Welcome to MilkyWay@home

new hard drive installed

Message boards : Number crunching : new hard drive installed
Message board moderation

To post messages, you must log in.

Previous · 1 · 2 · 3

AuthorMessage
Donnie
Avatar

Send message
Joined: 19 Jul 08
Posts: 67
Credit: 272,086,462
RAC: 0
Message 33653 - Posted: 24 Nov 2009, 17:21:55 UTC - in response to Message 33644.  

So if I may be so bold as to ask, what exactly is the new credit 'scheme'?
What will the net reduction- I assume it's going to be a reduction ;)- be?


I believe it can be found here -

Proposed new BOINC credit system:

New Credit Scheme

ID: 33653 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
arcturus

Send message
Joined: 20 Nov 07
Posts: 54
Credit: 2,663,789
RAC: 0
Message 33654 - Posted: 24 Nov 2009, 17:51:03 UTC

LOL I love how you point mongers get riled up at the first sign of credit adjustment. Face it, you'll still be here so long as it remains more generous than other projects that are up and active.
ID: 33654 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Dan T. Morris
Avatar

Send message
Joined: 17 Mar 08
Posts: 165
Credit: 410,228,216
RAC: 0
Message 33656 - Posted: 24 Nov 2009, 17:55:53 UTC - in response to Message 33623.  
Last modified: 24 Nov 2009, 17:59:16 UTC

Lets wait how the NEW credit reduction is gonna be
Or its gonna be here i come collatz
I even am thinking of turning off all projects as long as mister Anderson is alive.


The irony is that people complained for years about cross-project parity, and when steps are made to try and address that problem more complaints come across :)

The new credit scheme is honestly just a way to track credit in a project independent manner so when new projects come along they don't have to deal with all the credit issues.

I know you all want to think there's some big brother keeping you down in the credit ranks, but honestly Dave and the rest of us really want to make it so that credit is distributed correctly across projects in some kind of fair manner.

Honestly, we're really not so smart, which is why it's not been so good in the past. The new credit scheme is to try and make things better, really :)


When this happens I foresee a large exit happening me included. Hows that for cross project comparison. But hey its only me leaving .. so I would imagine you will find many of those 100 wu a day folks who will love it.

And the only ones who were complaining were the ones who do not have the investment in computers like some of us have. It has been a nice run here but hey we all have to do what makes us and DA happy. Chow
ID: 33656 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile GalaxyIce
Avatar

Send message
Joined: 6 Apr 08
Posts: 2018
Credit: 100,142,856
RAC: 0
Message 33657 - Posted: 24 Nov 2009, 18:18:38 UTC - in response to Message 33623.  

... but honestly Dave and the rest of us really want to make it so that credit is distributed correctly across projects in some kind of fair manner.

The only fair way is to keep fair credit earned in this project - in this project. I am not going to crunch here if you are going to distribute my credit to SETI or any other project. Where I earn credit is mine to choose. Tweak it here to help some other project and you'll be tweaking me out of here along with all the others who really don't want this adverse manipulation of credits.

But then again I made the favicon , so I have to stay here ;)


ID: 33657 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 33664 - Posted: 24 Nov 2009, 19:25:45 UTC - in response to Message 33635.  


Not going to say what I think about that. But to me you wouldn't question his dicisions with him grading your thesis papers.


If you want I can wait like 2 weeks for the last form to be filled out and have the PhD sitting on my desk :P

I've already passed the defense so there's not much that can be done, lol.

But honestly, from our discussions the scheme is to give the GPU applications the credit they should be getting. Which in our case is quite a lot more than what CPUs are getting, because our application scales so well to GPUs.

If the credit scheme doesn't end up awarding something somewhat similar to what you guys are getting now we probably won't end up using it until it's modified so it does.
ID: 33664 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
PeteS

Send message
Joined: 19 Mar 09
Posts: 27
Credit: 117,670,452
RAC: 0
Message 33666 - Posted: 24 Nov 2009, 19:35:20 UTC

Congrats to the new Dr. and great to see some positive movement in MW! I'm back to MW now after quite a long break!

Hope you still have the following on todo-list:
- GPU only side or very much longer wu's
- increase the limit of wu's from 6/cpu to xxx/gpu
- These fitness calculations do not reveal what the milky looks like, do we have graphical rendered presentations? ;) I wish to see my contribution to science in more detail..
- Having working GPU scheduling in Boinc Manager and an updated ATI-client that works on all OS/driver combinations without VPU errors!
- Having active admins / project personnel on forums surely help in attracting volunteers and makes us feel appreciated. Someone is actually listening!

