhile, one of the assimilators is failing with a brand new error. If any of us have time we'll try to check into that over the coming days, but we may be out of luck until we're all in the lab doing "extreme debugging" together on Wednesday. Hang in there! - Matt |
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Everything has now gone green except the splitters - let's hope they're "extreme debugging" the code that generated all the AR 2.47 shorties. |
Blurf Send message Joined: 2 Sep 06 Posts: 8962 Credit: 12,678,685 RAC: 0 |
|
Richard Haselgrove Send message Joined: 4 Jul 99 Posts: 14650 Credit: 200,643,578 RAC: 874 |
Still no mention of the source / cause of all the shorty work - just its effects. |
Jim-R. Send message Joined: 7 Feb 06 Posts: 1494 Credit: 194,148 RAC: 0 |
Still no mention of the source / cause of all the shorty work - just its effects. The "source" or "cause" is simply the high angle range of the signals we were/are splitting. All of the work units are the same length, and contain the same amount of data, but when the data is recorded while the telescope is scanning the sky quickly (high angle range) or tracking on a certain star/point in the sky, (very low angle range) there are not as many calculations done on the data so the work units finish quickly. So "short" wu's is a misnomer in that they are the same length as any other wu, but they just require fewer calculations and finish quickly. Jim Some people plan their life out and look back at the wealth they've had. Others live life day by day and look back at the wealth of experiences and enjoyment they've had. |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
|
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Scotty!!! We need more power!! Shields are down, the crunchers are dropping like flies... Oh well....... All humour aside, could Matt or Eric respond to whether this is just an anomoly of the recent datasets, or if there is some other evil force at work?? "Freedom is just Chaos, with better lighting." Alan Dean Foster |
BarryAZ Send message Joined: 1 Apr 01 Posts: 2580 Credit: 16,982,517 RAC: 0 |
So if we all them short processor time/credit work units, you'll feel better, right? (smile). The thing is, they are inefficient for distributed processing handling big time. That is true for both the client (as the download/upload time is not different from 'regular' work units, along with the start/complete a new work unit overhead), and the server (as the download/upload load is not different and the process the completed work cycle is about the same as well).
|
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Inefficient??? How the heck much work being transmitted back and forth does the current Cricket Graph represent???? I seem to recall a short while ago that the servers could not even sustain this sort of bandwidth for more than an hour or so....... Anybody know of any other project that tries to handle this sort of bandwidth in and out of the servers? If you are sending out WUs that process in 20 minutes instead of 60 or more, of course your infrastructue is gonna be taxed. Don't give me any grief about 'inefficient for distributed processing'. Puleeez...... "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Dissident Send message Joined: 20 May 99 Posts: 132 Credit: 70,320 RAC: 0 |
*Ahem* I know I'm a noob here but perhaps what needs to be config'd on the servers is a graduated scale of workload i.e. a rig like Mark's would get a '16 hour' wu while a rig that turns 1 every 12 days would get a shortie and so on and so forth... Am I out of line/ way off base here? Someone (please?) tell me I'm not insane wih this idea. |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
*Ahem* Not insano at all....that's my department... This idea has been discussed in this forum before.. Allocating the WU to fit the host, and pairing like hosts together as wingmen rather that a fast cruncher with a 233mmx or such. The merits are good, the implementation a lot of work, the benefit to the project itself questionable. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
Dissident Send message Joined: 20 May 99 Posts: 132 Credit: 70,320 RAC: 0 |
Whew, good to hear! Glad you're in charge of the Ministry of the Mentally Wobbly... :p I've merely noticed there is a lot of frustration on the board and feel there has to be a better way through this mire. Not a critique, just an observation. Cheers! |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
Whew, good to hear! Glad you're in charge of the Ministry of the Mentally Wobbly... :p No problemo.... You have just noticed what others have as well...... 'Mentally Wobbly'...is that a politically correct term for 'unstable'? Ohh.....I resemble that remark..LOL. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
vordhosbn Send message Joined: 22 Mar 07 Posts: 6 Credit: 80,527 RAC: 0 |
About this optimization... Why "a lot of work"? Pairing computers with similar Recent Average Credit for example seems simple enough. There is no spoon. |
Dissident Send message Joined: 20 May 99 Posts: 132 Credit: 70,320 RAC: 0 |
'Mentally Wobbly'...is that a politically correct term for 'unstable'? Been accused of that myself. Heh! ;) |
W-K 666 Send message Joined: 18 May 99 Posts: 19062 Credit: 40,757,560 RAC: 67 |
About this optimization... If computers were to be paired it would have to be on "Average turnaround time" |
kittyman Send message Joined: 9 Jul 00 Posts: 51468 Credit: 1,018,363,574 RAC: 1,004 |
About this optimization... Or RAC might be a more accurate indicator...I think ATT might fluctuate more.. "Freedom is just Chaos, with better lighting." Alan Dean Foster |
zoom3+1=4 Send message Joined: 30 Nov 03 Posts: 65746 Credit: 55,293,173 RAC: 49 |
Scotty!!! We need more power!! Shields are down, the crunchers are dropping like flies... Scotty beam the tribbles back to that Klingon ship, So there won't be any tribbles with the engines. ;) Yep lots of short work to do, Shoveling those shorties. :D And here's something for Scotties engines(bairns?): The T1 Trust, PRR T1 Class 4-4-4-4 #5550, 1 of America's First HST's |
PhonAcq Send message Joined: 14 Apr 01 Posts: 1656 Credit: 30,658,217 RAC: 1 |
Are those round things in the middle of your dinner plate truffles, tribbles, or something worse? |
©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.