Posts by -= Vyper =-


log in
1) Message boards : Number crunching : I'm falling, I bought a parachute. From 100% AP, to 100% MB. (Message 1492249)
Posted 27 days ago by Profile -= Vyper =-
Well i can only chime in that i noticed this early on.
This is quote from me in our team forum 254 Days ago..


"Well now it's confirmed that running AP on CPU also gives credit boost.
It's a 3930K machine with HT turned off (Yes it still sucks too much to be useful) and is running Lunatics 0,41 optimised executables.

On one of my hosts i'm running s@h7 and ap6 in parallell and compares credit/second on the cpu..

A AP WU ~ 0,041 - 0,043 credits/cpu second
A S@H7 WU ~ 0,019 - 0,021 credits/cpu second

So by only adjusting that our primary tasks are AP tasks and tick the "If no work for selected applications is available, accept work from other applications?" box.
That would make our boxes out there try to download AP's first of all. I didn't Think i would see the light of the day that i tried to encourage you to run AP here in GPUUG but as their now creditnew invention goes on"

Notice that i'm Writing CPU second and not GPU so even the CPUs got more credits doing AP tasks.

This is a shame and should be addressed and fixed asap in my opinion! I mean how hard can it be to adjust the earning (credit) parameters of MB7 up to atleast AP levels. MB6 is of the charts anyhow so they don't even need to try to match it to older multibeam.

I hope you land soft Sten! Thanks for bringing this topic up that has been going on since around June last year.

Kind regards Vyper
2) Message boards : Number crunching : The (Preliminary and Premature) Green Bank SERENDIP Fundraiser - Help SETI@Home Build a New (non-Arecibo) Receiver! (Message 1492156)
Posted 27 days ago by Profile -= Vyper =-


Salut and thank you, founder of GPUUG. I never would have done anything like this without this team that you started. :^)

And here is mine in return...


Aah hush. I devoted hard work getting devoted people like you!
Thank you for beeing so generous so we can extend our hobby more years to come :)

//Vyper
3) Message boards : Number crunching : The (Preliminary and Premature) Green Bank SERENDIP Fundraiser - Help SETI@Home Build a New (non-Arecibo) Receiver! (Message 1491856)
Posted 28 days ago by Profile -= Vyper =-
Ok here is mine:

Dear -= Vyper =-,

Thank you for your gift of $234.00 via Visa on 03/20/2014.
Your gift will benefit the following area:

SETI@home - $234.00

Your confirmation number is: 130690

I targeted the donation to the GPUUG Campaign!

Kind regards Vyper
4) Message boards : Number crunching : Intel GPU error (Message 1471322)
Posted 75 days ago by Profile -= Vyper =-
Anyone know what to write?
Downloaded Mikes latest AP Intel OpenCL 2083 and gets this!

I've updated the app_info as what is written into the aistub file!

2014-01-31 21:51:32 | | OpenCL CPU: Intel(R) Core(TM) i5-3570K CPU @ 3.40GHz (OpenCL driver vendor: Intel(R) Corporation, driver version 3.0.1.10878, device version OpenCL 1.2 (Build 76413))
2014-01-31 21:51:32 | SETI@home | Found app_info.xml; using anonymous platform
2014-01-31 21:51:32 | | app version refers to missing GPU type intel_gpu
2014-01-31 21:51:32 | | app version refers to missing GPU type intel_gpu

EDIT: I have no screen attached but GPU Caps Viewer can use the Intel Opencl gpu in it's test and its working!

Kind regards Vyper
5) Message boards : Number crunching : A new kitty auction.......... (Message 1423022)
Posted 197 days ago by Profile -= Vyper =-
Oh sorry!
Thanks for clarifying. Haven't read up.

//Vyper
6) Message boards : Number crunching : A new kitty auction.......... (Message 1423007)
Posted 197 days ago by Profile -= Vyper =-
Did our GPUUG member MJS bid for the Racoons?
Wtf!

//Vyper
7) Message boards : Number crunching : A new kitty auction.......... (Message 1422281)
Posted 199 days ago by Profile -= Vyper =-
Here is my two donations. Even though i was late to the show. (Late vacation from Malta)..

"Dear -= Vyper =-,

Thank you for your gift of $74.00 via Mastercard on 09/29/2013.
Your gift will benefit the following area:

SETI@home - $74.00

Your confirmation number is: 117737"

"Dear -= Vyper =-,

Thank you for your gift of $33.00 via Mastercard on 09/29/2013.
Your gift will benefit the following area:

SETI@home - $33.00

Your confirmation number is: 117735"


Kind regards Vyper
8) Message boards : Number crunching : A new kitty auction.......... (Message 1421598)
Posted 200 days ago by Profile -= Vyper =-
.. and i will donate anyhow whichever ..
Just for the record.

