Average Credit Decreasing?

Message boards : Number crunching : Average Credit Decreasing?
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 32 · Next

AuthorMessage
Profile Zalster Special Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 27 May 99
Posts: 5517
Credit: 528,817,460
RAC: 242
United States
Message 1782296 - Posted: 25 Apr 2016, 6:52:44 UTC - in response to Message 1782277.  

On the upside, I feel like I have 980 Titans now with how many files I am going though :)


Validation pending is approaching 2000 due to these things, lol
ID: 1782296 · Report as offensive
Profile Mr. Kevvy Crowdfunding Project Donor*Special Project $250 donor
Volunteer moderator
Volunteer tester
Avatar

Send message
Joined: 15 May 99
Posts: 3776
Credit: 1,114,826,392
RAC: 3,319
Canada
Message 1782337 - Posted: 25 Apr 2016, 11:40:05 UTC - in response to Message 1782296.  

Validation pending is approaching 2000 due to these things, lol


7942 currently. :^)
ID: 1782337 · Report as offensive
Profile Brent Norman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester

Send message
Joined: 1 Dec 99
Posts: 2786
Credit: 685,657,289
RAC: 835
Canada
Message 1782341 - Posted: 25 Apr 2016, 11:52:20 UTC

I found a new game

Disable all tasks, then enable only MESSIER then see how fast they disappear :)
ID: 1782341 · Report as offensive
Profile John Neale
Volunteer tester
Avatar

Send message
Joined: 16 Mar 00
Posts: 634
Credit: 7,246,513
RAC: 9
South Africa
Message 1782393 - Posted: 25 Apr 2016, 15:00:41 UTC

Is it fair to assume that these MESSIER031 tasks were created when the GBT was pointed at the Andromeda Galaxy?
ID: 1782393 · Report as offensive
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1513
Credit: 370,893,186
RAC: 340
United States
Message 1782415 - Posted: 25 Apr 2016, 17:33:45 UTC

Seems like over the last couple of days, my RAC has stabilized (on one machine) and even increased for two days (on the other), after falling for several days on both.

Is this just a fluke, or is the worst over?
ID: 1782415 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51468
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1782418 - Posted: 25 Apr 2016, 17:37:27 UTC - in response to Message 1782415.  
Last modified: 25 Apr 2016, 17:51:30 UTC

Seems like over the last couple of days, my RAC has stabilized (on one machine) and even increased for two days (on the other), after falling for several days on both.

Is this just a fluke, or is the worst over?

Mine seems to be recovering a bit as well.
However, that's gonna go out the window if the servers can't continue to feed the GPUs enough kibbles.
"Freedom is just Chaos, with better lighting." Alan Dean Foster

ID: 1782418 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1782424 - Posted: 25 Apr 2016, 18:15:35 UTC - in response to Message 1782415.  
Last modified: 25 Apr 2016, 18:16:14 UTC

Seems like over the last couple of days, my RAC has stabilized (on one machine) and even increased for two days (on the other), after falling for several days on both.

Is this just a fluke, or is the worst over?


I predict oscillation with work mix, around +/- 37% of some approximate midpoint. As oscillation instabilities work, some hosts will stay on the low side, and others stay on the high side of that, 5% in each end case never crossing the middle over extended periods, and the other 90% of us flopping around.
"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.
ID: 1782424 · Report as offensive
Profile HAL9000
Volunteer tester
Avatar

Send message
Joined: 11 Sep 99
Posts: 6534
Credit: 196,805,888
RAC: 57
United States
Message 1782427 - Posted: 25 Apr 2016, 18:29:56 UTC - in response to Message 1782424.  

Seems like over the last couple of days, my RAC has stabilized (on one machine) and even increased for two days (on the other), after falling for several days on both.

Is this just a fluke, or is the worst over?


I predict oscillation with work mix, around +/- 37% of some approximate midpoint. As oscillation instabilities work, some hosts will stay on the low side, and others stay on the high side of that, 5% in each end case never crossing the middle over extended periods, and the other 90% of us flopping around.

