September 2009 - Darwin-dev

Re: fread multiple files
by Terry Lambert 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com -- Terry David, Terry, Jens - thanks for your answers Here are the attached shark and sampler outpus: http://people.civil.gla.ac.uk/~koziara/solfec.shark.txt.gz http://people.civil.gla.ac.uk/~koziara/solfec.trace.tar.gz For each time t For each output file f Seek (f, t) Read (f) Tomek On 25 Sep 2009, at 20:20, Jens Alfke wrote: _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… You could definitely be doing your three I/Os concurrently instead of serailly, assuming the index data is not needed to drive the I/O, at which point you could do your index I/O concurrently and then do your data file I/O after that (semiserially, in other words). But if you arranged your code to do the first index read up front before going into the loop, you could do your *next* index reads concurrently with your data I/O for your current indices. You could further increase concurrency , assuming the data is index linked instead of data linked (i.e. you don't need each set of data prior to the next set of data) by inlementing a producer/consumer model, and issuing your I/Os as fast as you can and queueing the buffered data records for processing (you'd want to rate limit the number of outstanding I/Os to limit the data you have in hand at one time, before you ask for more data). If your data comes in at a rate higher than 1.5 times faster than you can process it, if you have a sufficiently large pool of work item queue elements, you could establish high and low water marks on number of elements in the queue, and then only go back for more data when you hit the low water mark. On Sep 25, 2009, at 1:40 PM, Tomasz Koziara wrote: Indeed - I do not use MPI IO. I use an XDR based binary output instead. Would be desirable if the post-processor would not require MPI, hence I stick to XDR for now. Of course it could change, once I am not able to overcome the efficiency issues. But then I would probably go for HDF5. The post-processing process (single-thread) simply opens all 3 * N files and keeps them open. Of those 3 * N files, only N store some substantial data (are large), while the rest stores indexing and labeling data. So basically what happens is: Could you post the actual output of running 'sample' on your single- threaded tool? I think you may be misinterpreting the sampling. —Jens This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by Jens Alfke 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On Sep 25, 2009, at 1:29 PM, Tomasz Koziara wrote: For each time t For each output file f Seek (f, t) Read (f) —Jens _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… The post-processing process (single-thread) simply opens all 3 * N files and keeps them open. Of those 3 * N files, only N store some substantial data (are large), while the rest stores indexing and labeling data. So basically what happens is: Sounds like you're reading a lot of data, so I would expect that your process spends most of its time blocked in disk I/O, which is what the profiles show. It sounds like your access patterns are pretty random too, which makes performance worse. I don't know anything at all about MPI or XDR, so I don't think I can give any advice here. This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by Tomasz Koziara 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com David, Terry, Jens - thanks for your answers Here are the attached shark and sampler outpus: http://people.civil.gla.ac.uk/~koziara/solfec.shark.txt.gz http://people.civil.gla.ac.uk/~koziara/solfec.trace.tar.gz For each time t For each output file f Seek (f, t) Read (f) Tomek On 25 Sep 2009, at 20:20, Jens Alfke wrote: _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Indeed - I do not use MPI IO. I use an XDR based binary output instead. Would be desirable if the post-processor would not require MPI, hence I stick to XDR for now. Of course it could change, once I am not able to overcome the efficiency issues. But then I would probably go for HDF5. The post-processing process (single-thread) simply opens all 3 * N files and keeps them open. Of those 3 * N files, only N store some substantial data (are large), while the rest stores indexing and labeling data. So basically what happens is: Could you post the actual output of running 'sample' on your single- threaded tool? I think you may be misinterpreting the sampling. —Jens This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by Alastair Houghton 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha1; c=relaxed; d=alastairs-place.net; h= subject:mime-version:content-type:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; q=dns/txt; s= aug07; bh=wQ9t9dnaucrW8EkvPs6KqX7tTsY=; b=e7zeE00XKfDVacZXv8ecFi 2cymgUr9inZ5LmFuI/Cj6ui16X0s+UoLA+D/Y3/SPVBQwL5h3AewMO8cCFikJ9XI 6YajiFb26zIkS6/NNQ047npBlP499iI34exWm3xwgo9h12RZD7vRKrQmMnSlSnNh 5AT1VsuaZSmrc5g9JVWD8= On 25 Sep 2009, at 20:07, Terry Lambert wrote: press? I don't recognise that in this context, but I'm curious? Kind regards, Alastair. -- http://alastairs-place.net _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… If the same file is being contended in multiple threads (the mutex usage will normally not kick in unless your program is multithreaded in the first place), then you need to stop using stdio and switch to using aio_read or press instead, and managing any block buffering, like that normally provided by stdio, in your own code. And in which version of OS X did support for aio_read() debut? I don't recall noticing support for the REALTIME stuff before (maybe I just missed it somehow). This email sent to site_archiver(a)lists.apple.com