I'm in this all for the science! Have to also wonder about this credit issue, as long as it's the same for all participants it should not matter what the amounts are. Constant adjustments just make people angry. Like many have stated, credits are just numbers, they are not worth anything! :)
ID: 33666 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Bryan

Send message
Joined: 26 Jun 09
Posts: 47
Credit: 276,827,695
RAC: 0
Message 33668 - Posted: 24 Nov 2009, 20:36:19 UTC - in response to Message 33666.  

Hope you still have the following on todo-list:
- GPU only side or very much longer wu's
- increase the limit of wu's from 6/cpu to xxx/gpu


I would certainly vote for that. With a 5870 on a quad I have a whole 8 minutes of cache :( Also, I don't like having to hit the server every 20 seconds with up/dl.

Bryan

ID: 33668 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Beyond
Avatar

Send message
Joined: 15 Jul 08
Posts: 383
Credit: 729,293,740
RAC: 0
Message 33671 - Posted: 24 Nov 2009, 21:22:37 UTC - in response to Message 33668.  

Hope you still have the following on todo-list:
- GPU only side or very much longer wu's
- increase the limit of wu's from 6/cpu to xxx/gpu

I would certainly vote for that. With a 5870 on a quad I have a whole 8 minutes of cache :( Also, I don't like having to hit the server every 20 seconds with up/dl.

This issue was extensively discussed here:

http://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=1171

So far there hasn't been any official input on the subject. Some excellent solutions were suggested by Crunch3r, Extraterrestrial Apes and Paul Buck among others.

Here's two of the proposed ideas:

From ET:

The ideal solution would be a perfect balance between latency and throughput. This is something only the project staff could determine: if they'd like to run more searches in parallel, they need higher throughput. If they're waiting for results to return they'd need faster turn around times / lower latencies.

That leads me to the conclusion that it would be a good idea to tie the allowed number of concurrent WUs to a hosts average turn around time, a value easily accessible to the server. The project could set the desired value quite flexible / dynamically (within sane limits), depending on the current search state. The allowed WUs per host could be updated daily to keep the number of additional database requests low.

MrS

The other by Crunch3r:

No-one want's to single out the CPUs. What we need is max_jobs_on_host_gpu=200 ...
that should reduce the server load a bit since the clients won't hammer it every 60 sec..

But you really should ytake a look at the whole thread. Hopefully we'll get some feedback from the Admins.
ID: 33671 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Ziffen63

Send message
Joined: 26 Mar 09
Posts: 7
Credit: 702,781
RAC: 0
Message 33674 - Posted: 24 Nov 2009, 22:24:37 UTC

Travis, what kind of hardware is the project running on? What hard drives are you using?
ID: 33674 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Ironworker16
Avatar

Send message
Joined: 31 Jan 09
Posts: 31
Credit: 69,908,565
RAC: 0
Message 33679 - Posted: 25 Nov 2009, 0:37:20 UTC - in response to Message 33639.  

MW is Back !!
That's the Main Thing !!

Crunching WUs for a project you have interest in is way better than staring at an empty Task Queue, regardless whether you get unpteem extra credits than down the street or simply parity or a few less (which simply encourages the optimizations). In any case...

MW is Back !!

(Travis: Slather on an extra dose of Skin Hardener, and keep an eye on "our" new Server Hard Drives !! And... THANKS !!)


I agree, I'm not going anywhere. Also looking the software updates.
ID: 33679 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile verstapp
Avatar

Send message
Joined: 26 Jan 09
Posts: 589
Credit: 497,834,261
RAC: 0
Message 33682 - Posted: 25 Nov 2009, 0:57:34 UTC - in response to Message 33679.  

Also looking the software updates.


Perhaps my 5870 will finally get to do something other than Collatz...

Cheers,

PeterV

.
ID: 33682 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
BarryAZ

Send message
Joined: 1 Sep 08
Posts: 520
Credit: 302,524,931
RAC: 15
Message 33684 - Posted: 25 Nov 2009, 0:58:26 UTC - in response to Message 33584.  

Before those next steps, perhaps a 'viability study' -- see if you can keep the servers up and running reliably for say 168 hours or more before you mess with other changes.

Here we are 24 hours 'back to life' and things went bad again.

As to the credit schema thing, it presents a real problem, simply because of multiple variables.

1) CPU performance
2) GPU (ATI) performance
3) GPU (CUDA) performance
4) Application code performance at the project level for CPU, ATI, CUDA
5) Then of course the soft factors
A) Value of science across multiple project
B) 'Social' value of science
C) Factors to encourage new and low resource projects
6) Factors to reward projects for reliability
7) Factors to reward projects for user responsiveness

I can seriously muddy the waters up quite nicely. Currently the award 'schema' is pretty random -- and given the population, that might well be a good thing.


Me too :) I think everything should be fine unless the construction starts to cause a lot more vibration, but that shouldn't happen as it's 99% done.