Prior to v8 & GBT work I a RAC oscillation of +/- 25-30% was not uncommon on some of my machines. The CPU only machines had a tendency to wiggle the least.
SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[
ID: 1782427 · Report as offensive
Profile ReiAyanami
Avatar

Send message
Joined: 6 Dec 05
Posts: 116
Credit: 222,900,202
RAC: 174
Japan
Message 1782442 - Posted: 25 Apr 2016, 19:30:30 UTC
Last modified: 25 Apr 2016, 19:55:12 UTC

All mine have been going down for over 2 weeks (between 9 to 12%) and still going.
It seems that effects are at least similar to many machines.
My fastest machine's been staying in the middle of the same page of the Top hosts for at least 6 months and I don't see too many position changes in my neighborhood.
Since I don't do anything other than SETI, this is not too bad.
Decreasing RAC certainly doesn't motivate me, though.
I prefer numbers reflect calculation power and time of my machines and something more stable....
ID: 1782442 · Report as offensive
Profile AllenIN
Volunteer tester
Avatar

Send message
Joined: 5 Dec 00
Posts: 292
Credit: 58,297,005
RAC: 311
United States
Message 1783046 - Posted: 28 Apr 2016, 1:55:18 UTC
Last modified: 28 Apr 2016, 1:55:36 UTC

From April 3rd (22,000 RAC) I was on a steady uphill to 24,200 RAC on April 10th, but on April 14th I was on a fast downhill to 22,100 RAC by April 24th.

Now it seems to be steady at this point not moving up or down more than about 70 RAC. This is for 5 machines, 4 with GPU's.

I was thinking of suspending all of the guppis and see if the RAC starts to climb again.
ID: 1783046 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11358
Credit: 29,581,041
RAC: 66
United States
Message 1783048 - Posted: 28 Apr 2016, 2:07:36 UTC - in response to Message 1783046.  

guppis need too be pocessed
ID: 1783048 · Report as offensive
Lionel

Send message
Joined: 25 Mar 00
Posts: 680
Credit: 563,640,304
RAC: 597
Australia
Message 1783049 - Posted: 28 Apr 2016, 2:11:51 UTC - in response to Message 1782442.  

I went back and had a look at some old data.

Prior to v7, I had a daily credit run rate of 220+k/day. Then came v7. Now GBT data. My daily run rate is now bouncing around between 130k and 170k per day. On the surface some would same not to bad given passed history.

The twist in here is that I used to run with 2 x GTX580s in two boxes (4 cards), 2 x GTX470s in the third box and 2 x GTX295s in the fourth box. Some time after v7 I began changing things. I replaced the GTX580s with GTX770 SCs and then replaced these with GTX780 TI SCs. The GTX470s were replaced with GTX680 Classifieds and the GTX295s I replaced with 2 of the old GTX580s. All the machines now process more WUs than before, upwards of 50% in 2 cases.

So I look at what recognition I receive today vs 3 years ago and considering that the machines are doing considerably more than they used to, I wonder how on earth anyone at Berkerley could say that there isn't an issue with the credit system and not want to understand why it is behaving in the manner that it is.
ID: 1783049 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11358
Credit: 29,581,041
RAC: 66
United States
Message 1783051 - Posted: 28 Apr 2016, 2:22:24 UTC - in response to Message 1783049.  

I wonder how on earth anyone at Berkerley could say that there isn't an issue with the credit system and not want to understand why it is behaving in the manner that it is.

It is obvious that creditnew is not reflecting thru put so the real question I have is what is it supposed to show?
ID: 1783051 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1783198 - Posted: 28 Apr 2016, 13:57:31 UTC - in response to Message 1783051.  

I wonder how on earth anyone at Berkerley could say that there isn't an issue with the credit system and not want to understand why it is behaving in the manner that it is.

It is obvious that creditnew is not reflecting thru put so the real question I have is what is it supposed to show?


