Observation of CreditNew Impact

Message boards : Number crunching : Observation of CreditNew Impact
Message board moderation

To post messages, you must log in.

Previous · 1 . . . 10 · 11 · 12 · 13 · 14 · 15 · Next

AuthorMessage
Profile cov_route
Avatar

Send message
Joined: 13 Sep 12
Posts: 342
Credit: 10,270,618
RAC: 0
Canada
Message 1383222 - Posted: 20 Jun 2013, 22:29:19 UTC
Last modified: 20 Jun 2013, 22:29:33 UTC

I wonder if running multi-instances on a GPU screws up CreditNew? From what I could gleen from the description (before my eyes glazed over and started spinning like a slot machine and then fell right out) the system is supposed to track the capabilities of devices. Well, if you run 2-up on a gpu when the baseline is 1-up, maybe that borks some timing numbers somewhere?

Like this right here:

For jobs done by anonymous platform apps, the server knows the devices involved and can estimate PFC. It maintains host_app_version records for anonymous platform, and it keeps track of PFC and elapsed time statistics there. There are separate records per resource type. The record's app_version_id encodes the app ID and the resource type (-2 for CPU, -3 for NVIDIA GPU, -4 for ATI).

To be honest I don't know what any of it means. It *looks* like english but...
ID: 1383222 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13215
Credit: 208,696,464
RAC: 304
Australia
Message 1383285 - Posted: 21 Jun 2013, 6:50:17 UTC - in response to Message 1383163.  

My RAC went fron 41K on friday to 62K today

I don't do AP and my RAC is still continuing to fall.


Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.
Grant
Darwin NT
ID: 1383285 · Report as offensive
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 1383373 - Posted: 21 Jun 2013, 14:12:21 UTC - in response to Message 1383285.  
Last modified: 21 Jun 2013, 14:13:01 UTC

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

For an Nvidia card there are 2 ways to do it.
1) Run the Lunatics installer or,
2) Download the AP files from Lunatics and copy the exe file and dll's into your SAH directory, copy the NV aistub file and paste it into your existing app_info file deleting the extra <app_info> entries.(My preferred method, it makes for a neater app_info file)
3) Copy the suggested default entries that suit your card from the bottom of the Readme file into into the AP**NV.txt file (can't remember the full name).
4) Allow AP's in your SAH preferences file.
5) Run BOINC.

This takes about 15 minute to get a going machine.

Some tips.
The AP app is most efficient when running at least 2 instances per card.

While allocating a full CPU per AP instance is recommended, I've found that allocating 0.5 CPU's per instance make no difference to the run times and allows for more efficient CPU usage. Change the <av_cpus> entry in the AP section of you app_info file from 0.04 to 0.5.

Make sure you have at least V310.xx drivers installed

CPU usage and run times vary with the amount of blanking in the WU so run times can vary from ~25 mins. to ~1.5 hours.

Running 2 instances, the average crunching time per unit is around 45mins for a GTX580 and around 60 minutes for a GTX470

Have fun :)

T.A.
ID: 1383373 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13215
Credit: 208,696,464
RAC: 304
Australia
Message 1383489 - Posted: 21 Jun 2013, 21:40:31 UTC - in response to Message 1383373.  

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

Neither do i, so I might wait to see just how far my RAC does fall, then give it a go.
If it's just a case of running Lunatics, changing the CPU value from .04 to .5 it may be worth the effort.
Grant
Darwin NT
ID: 1383489 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51396
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1383495 - Posted: 21 Jun 2013, 21:46:35 UTC - in response to Message 1383489.  

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

Neither do i, so I might wait to see just how far my RAC does fall, then give it a go.
If it's just a case of running Lunatics, changing the CPU value from .04 to .5 it may be worth the effort.

That should be all it takes.
If you get errors, it might require one more copy and paste operation to add some parameters to override the default settings.
Should be no muss, no fuss with NV cards.
Excuse me if I am hard to understand at times.......I've had a difficult few lives.

ID: 1383495 · Report as offensive
Lionel

Send message
Joined: 25 Mar 00
Posts: 680
Credit: 563,640,304
RAC: 597
Australia
Message 1383498 - Posted: 21 Jun 2013, 21:49:30 UTC - in response to Message 1382463.  

Note: Data for last 6 days added and revised running average.


