* Looking for MT/NLP opportunities * Hieu Hoang http://moses-smt.org/
On 16 March 2017 at 14:51, Ivan Zapreev <[email protected]> wrote: > Dear Hieu, > > Well, whether it is 20 or 40 cores is irrelevant as the trend on up to 20 > threads still contradicts the data provided on the Moses2 page. > > Regarding using the binary models, I shall check this out, as far as I > remember I use plain text models. I will look it up later today. But I see > that loading the models takes about 763 seconds, so it is 12-13 minutes ... > does not look binary to me. At least not all of them. > ok, if you are using plain text models, then this is the 1st difference to my results. I don't optimize the plain text models, I don't wanna wait for 13 minutes and this is not how most users use the decoder. But if you see bad scaling in Moses2 with binary models, please let me know If you do load the plain text files, you should check that it doesn't use up all memory and has to disk swap > > By the way, did you look at the data I provided? Moses2 and Moses do NOT > have the same speed using single thread. Moses2 is about 2 times faster and > by reaching 40 threads it is just 1.5 faster than Moses. This means that > Moses2 is by itself using faster algorithms or data structures, and its > better performance in my experiments comes from there and not from a better > multi-threading. > I know it's faster on 1 thread, it should be much faster on lots of threads. > You can consider also the Moses2 speed-up plots with respect to single > thread (provided in my experiments). Moses2 seems to scale worse than Moses > in the number of threads. I use the same models for Moses and Moses2 so any > possible memory hits, unless the data structures are different, should be > the same and thus not influencing the overall picture of scaling with the > increased number of threads. > By the way, should pre-loading of plain text models help? As far as I see > it should not make any difference ... > no, just pre-load the binary files > > Kind regards, > > Dr. Ivan S. Zapreev > > On Thu, Mar 16, 2017 at 3:11 PM, Hieu Hoang <[email protected]> wrote: > >> >> >> * Looking for MT/NLP opportunities * >> Hieu Hoang >> http://moses-smt.org/ >> >> >> On 16 March 2017 at 13:16, Ivan Zapreev <[email protected]> wrote: >> >>> Dear Hieu, >>> >>> Thank you for a prompt and detailed reply! >>> >>> >>> So your server has 20 cores (40 hyperthreads) and 16GB RAM? If >>> that's correct, then the RAM size would be a problem - you need as much RAM >>> as the total size of your models, plus more for working memory and the OS. >>> >>> The amount of memory is 256 Gb and not 16. There are a number of 16 Gb >>> plates installed. >>> To my knowledge the machine is not hyperthreaded but just has 40 cores, >>> although I am now getting a bit doubtful about that. >>> >> 256GB is good. 20/40 core/hyperthreads is not important for the moment, >> but you should find out exactly what it is >> >> >>> >>> >> Do you run Moses command line, or the server? My timings are based on >>> the command line, the server is a little slower. >>> >>> Both Moses and Moses2 are run in the console mode (not server). The >>> model loading time is excluded from the measurements. I could not manage to >>> get the asynchronous XML-RPC to work, so for my experiments that would be >>> as if I used Moses/Moses2 in a single-thread mode. Therefore I used the >>> command-line version. >>> >>> >>> Do you run Moses directly, or is another evaluation process running >>> it? Are you sure that evaluation process is working as it should? >>> >>> Moses is run from command time under the "time" command of Linux, and so >>> are other systems we used in comarison. We look at the runtime and not the >>> CPU times, but we perform a number of experiments to measure the average >>> times and control the standard deviations. >>> >>> >>> Do you minimise the effect of disk read by pre-loading the models >>> into filesystem cache? This is usually done by running this before running >>> the decoder cat [binary model files] > /dev/null >>> >>> Nop, we did not do pre-loading, for none of the tools but perhaps this >>> is not an issue as we just measure the average model loading times and >>> subtract them from the average run-time with decoding. So the model loading >>> times are excluded from the results. Our goal was to measure and compare >>> the decoding times and how they scale in the number of threads. >>> >> You're using the Probing phrase-table with integrated reordering model, >> and a binary KenLM, right? >> >> If so, the loading time will be minimal (1-2 secs) since the binary >> format just memory map the whole data but doesn't actually load them into >> memory. However, the overwhelming amount of the time taken for decoding >> will be page faults while doing LM and pt random lookups. >> >> It would be no surprise that decoding speed for Moses and Moses2 would be >> similar without pre-loading - they are looking up the same data. >> >> >>> >>> >>> it may take a while, but I can't replicate your results without it. >>> Alternatively, I can provide you with my models so you can try & replicate >>> my results. >>> >>> The experiments are run on an internal server which is not visible from >>> outside. I shall explore the possibilities of sharing the models, but I am >>> doubtful it is possible. The university network is very restricted. Yet, I >>> am definitely open to re-running your experiments. If possible. >>> >> i can make it available. But the results will be the same unless you sort >> out your pre-loading >> >>> >>> Kind regards, >>> >>> Ivan >>> >>> <http://www.tainichok.ru/> >>> >> >> > > > -- > Best regards, > > Ivan > <http://www.tainichok.ru/> >
_______________________________________________ Moses-support mailing list [email protected] http://mailman.mit.edu/mailman/listinfo/moses-support
