ARGOMENTO:

ReLeon ha scritto:

Quelli di Universe le provano proprio tutte, oppure non sono capaci di usare le gpu..

Le ultime parole famose di ReLeon

First test I will do myself obviously. After it I will add GPU application as "test" and it will be available only for persons who mark test applications in their settings.
Especially on first revisions it will be very unstable as GPU coding is completely new area for us, so advice to use it only by volunteers who can accept risk of crashes...

hanno riaperto Universe-test oppure sempre su Universe ?

sabayonino ha scritto:

hanno riaperto Universe-test oppure sempre su Universe ?

Da quello che si capisce da quanto ho riportato, dovrebbero mettere una casella di scelta per la beta nel profilo utente del sito di produzione

boboviz ha scritto:

sabayonino ha scritto:

hanno riaperto Universe-test oppure sempre su Universe ?

Da quello che si capisce da quanto ho riportato, dovrebbero mettere una casella di scelta per la beta nel profilo utente del sito di produzione

ah ecco ...dovrebbero ...ieri non c'era

sabayonino ha scritto:

ah ecco ...dovrebbero ...ieri non c'era

E neanche oggi, per quello
Direi dopo le vacanze.....

I lavori sul client Gpu procedono, anche se a rilendo a causa di alcune sfasature di risultato rispetto al client cpu.
E, rullate i tamburi....

This is why we work on OpenCL
We will continue the fight until we get app working

boboviz ha scritto:

I lavori sul client Gpu procedono, anche se a rilendo a causa di alcune sfasature di risultato rispetto al client cpu.
E, rullate i tamburi....

This is why we work on OpenCL
We will continue the fight until we get app working

li mortacci loro!!!! opencl.... mah speriamo che vada su entrambe le marche allora...
Mi preparo un app_config per l'occasione...

ReLeon ha scritto:

li mortacci loro!!!! opencl.... mah speriamo che vada su entrambe le marche allora...

E' MOLTO più probabile che vadano su entrambe le marche (e pure sulle Intel) in OpenCl rispetto a Cuda

New models and television program about U@H (in polish)
We have finished a bunch of models, which will be published shortly. However, still a lot more are waiting for investigation. In the case of ULXs we are working on the influence of more primordial matter composition on formation of these mysterious objects. The BHSPIN application concentrate on black holes spin (i.e., rotation), which influences the mass accretion processes (necessary for the detection of this latent objects), and emission of gravitational waves (a.k.a. propagation of space-time wrinkles). The last process is recently raises a unprecedented interest due to the rumours of its first observations. Predicted 100 years (precisely) gravitational waves have not been detected yet. Our results may play a significant role in understanding the observational results.

In polish public television appeared in the end of previous year a program about scientific computing considering also our project. Unfortunately it is available only in polish. You may found it here

Intel 64-bit Mac OS 10.5 or later support
I had compiled Mac OS version of ULX application.

Apologise if application will crash on your system (please report this to me) as I have no experience in Mac OS compiling.

If the application provide right results I will also add support for BHspin app.

Allora, i lavori gpu opencl procedono e, a fronte di notizie interessanti riguardo le prestazioni

some functions are 30 times faster on my GPU (ATI 7770) then on CPU (E3-1230v2)

ci sono da annoverare anche alcuni problemi

we still have some problems with calculations precision and new problem with necessary temp file which uses about 4GB of HDD space for each work unit.

Tutti a comprare dischi da 2 Tb, forza!!

boboviz ha scritto:

Allora, i lavori gpu opencl procedono e, a fronte di notizie interessanti riguardo le prestazioni

some functions are 30 times faster on my GPU (ATI 7770) then on CPU (E3-1230v2)

ci sono da annoverare anche alcuni problemi

we still have some problems with calculations precision and new problem with necessary temp file which uses about 4GB of HDD space for each work unit.

Tutti a comprare dischi da 2 Tb, forza!!

Oppure ramdisk eseguendo qualche manciata di wu.

Non avevo notato, nel loro sito, la cartina geografica che indica i volontari per nazione . Non sarebbe male averla anche per altri progetti (tipo Tn-Grid)

Sono entrato da pochi giorni pure in questo progetto e devo dire che rende niente male in fatto di crediti. Dovevo entrarci prima...

boboviz ha scritto:

Non avevo notato, nel loro sito, la cartina geografica che indica i volontari per nazione . Non sarebbe male averla anche per altri progetti (tipo Tn-Grid)

però qualcuno mi deve spiegare a che serve inserire nello stesso diagramma a torta tre valori con unità di misura diversa come user, hosts e gigaflops

@astroale: penso che sia una effettiva comparazione tra utenti attivi, host attivi e gflops impiegati.

Ah, comunque.. news!!!

Recognition for gravitational waves discovery
I have a pleasure to inform you, unless you already know, that prof. Krzysztof Belczyński,
the PI of Universe@Home project, in recognition for his involvement in discovery of gravitational
waves, was awarded recently with the Medal of Nicolas Copernicus and the Breakthrough Prize.

The former is the most prestigious prize awarded by the Polish Academy of Sciences. The latter is an
international award bestowed for remarkable scientific outcome. It was awarded for the whole group,
which took part in the research.

The results from the Universe@Home project are still used to increase our understanding of
gravitational waves and their role in the Universe.

A domanda diretta, risposta diretta:

We are under preparing new app proces and the changes need to be ported to GPU as well...
At the current stage test GPU app I expect in about a month

Luglio, col bene che ti voglio, vedrai non finirà ahi, ahi, ahi

Qualche novità riguardo il client gpu, che sembra procedere

I am still working over GPU app. I invented the engine that help to run massive number of tasks efficiently (including its irregularity) on GPU. First tests of this engine was passed correctly. Now, I am working over rewriting main code of the application to this engine. I can't write about speedup and majority of the possible potential bottlenecks. This engine will require a very little transfer between host and GPU and it will move the program control to GPU. I hope that engine allow to achieve good results under modern GPU (even if it will run quite complex code)

Speriamo !!!

Ci sono quindi buone premesse !!!

In attesa del client gpu, quelli del progetto hanno pubblicato su Nature!!!
arxiv.org/pdf/1602.04531v2

Qualcosa si muove lato gpu

I can write something about progress in development of GPU app. I wrote tentative version of CLRunEngine (engine that can help efficiently run complex code on GPU, including a sparse and non-regular control flow). Now, I writing an utility to facilitate porting complex from original app to CLRunEngine in enough natural way. Utility is not finished so far. Conversion from original code to code accepted by my engine would be a enough complex and cumbersome to maintain in handmade way. Therefore, I am trying to write an utility that help to move original to my engine.
Non-regular way of doing computations by original application and a sophisticated and sensitive code is enough big challenge for me. Very likely utility will be finished in next week or slightly later.