I’ve had a look at the data around v7 and v6 WUs and below is a quick observational analysis of my data.

Under v6, I was roughly averaging 100 credits per Work Unit (WU).
Under v7, it seems that the average is sitting around 75-80 credits per WU.
In looking at run-times and taking the outliers out, cpu run time was around 600-660 seconds (10-11 minutes) per WU for v6, and appears to be around 800-1100 seconds (13+ to 18+ minutes) for a v7 WU. CPU time seems to have gone up by a factor of 2-3 from 50-60 seconds for v6 to 90-180 seconds for v7.

So doing a quick Back of the Envelope (10.5/15.83=0.66) shows that from a WU processing/throughput capability, I can expect to do roughly 66% of the volume of WUs that I did before (for example, if I was doing 400 WUs per day under v6, I can now expect to do around 264 WUs per day under v7).

Looking at the impact on credit gives 0.66*0.775 = 0.514 or 51.4%. In essence I can expect that daily credit for v7 will drop to circa 51% of what I was getting under v6.

I am aware of the comments around “that the system needs time to settle down” and that “it thinks all the WUs coming back at the moment are easy, hence the low credit” however, if the system continues to perform as is, then I can expect to see no change from current trajectory.

To test the assumption, I have looked at credit per day for the last 20 days. Below is the data:


    2013.05.16 – 244,130
    2013.05.17 – 220,168
    2013.05.18 – 231,098
    2013.05.19 – 226,353
    2013.05.20 – 224,723
    2013.05.21 – 210,477
    2013.05.22 - 0
    2013.05.23 – 431,485
    2013.05.24 – 229,312
    2013.05.25 – 228,767
    2013.05.26 – 239,021
    2013.05.27 – 231,271
    2013.05.28 – 231,050
    2013.05.29 – 0
    2013.05.30 – 392,635
    2013.05.31 – 209,556
    2013.06.01 – 123,072
    2013.06.02 – 94,061
    2013.06.03 – 102,333
    2013.06.04 – 99,896
    2013.06.05 - 65,653
    2013.06.06 - 112,209
    2013.06.07 - 102,538
    2013.06.08 - 110,760
    2013.06.09 - 89,757
    2013.06.10 - 96,018
    2013.06.11 - 111,653
    2013.06.12 - 90,091
    2013.06.13 - 119,848
    2013.06.14 - 99,884
    2013.06.15 - 104,561
    2013.06.16 - 110,566
    2013.06.17 - 110,603
    2013.06.18 - 102,856
    2013.06.19 - 85,268
    2013.06.20 - 140,694
    2013.06.21 - 70,247
    2013.06.22 - 109,698





The average daily credit prior to migration was 221,878. Following migration on 1st June, the average daily credit is showing as 102,376 which is circa 46.1% of the previous daily average under v6.

I suspect that many are starting to see their RAC decline, but as RAC is a lagging indicator and is currently composed of v6 and v7 numbers, the end effect is being masked. As this mask evaporates and RAC asymptotes towards the underlying v7 values, I suspect that the grumblings in the forum may get louder.

Also, the average has come up slightly as "the system" has been sending me more "enhanced" wus to do. In looking at granted credit for these wus, they appear to be getting almost twice as much credit as a v7 wu.



Comment on comments

Whilst some are focusing just on credit, the issue is not about credit as such. It’s about recognition.

There are many distributed computing projects to which people contribute resources. The manner and means in which those projects recognise individual contribution is through a system that is based on and allocates credits. Some projects choose to recognise a person’s contribution more than other projects, thus they grant a higher credit rate per contribution for that project. In short, credits are effectively an indication of a person’s contribution to a project.

In the case of "the New Credit System" implemented by Seti, recognition of personal contribution has been reduced. At present, the indication is that recognition for effort is effectively half that of what it was prior to the new recognition system being employed.


ID: 1383498 · Report as offensive
Profile Gatekeeper
Avatar

Send message
Joined: 14 Jul 04
Posts: 887
Credit: 176,479,616
RAC: 0
United States
Message 1383499 - Posted: 21 Jun 2013, 21:55:10 UTC - in response to Message 1383489.  

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

Neither do i, so I might wait to see just how far my RAC does fall, then give it a go.
If it's just a case of running Lunatics, changing the CPU value from .04 to .5 it may be worth the effort.