It's supposed to first dial in (adaptively converge) elapsed time estimates for sending the right amount of work to hosts, and then on client side hooks into the scheduling of tasks and projects.

Once validations occur, it then awards by the cobblestone scale for tasks, in theory RAC then directly indicative of successful throughput, yes (just a different scale). The elapsed times then go back into the first step to dial in estimates.

As an adaptive control system, it has some notable & classic engineering instabilities, that make each stage pretty fragile under anything but rare idealised (non existent) conditions. The rub is that it 'sortof works' as described, but not quite meeting the expectations of users. In that respect it's a proof-of-concept mechanism, rather than a complete system that should be in deployment. The reason for that is there are no health/quality-control indicators built in (other than user complaints) that would make refinement easier. Also semantically and structurally the code uses its own 'standards' that in some areas obscure what the elements actually are, making replacement of the pieces with 'better/more-standard' implementations harder than it should be.
"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.
ID: 1783198 · Report as offensive
Profile William
Volunteer tester
Avatar

Send message
Joined: 14 Feb 13
Posts: 2037
Credit: 17,689,662
RAC: 0
Message 1783206 - Posted: 28 Apr 2016, 14:22:42 UTC

In easier words, it is supposed to take the time you took for a task, award you credit according to how efficient you were (compared to all tasks and tasks in your group of tasks) in crunching the task, [I think in theory, you are supposed to get a bonus if you are faster than the average] and then use the time (in relation to how 'big' the task was) to estimate your speed for future tasks (APR).

I'm not sure that was an easier explanation of what it's supposed to do...

In praxi, we have noticed that it falls sadly short of both making a good estimate of your speed (it lacks adaptability) and of awarding credit relative to your efficiency.

In short, the way in which it was implemented (coded) cannot work. It leads to chaotic behaviour.
A person who won't read has no advantage over one who can't read. (Mark Twain)
ID: 1783206 · Report as offensive
Ulrich Metzner
Volunteer tester
Avatar

Send message
Joined: 3 Jul 02
Posts: 1256
Credit: 13,565,513
RAC: 13
Germany
Message 1783212 - Posted: 28 Apr 2016, 15:03:20 UTC

Interestingly the slower you crunch, the higher the 'award' with this strange 'system'. I can easily recognize this, cause i have two different GPUs, the GT 640 being nearly three times faster than the GT 430. On similar WUs the GT 430 always get more credit, only topped by the CPU (Core2Quad), which again earns more credit for a similar WU. I think "borked" is way too friendly for this behavior...
Aloha, Uli

ID: 1783212 · Report as offensive
Profile jason_gee
Volunteer developer
Volunteer tester
Avatar

Send message
Joined: 24 Nov 06
Posts: 7489
Credit: 91,093,184
RAC: 0
Australia
Message 1783217 - Posted: 28 Apr 2016, 15:21:47 UTC - in response to Message 1783212.  
Last modified: 28 Apr 2016, 15:29:07 UTC

Interestingly the slower you crunch, the higher the 'award' with this strange 'system'. I can easily recognize this, cause i have two different GPUs, the GT 640 being nearly three times faster than the GT 430. On similar WUs the GT 430 always get more credit, only topped by the CPU (Core2Quad), which again earns more credit for a similar WU. I think "borked" is way too friendly for this behavior...


Yep, that's right. This was discovered during controlled tests on Albert, which had tasks that process identically at the time. If you crunch less efficiently (i.e. take longer and generate more heat) you end up on the high side of the claim. [i.e. process much slower than the expected ~5%+/- of peak_flops (GPU)]

Higher credit will happen when you crunch as slowly as possible (claiming lots of operations), and are paired with a blazingly efficient wingman [crunches at higher % of his peak_flops].

Importantly the average claim on the quorum is scaled back to a reference application anyway, in our project case making it a gross underclaim because of no compensation for that being SIMD enabled.

Crunching slower at some point will lead to overall lower credit, but the loophole is you have enough memory these days to pile on tasks [to run slowly in parallel].