//Vyper
9) Message boards : Number crunching : A new kitty auction.......... (Message 1421587)
Posted 200 days ago by Profile -= Vyper =-
And i will first change some BTC -> $ .. So 100$ from me.. no later than tomorrow.

//Vyper
10) Message boards : Technical News : Colo! (Mar 28 2013) (Message 1353605)
Posted 377 days ago by Profile -= Vyper =-
This is the best thing that have happened in years!
Really good work and good move to keep this Project alive and sound.

Perhaps all work could get to their ends without timing out and requiring resends all the time like Before!

I'm so thrilled and thankful!

Kind regards Vyper
11) Message boards : Technical News : Upgrades (Jan 30 2013) (Message 1332948)
Posted 441 days ago by Profile -= Vyper =-
Matt!

Check the motherboard to see what the brand is and model no., perhaps it's a watchdog timer in bios that locks the machine (fails to reset) when the context/interrupt switching is high.
Todays virtualization Technologies stresses that to the max.
If so tweak all buffers and NIC's to reduce the thing mentioned above and if it "lasts" longer then you're on the right track!

Kind regards Vyper
12) Message boards : News : Scheduler crashed. (Message 1308909)
Posted 510 days ago by Profile -= Vyper =-
Maybe a openvpn server and a pdu is something to invest in so you could reboot stuff from example a android phone. :)

Perhaps should persuade Slavac to set it up on the gpuug site.

You always want a safe connection to a pdu so you could restart the server while eating the turkey :D

Happy thanksgiving over there.
Kind regards Vyper
13) Message boards : Number crunching : Average RAC for GTX 670/680.? (Message 1294940)
Posted 550 days ago by Profile -= Vyper =-
Hi there.

One of my machines have a single non OCed 670 so i'll post some numbers.
Running 2 WU's in parallell = 0.5

Lower AR shorties takes around 130 seconds / WU gives around 25 credits. (AR = 4.319562)
Higher AR shorties takes around 148 seconds / WU gives around 28 credits. (AR = 2.716736)
Upper MID AR takes around 272 seconds / WU, gives around 50 credits. (AR = 1.122237)
MID AR takes around 580 seconds / WU, gives around 118 credits. (AR = 0.420551)
Lower MID AR takes around 856 seconds / WU, gives around 165 credits. (AR = 0.238529)

So that equals to:
86400/130 * 2 * 25 = 33230 credits
86400/148 * 2 * 28 = 32691 credits
86400/272 * 2 * 50 = 31764 credits
86400/580 * 2 * 118 = 35155 credits
86400/856 * 2 * 165 = 33308 credits.

If you add them all 33230+32691+31764+35155+33308 and divides by 5 = Median of 33283 credits and if you take into consideration that main distribution is shorties then you need to slighly lower the RAC..

There you have it then.
A single GTX670 stock clock gives around 33100 in RAC by itself 24/7/365.
Running setiathome enhanced x41z, Cuda 4.20

Mbcuda.cfg =
processpriority = abovenormal
pfblockspersm = 15
pfperiodsperlaunch = 200

Running on a 2600K @ 4GHZ non HT.

Kind regards Vyper

14) Questions and Answers : GPU applications : Fermi and pre-Fermi mix? (Message 1211817)
Posted 748 days ago by Profile -= Vyper =-
Yes

http://vyper.kafit.se/wp/index.php/2011/02/04/running-different-nvidia-architectures-most-optimal-at-setihome/

Kind regards Vyper
15) Message boards : Number crunching : Analogue Website (Message 1192180)
Posted 800 days ago by Profile -= Vyper =-
Hey

Please jump in and navigate a few dudes which have a website which they call a "Analogue Website".

You can watch a video, fax them messages, like, Beer (after 6 PM) etc etc.
It's quite fun actually.
Warning for a pixly posters when in 18+ mode! :D

Kind regards Vyper

http://madebypeterandrasmus.com/
16) Message boards : Cafe SETI : Analogue Website (Message 1192178)
Posted 800 days ago by Profile -= Vyper =-
Hey

Please jump in and navigate a few dudes which have a website which they call a "Analogue Website".

You can watch a video, fax them messages, like, Beer (after 6 PM) etc etc.
It's quite fun actually.
Warning for a pixly posters when in 18+ mode! :D

Kind regards Vyper

http://madebypeterandrasmus.com/
17) Message boards : Number crunching : Can I select which WU/project goes to which GPU? (Message 1190289)
Posted 807 days ago by Profile -= Vyper =-
there is always the (Vyper style) option to go to old drivers


Lol! I'm refered to as Vyper style!
Wth! It's the only way i can fully load those beasts on that particular platform and god know why really.
As long as it works it works because i don't think that the speed benefits would be that much better on pre-fermi cards anyway.