No doubt that we NV users have an easier go of it than ATI folks. I started running the opencl AP's with the new installer release, but stopped when I ran into errors that were apparently caused by BOINC forcibly terminating jobs prior to their normal EOJ exit. I don't know whether or not a fix for that issue has been accomplished.
ID: 1383499 · Report as offensive
kittyman Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 9 Jul 00
Posts: 51396
Credit: 1,018,363,574
RAC: 1,004
United States
Message 1383503 - Posted: 21 Jun 2013, 22:09:57 UTC - in response to Message 1383499.  
Last modified: 21 Jun 2013, 22:10:15 UTC

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

Neither do i, so I might wait to see just how far my RAC does fall, then give it a go.
If it's just a case of running Lunatics, changing the CPU value from .04 to .5 it may be worth the effort.


No doubt that we NV users have an easier go of it than ATI folks. I started running the opencl AP's with the new installer release, but stopped when I ran into errors that were apparently caused by BOINC forcibly terminating jobs prior to their normal EOJ exit. I don't know whether or not a fix for that issue has been accomplished.


Did you try the advice in the last couple of posts in this thread?
It worked for me.
Excuse me if I am hard to understand at times.......I've had a difficult few lives.

ID: 1383503 · Report as offensive
TBar
Volunteer tester

Send message
Joined: 22 May 99
Posts: 5204
Credit: 840,779,836
RAC: 2,768
United States
Message 1383513 - Posted: 21 Jun 2013, 22:33:50 UTC - in response to Message 1383499.  

Curious, why don't you do AP work units?

Going by the enquiries on the forum they take a lot more effort to set things up to process without issues.

Like you I was suspicious of crunching AP's but the crashing RAC of my major cruncher since June 1 forced me to do something. The effort required I think depends on whether you have an ATI or an Nvidia card(s). Setting up an ATI card seems to involve a bit more fuss. But I can't really comment as I don't have any ATI cards

Neither do i, so I might wait to see just how far my RAC does fall, then give it a go.
If it's just a case of running Lunatics, changing the CPU value from .04 to .5 it may be worth the effort.


No doubt that we NV users have an easier go of it than ATI folks. I started running the opencl AP's with the new installer release, but stopped when I ran into errors that were apparently caused by BOINC forcibly terminating jobs prior to their normal EOJ exit. I don't know whether or not a fix for that issue has been accomplished.

I told you about the 'fix' for that issue back here, on page one, http://setiathome.berkeley.edu/forum_thread.php?id=71889&postid=1379669#1379669

It works.
ID: 1383513 · Report as offensive
Profile Donald L. Johnson
Avatar

Send message
Joined: 5 Aug 02
Posts: 8240
Credit: 14,654,533
RAC: 20
United States
Message 1383596 - Posted: 22 Jun 2013, 6:39:03 UTC

I have noticed, in the past couple of days, that mid-range AR tasks running about 25,000 seconds validate with 105-120 credits, about what thee got under S@H v6, whereas right after v7 rollout they were getting about 80 credits.
Donald
Infernal Optimist / Submariner, retired
ID: 1383596 · Report as offensive
juan BFP Crowdfunding Project Donor*Special Project $75 donorSpecial Project $250 donor
Volunteer tester
Avatar

Send message
Joined: 16 Mar 07
Posts: 9786
Credit: 572,710,851
RAC: 3,799
Panama
Message 1383636 - Posted: 22 Jun 2013, 12:38:54 UTC

My RAC finaly reaches the bottom of the sea, and start to rise (very...very slowly) worst case cenario 150K, not rises to 159K was 450K before.

As you all could imagine I´m so happy about that! I love Creditnew.

But i´m still crunching SETI until the last beer!
ID: 1383636 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13215
Credit: 208,696,464
RAC: 304
Australia
Message 1383733 - Posted: 22 Jun 2013, 21:05:36 UTC - in response to Message 1383636.  


Still falling here, but not as fast as it was.
Grant
Darwin NT
ID: 1383733 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 22842
Credit: 261,360,520
RAC: 489
Australia
Message 1383750 - Posted: 22 Jun 2013, 21:46:54 UTC - in response to Message 1383733.  