It's just a matter of time until we start adding scaling/throttle controls into the various apps for other reasons (mainly safety and usability control), but in the meantime you can pile on the instances and downclock the GPU after Boinc starts, then you'll consistently claim high. In an extreme case you would use as many instances of the lowest performing application as will fit. [Will always be working toward reduced memory, Ram and Vram, modes to support more devices]

A problem can come with that if you are paired with someone doing the same (intentionally or unintentionally), but I think even if everyone that reads this did it, still enough host-owners never visit the forums that this low effective claim pairing would remain rare.
"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.
ID: 1783217 · Report as offensive
Cruncher-American Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor

Send message
Joined: 25 Mar 02
Posts: 1513
Credit: 370,893,186
RAC: 340
United States
Message 1783226 - Posted: 28 Apr 2016, 15:53:25 UTC
Last modified: 28 Apr 2016, 16:00:43 UTC

All this kvetching is so unseemly. You are all AGCNDs (that's Anthropogenic Global Credit New Deniers). Shame on you!

You should just grin and bear it, like good little Computation Drones; your Betters have decided what is good for you.

----------------------------------------------------------------------

Oh, and by the way, what use are the time estimates for WUs, anyway? We all just get our 100 for CPU and 100/GPU anyway, so that is pointless at this time, is it not?
ID: 1783226 · Report as offensive
Al Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Avatar

Send message
Joined: 3 Apr 99
Posts: 1682
Credit: 477,343,364
RAC: 482
United States
Message 1783251 - Posted: 28 Apr 2016, 16:04:28 UTC - in response to Message 1783212.  

Interestingly the slower you crunch, the higher the 'award' with this strange 'system'. I can easily recognize this, cause i have two different GPUs, the GT 640 being nearly three times faster than the GT 430. On similar WUs the GT 430 always get more credit, only topped by the CPU (Core2Quad), which again earns more credit for a similar WU. I think "borked" is way too friendly for this behavior...

Absolutely true it appears. In this example, I have 2 machines, a Genuine Intel(R) Atom(TM) CPU 330 @ 1.60GHz [Family 6 Model 28 Stepping 2] (4 processors) and a Genuine Intel(R) Pentium(R) 4 CPU 3.80GHz [Family 15 Model 4 Stepping 10] (2 processors) and have recently been keeping an eye on them, and wondering what might be going on. They are both configured to only run GPU tasks, as the Atom would pretty much choke on anything CPU, it really wouldn't be worth it.

As the P4 3.8 proc is unfortunately fairly pitiful as well, I decided to set that up as a GPU only cruncher too. Now to be honest, the Atom just sits upstairs happily crunching away with no one bothering it at all, and the P4 is used by my daughter for web based games and such, maybe 1 hour each day of the week and on one of those days for a total of probably 2-3 hours, where crunching is suspended because it really effects the usability of the system if it is enabled.

So because of that, it's not an _exact_ apples to apples comparison, but pretty dang close I'd say, and the difference in RAC (or should I say lack of significant difference) is pretty surprising. And actually, the Atom is (and has been for a while) running higher (6,304 ) than the P4 (5,765), and it only has a GTX260 in it, whereas the P4 has a GTX950, which is 5 generations newer then the 260, and both the 260 and the 950 are pretty much the lower end of their respective spectrums when they were new.

Just thought I'd toss that out there as an interesting comparison.

ID: 1783251 · Report as offensive
Profile betreger Project Donor
Avatar

Send message
Joined: 29 Jun 99
Posts: 11358
Credit: 29,581,041
RAC: 66
United States
Message 1783263 - Posted: 28 Apr 2016, 16:46:07 UTC - in response to Message 1783258.  

MY stats have that same flavor, the question I have is when will it flatten out?
ID: 1783263 · Report as offensive
Previous · 1 . . . 6 · 7 · 8 · 9 · 10 · 11 · 12 . . . 32 · Next

Message boards : Number crunching : Average Credit Decreasing?


 
©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.