It's simply as running the platform on what makes it works the best and for me it is to use the good old 190.38 driver with a quad cpu feeding octo gpu with no cpu tasks congesting the cpu pipes to free it for the GPU's..

Simply as that :)

Kind regards Vyper
18) Message boards : Number crunching : Idea for Boinc..... (Message 1166637)
Posted 899 days ago by Profile -= Vyper =-
But the client can't send a compressed request file unless it is known that the server can accept compressed request files. BOINC != SETI.


Ofcourse!

This needs to be implemented as follows.

1. Those who "design" BOINC needs to get this information and apply this on their application.
2. They need to test that the solution works and figure out a futureproof design of the implementation.
3. When they're done, they can release a new "alpha" server version to be installed followed by an "alpha" BOINC client.
4. Seti would update their "Boinc" seti installation on the beta site and request people to try the latest "alpha" boinc client installation.
5. Tests would be performed for perhaps two weeks to see that it works as intended or they would report it to the boinc developers.
6. We pretend that there were errors and a new version has been released. Seti updates beta site and people are encouraged to try the new alpha Boinc client again.
7. Things worked as intended for a week without any hickup. Seti starts to design a implementation protocol and schedule to implement this on seti main.
8. One week has passed with the design and is about to be implemented on the next Tuesday outage. People are beeing informed about the upgrade and could be encouraged to download a new Boinc client and two days before this Tuesday outage occurs the new Boinc alpha client which hopefully hasn't have had any sort of "side effect" is beeing upgraded to be a "latest stable" version.
9. Seti is monitoring their sides and perhaps notices an "ooops" and noticed that the server process which monitor and processes the in/out flow of status files is beeing "capped" by a high load and starts to fall behind.
10. They implement four virtual servers (or instances, scripts on four different loaded machines) , two for each drawer which by filtering is checking for differnt files thus the handling is divided by four different processes guarding those in/out directories and notices that it would now work even if they manage to have a weeks outage that could and will occur in the future by Murphys Law.
11. Things are smooth but they need to trim the servers so they wouldn't drop connections if things aren't as sleek as it should behave.
12. They are an a roll within a week and all major flaws has been ironed out, the patch has been implemented and they never again looked back on uncompressed datafiles and the bandwidth/transactions/requests/resends wouldn't be as capped as it were two months earlier.


There you have it, a small miniproject of a implementation basis and schedule for Boinc infrastructure!
It wasn't that hard was it! ;-)

Kind regards Vyper

P.S All this are based on pure speculations from my PoV D.S
19) Message boards : Number crunching : Idea for Boinc..... (Message 1166544)
Posted 899 days ago by Profile -= Vyper =-
I just wanted to correct myself about the statement of the approach.
Actually the boinc client which send units to the servers could be named in a way in the filename so the servers know which boinc id which is sending compressed files.
In that way the servers automatically know which client is using compression just to list the filename for example:

_Idname_filename.zip
_3050453_xxxxxxx.zip

In that way it's easy implemented that boinc at the other end is responding to compressed files as well so the servers can tag 3050453 computer id with compressed files to be sent and recieved.
And if everybody thinks of security and manipulation there is no issue because pure uncompressed files is not secure either at all so there is no problem with that approach either.

//Vyper
20) Message boards : Number crunching : Idea for Boinc..... (Message 1166416)
Posted 900 days ago by Profile -= Vyper =-
And i endorse my opinion of compressing the data files before they are sent!
Cpu's are getting faster and faster, ram cheaper and cheaper. Just look on the latest three servers that the s@h community has sponsored.

Every server has 96GB of ram!

That's alot.

I just tried compressing sched_request and it was 510KB and the file became 17Kb in size.
The stress involved on the cpu's on s@h would increase ofcourse but it could be implemented soo simple, the upload and download servers do their work as usual but they can have a script that watches the up/download queues and every minute a ls *.zip process can be run to pipe the zip files to a decompressor and then delete the file afterwards.
The routine for this doesn't need to be run at the up/download servers, another server internally could just "poll" the directorys and handle the compressing actually freeing that duty from up/download servers.
Ofcourse you need to have some "how much space is there left" in the batch script but this could be implemented as well with a "no biggie" in the boinc infrastructure.
So could we implement Marks and my ideas then the bandwidth would drop tremendously keeping more bandwidth available for true workunit transfers as it should!

(edit)And you need to have a routine that polls what version of Boinc you are using so that this would be plausible. So that the scripts can watch for that version number so it could be supported otherwise leave it uncompressed as usual..(/edit)

Thoughts and ideas ppl?

Kind regards Vyper


Next 20

Copyright © 2014 University of California