Our next step is to get the server code upgraded so we can recognize ATI GPUs and also test out the new BOINC credit scheme. I'm hoping that'll happen tomorrow. We'll also be recompiling the feeder with some changes that I think should help keep work flowing and improve the performance of the webpage.

--Travis


ID: 33684 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile banditwolf
Avatar

Send message
Joined: 12 Nov 07
Posts: 2425
Credit: 524,164
RAC: 0
Message 33689 - Posted: 25 Nov 2009, 20:02:27 UTC - in response to Message 33614.  

Lets wait how the NEW credit reduction is gonna be
Or its gonna be here i come collatz
I even am thinking of turning off all projects as long as mister Anderson is alive.


Here we go again. Playing with the credits again, and I was just thrilled to see the project up and running. Oh well I guess that the....cops are back before we even get going. lets see 4 credit changes in one year. Oh my.. Get the flame proof jackets ready folks!


hmmmm... I was considering running some MW again in a week or two when I hit 100k on rosetta, now I am going to definately wait to see how the credits work out. (also see if the server doesn't go kerpow again)

Didn't take as long as I thought.
Doesn't expecting the unexpected make the unexpected the expected?
If it makes sense, DON'T do it.
ID: 33689 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 33691 - Posted: 25 Nov 2009, 20:21:58 UTC - in response to Message 33689.  

It looks like we may have a problem with the hard drive controller, because we took the new drive out and it seemed to pass all the diagnostics just fine.

I've started things back up and we're going to be keeping an eye on it, but it looks like we may have to order more hardware.
ID: 33691 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Divide Overflow
Avatar

Send message
Joined: 16 Feb 09
Posts: 109
Credit: 11,089,510
RAC: 0
Message 33695 - Posted: 25 Nov 2009, 20:27:58 UTC - in response to Message 33689.  
Last modified: 25 Nov 2009, 20:29:49 UTC

(also see if the server doesn't go kerpow again)

Didn't take as long as I thought.

They did say that they would be making some adjustments to the server to help improve things, as well as the anticipated updates to the server code and applications. I'd expect the server to go offline again when these changes are implemented.

Edit: Owch. More hardware problems. :(
ID: 33695 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile banditwolf
Avatar

Send message
Joined: 12 Nov 07
Posts: 2425
Credit: 524,164
RAC: 0
Message 33701 - Posted: 25 Nov 2009, 21:24:37 UTC - in response to Message 33695.  

(also see if the server doesn't go kerpow again)

Didn't take as long as I thought.

They did say that they would be making some adjustments to the server to help improve things, as well as the anticipated updates to the server code and applications. I'd expect the server to go offline again when these changes are implemented.

Edit: Owch. More hardware problems. :(


The code and all I understand that the server needs to be off for that. But they weren't messing with that.
ID: 33701 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 29 Aug 07
Posts: 115
Credit: 501,600,397
RAC: 5,019
Message 33703 - Posted: 25 Nov 2009, 21:45:56 UTC - in response to Message 33691.  

It looks like we may have a problem with the hard drive controller, because we took the new drive out and it seemed to pass all the diagnostics just fine.

I've started things back up and we're going to be keeping an eye on it, but it looks like we may have to order more hardware.


Any progress on updating the BOINC server software?

ID: 33703 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Travis
Volunteer moderator
Project administrator
Project developer
Project tester
Project scientist

Send message
Joined: 30 Aug 07
Posts: 2046
Credit: 26,480
RAC: 0
Message 33717 - Posted: 25 Nov 2009, 23:03:19 UTC - in response to Message 33703.  

Any progress on updating the BOINC server software?


Probably won't happen until early next week when the guy in labstaff who has been working on that with us is back from thanksgiving break.
ID: 33717 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Profile Bymark
Avatar

Send message
Joined: 6 Mar 09
Posts: 51
Credit: 492,109,133
RAC: 0
Message 33722 - Posted: 25 Nov 2009, 23:09:03 UTC - in response to Message 33717.  

Any progress on updating the BOINC server software?


Probably won't happen until early next week when the guy in labstaff who has been working on that with us is back from thanksgiving break.


Thanks, I hate updates, I hope it will be next year....
ID: 33722 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
zombie67 [MM]
Avatar

Send message
Joined: 29 Aug 07
Posts: 115
Credit: 501,600,397
RAC: 5,019
Message 33744 - Posted: 26 Nov 2009, 4:00:00 UTC - in response to Message 33717.  

Any progress on updating the BOINC server software?


Probably won't happen until early next week when the guy in labstaff who has been working on that with us is back from thanksgiving break.


Got it. Thanks for the update.

ID: 33744 · Rating: 0 · rate: Rate + / Rate - Report as offensive     Reply Quote
Previous · 1 · 2 · 3

Message boards : Number crunching : new hard drive installed

©2024 Astroinformatics Group