Posts by Honza

Message boards : Number crunching : Who is overclocking their machine?

( Message 2165 )
Posted 3857 days ago by Honza

I used to not overclock machines but when upgraded to Core 2 Duo E6600s, it is so tempting to overclock that I could not prevent myself from running it at +30%.
http://docking.utep.edu/show_host_detail.php?hostid=92

I'm also running CPDN for years which is known to be very sensitive to overall machine stability.

re: David Ball's post - CPDN has been monitoring CPU/OS differences in results over the years and it a known issue (and not an easy one solve).

Message boards : Application Info : Is there a difference between the work done by Lin and Win WUS?

( Message 620 )
Posted 3970 days ago by Honza

Is there a possibility to measure the scientific output at the servers?

I think this question suggest possibility of credit based on server-side (a-la CPDN, Einstein) - fixed credit per work done and abandom ill-numbered benchmarks.
Credit award was never fair among project (weak correlation) and differences withing project when difference platforms gives different numbers doesn't help the situation. This can be solved - on some projects - using server-side credit award, which also eliminate those with unrealistic benchmarks.

EDIT: ha, found it on the other thread ...at least my assumption was correct.

Mentioning early 70's and Fortran77...
I'm not sure how complex the code is but probably way more simpler than climate models...which were developed in Fortran over the decades as well.
A cooperation with CPDN on compilation of Fortran stuff and cross-platform consistency may be usefull...CPDN is quite mature among BOINC projects and porting from 64-bit Cray over to Linux and 32-bit Windows was done even prior BOINC era.

Message boards : Number crunching : zero credits for valid result??

( Message 448 )
Posted 3977 days ago by Honza

Honza,
Could you send me Bernhard's email address by normal email?
Andre

Done.

Message boards : Docking@Home Science : What is the main goal of Docking@home ?

( Message 404 )
Posted 3978 days ago by Honza

Thanks, that answered (not only) my questions very well.

Message boards : Number crunching : Too Many WUs!

( Message 333 )
Posted 3978 days ago by Honza

Maybe Docking should send a max of 10 WUs. Last time I contacted the server I got 29. This is work for 3 days and I don`t WANT to stop all other Projekts because of Docking.

Sounds like a good idea.
I'm quite sure SETI uses this method to limit fetching large portion of results at once (to prevent overloading servers).
(it's a pain if you want to fetch full cache on a BOINC installation intended for non-internet machines...takes hours due to deffered communitation)

Message boards : Number crunching : Charmm 5.02

( Message 332 )
Posted 3978 days ago by Honza

@ JShadic - No heartbeat means that BOINC core is having trouble finiding science application alive. Application should send "alive" message periodically.

It can happen when another task takes too much CPU cycles hence BOINC project application doesn't get any since it runs on low priority.
Or, it can happen on windows machines when the clock on XP is updated to the correct time (done automatically in Windows), and BOINC core gets out of sync with application.

More on wiki
http://boinc-wiki.ath.cx/index.php?title=No_heartbeat_from_core_client_-_exiting

Message boards : Number crunching : zero credits for valid result??

( Message 325 )
Posted 3978 days ago by Honza

We might do the same if we find out how... If anybody can offer any help as to how to do this it will be appreciated. We are not boinc-ignorants, but also not complete experts on the system yet :-)
Andre

I believe Bernhard of RCN used semi-automatic script for that; better ask him.

Message boards : Number crunching : zero credits for valid result??

( Message 302 )
Posted 3979 days ago by Honza

I'm not getting what would you like to change? A higher number for max errors?

btw, I would not expect last results to get back valid as last computer has 3 results invalid so far and not a single one valid.

(I have also 2 valid results belonging to non-validated work unit...those WU were cancelled in early stages. Also, I have 3 completed results with errors on final upload. There is nothing much to "change" - only generation correct work units and better application...this is what alpha stage suppose to do. Albeit, some projects during alpha stage gives credit to machines that completed valid results regardless if WU as validated as a whole.)

Message boards : Docking@Home Science : What is the main goal of Docking@home ?

( Message 293 )
Posted 3979 days ago by Honza

Thanks for the answers, it is more obvious what Docking project is doing in term of science.

There are still points not very clear to me so let me restate my questions:
Is Docking pure non-profit effort or is project going to make money from any results?
What is the connection between Predictor and Docking?

10)

Message boards : Number crunching : How long are the Work Units going to be?

( Message 275 )
Posted 3980 days ago by Honza

If only running 1 thread then a HT CPU is really only running at 50% of its capacity, not 50% speed.

I can't agree there.
HT is not a dual-core, nor dual-CPU...actually far from it.
If you are running without HT (or running only single thread/application), you are at worst loosing 25% of total performance.
Older Xeon are known to do poor on HT and there are cases when running both threads on HT actually makes whole system running slower and overall performance goes down (due to high latencies to CPU cache and memory, all running via FSB etc.); for example on highly optimalized code like Einstein akosf's optimalization was.

You may find some interesting vaules in my BOINC table .
(I have been running Intel HT machine but abandomed them 1+ years ago when I went dual-cores and again upgraed recently to Conroes).

Next 10 posts