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

Cheers.
ID: 1383750 · Report as offensive
Grant (SSSF)
Volunteer tester

Send message
Joined: 19 Aug 99
Posts: 13215
Credit: 208,696,464
RAC: 304
Australia
Message 1383756 - Posted: 22 Jun 2013, 22:14:42 UTC - in response to Message 1383750.  

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

I'm doing MB only, people doing AP as well won't take as big a hit.
It's MB that's had it's payout slashed.

Grant
Darwin NT
ID: 1383756 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 22842
Credit: 261,360,520
RAC: 489
Australia
Message 1383767 - Posted: 22 Jun 2013, 22:52:46 UTC - in response to Message 1383756.  

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

I'm doing MB only, people doing AP as well won't take as big a hit.
It's MB that's had it's payout slashed.

I only did AP's on my 2500K CPU, when they came along and you could grab 1, but now it seems to be all it does these days, with the odd MB V6 or V7 thrown in now and again to break the monotony (which is why it's now my higher RAC rig again).

Cheers.
ID: 1383767 · Report as offensive
Terror Australis
Volunteer tester

Send message
Joined: 14 Feb 04
Posts: 1817
Credit: 262,693,308
RAC: 44
Australia
Message 1383770 - Posted: 22 Jun 2013, 22:53:26 UTC - in response to Message 1383750.  
Last modified: 22 Jun 2013, 22:54:25 UTC

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

Cheers.

My MB only cruncher is back on the rise. Its old V6 RAC was ~32k, it bottomed at around 18k but is now back up to 19.5k.

The V7 credits seem to have increased in value since last week.

T.A.
ID: 1383770 · Report as offensive
bill

Send message
Joined: 16 Jun 99
Posts: 861
Credit: 29,352,955
RAC: 0
United States
Message 1383771 - Posted: 22 Jun 2013, 22:53:47 UTC - in response to Message 1383756.  

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

I'm doing MB only, people doing AP as well won't take as big a hit.
It's MB that's had it's payout slashed.


Mines gone up. Only using the GPUs.

Crunching Milkyway on the cpu at the momenr.
ID: 1383771 · Report as offensive
Profile Wiggo
Avatar

Send message
Joined: 24 Jan 00
Posts: 22842
Credit: 261,360,520
RAC: 489
Australia
Message 1383774 - Posted: 22 Jun 2013, 22:59:57 UTC - in response to Message 1383770.  

I seem to have hit bottom now.

Not bad, from 90K down to 54K, but things seem to be slowly on the up.

Cheers.

My MB only cruncher is back on the rise. Its old V6 RAC was ~32k, it bottomed at around 18k but is now back up to 19.5k.

The V7 credits seem to have increased in value since last week.

T.A.

I've noticed the same thing myself, about a 25-30% increase by the looks so far but they still have a long way to go yet (they may get back to right by Xmas).

Cheers.
ID: 1383774 · Report as offensive
EdwardPF
Volunteer tester

Send message
Joined: 26 Jul 99
Posts: 389
Credit: 236,772,605
RAC: 374
United States
Message 1383789 - Posted: 23 Jun 2013, 0:37:58 UTC - in response to Message 1383774.  

Here are 3 identical crunchers spanning the changeover - for what is's worth (if anything) all 3 have identical CPU load.

(I hope this works!!)

Ed F



https://dl.dropboxusercontent.com/u/42596478/SETI%20Graph%20CPU%202.jpg
https://dl.dropboxusercontent.com/u/42596478/SETI%20Graph%20CPU%204.jpg
https://dl.dropboxusercontent.com/u/42596478/SETI%20Graph%20CPU%206.jpg

ID: 1383789 · Report as offensive
Michael Cruz
Avatar

Send message
Joined: 23 Jan 00
Posts: 35
Credit: 323,653,343
RAC: 30
United States
Message 1383818 - Posted: 23 Jun 2013, 3:55:39 UTC

I have also noticed the same "bottoming" out. My RAC dropped from ~102K to about ~60K and for the last few days I have seen a slight upturn of RAC on most of my rigs. I'm now "up" to about 69K. I'm going to have to add a couple more computers to the farm :)



Seti Classic: 204,777 WU /113.636 Yrs
ID: 1383818 · Report as offensive
Previous · 1 . . . 10 · 11 · 12 · 13 · 14 · 15 · Next

Message boards : Number crunching : Observation of CreditNew Impact


 
©2021 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.