1 0

Re: conversion printf format to icu style format
by Alastair Houghton 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha1; c=relaxed; d=alastairs-place.net; h= subject:mime-version:content-type:from:in-reply-to:date:cc: content-transfer-encoding:message-id:references:to; q=dns/txt; s= aug07; bh=0tfesNYtGNjSlCBKZWkYXuf+2UI=; b=nevxOvZ66nMQ4XGo6IZNVw S6C/MmmLe7aeMeszv5YFbOsySaNma3Ng+64rA3MBTcsXlDWvulFtK8Xp6zhhCNAX OFP0CzWA6x0hyXbp4kLKZ++XPYvcEPwGeGRfcjQP+jxIO9TqgaK0/265aguHFyUd WI/7tvfC1TZ0R7VmNdxrw= On 25 Sep 2009, at 19:54, Terry Lambert wrote: Apple? :-D Kind regards, Alastair. -- http://alastairs-place.net _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I don't know if anyone has done this, as I don't know of anyone who actually uses ICU in production software. It seems to me, though, that that question might be better asked at the ICU site than on this list, since there's no vested interest here in tools to start using ICU, whereas the ICU people are the ones most likely to provide such a tool as part of advocating their library. Well, whoever it is that works on CoreFoundation is probably interested in this kind of thing also. And the Core Text/Cocoa Text people probably are too... Not to mention some of the other framework guys (NSPredicate uses ICU as well). This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by Jens Alfke 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com —Jens _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… Could you post the actual output of running 'sample' on your single- threaded tool? I think you may be misinterpreting the sampling. This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by Terry Lambert 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Hi __spin_lock pthread_mutex_lock fread pthread_mutex_unlock flockfile funlockfile -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Sep 25, 2009, at 10:27 AM, Tomasz Koziara <t.koziara(a)civil.gla.ac.uk> wrote: An MPI code I use outputs 3 * N files, where N is the number of processors. Then I am using a serial code to post-process the results. The post-processor is single-threaded and for every time instant reads a chunk of data from each of 3 *N files. I noticed that most of the time (nearly 50%) is spent in: I am wondering whether those of you who know more about optimizing IO could advice me on how to make this reading more efficient. What you are seeing with the fread is serialization of access to the FILE/fd pair being read in order to protect the FILE's stdio buffer contents and the fd's file offset from being simultaneously accessed and therefore subject to races. This is normal, and unless you are accessing the same file from multiple threads simultaneously, the lock being held is not contended, and you are misinterpreting your statistics by not subtracting the fread time from the time the lock is being held to get the actual lock overhead. If the same file is being contended in multiple threads (the mutex usage will normally not kick in unless your program is multithreaded in the first place), then you need to stop using stdio and switch to using aio_read or press instead, and managing any block buffering, like that normally provided by stdio, in your own code. PS: It seems to me that if there isn't simultaneous access to the same file happening, the main serialization overhead your application has is in your post-processing code. This email sent to site_archiver(a)lists.apple.com

1 0

Re: conversion printf format to icu style format
by Terry Lambert 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Ah. Misunderstood because of where he asked, then. -- Terry _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… On Sep 25, 2009, at 11:06 AM, Fritz Anderson <fritza(a)manoverboard.org> wrote: On 25 Sep 2009, at 12:24 PM, Terry Lambert wrote: I believe the ICU libraries can do ICU formatting; what does this have to do with printf, whose formatting is dictated by the ISO/ ANSI C standards and the Single UNIX Specification v3/POSIX 1003.1-2003 standards, and can not be changed without starting by submitting an RFP to The Austin Group? I would suggest you just use the ICU libraries if you want ICU formatting. I imagined the OP's situation was that he had thousands of printf- format strings, and wanted to know if he could automate the task of porting them to ICU. I think it would be pretty trivial to automate this by starting with a printf implementation, and instead of processing arguments while traversing the format string, emitting ICU formatting instead. I don't know if anyone has done this, as I don't know of anyone who actually uses ICU in production software. It seems to me, though, that that question might be better asked at the ICU site than on this list, since there's no vested interest here in tools to start using ICU, whereas the ICU people are the ones most likely to provide such a tool as part of advocating their library. This email sent to site_archiver(a)lists.apple.com

1 0

Re: fread multiple files
by David Leimbach 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com Dkim-signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:mime-version:received:in-reply-to:references :date:message-id:subject:from:to:cc:content-type; bh=mToVbDs+weOoaHDr/V+nCKDMX30azVsiVtlCXa1EOYs=; b=sWNm0nuImVN/poOFzThFpqYk5DmTJUVFAGRUAFXbl8Y0HzCYdl7iaC6KB/H3q/XuKU LCzSZLdF72lG2eLUMZwjL2IloUIEar3OgQfjzaI9YtKvsG6nmE6gtc+PYZFIrtNZ4c3E HTgdTzmuqEeHv98ATaH9cdKFCfLBhixQW8R5s= Domainkey-signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; b=nlQDlSshEj3vYDzO5Jfe6GnEQ4QgzO2ExdrBSLtRIwL/634ytlXKZMpWvSV4IneI0/ aJbXQdU4DAdsvekuGA6AhX5V+l9EV3tGExnV/8PqsU/6SoJFJeHDVOIWVSFn4zzfaWj+ /ZkxQVwLe0SR+PWVTuA8sWn5iE5mvs27fpznE= _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… This email sent to site_archiver(a)lists.apple.com

1 0

Re: conversion printf format to icu style format
by Fritz Anderson 25 Sep '09

25 Sep '09

site_archiver(a)lists.apple.com Delivered-To: darwin-dev(a)lists.apple.com On 25 Sep 2009, at 12:24 PM, Terry Lambert wrote: — F _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.app… I believe the ICU libraries can do ICU formatting; what does this have to do with printf, whose formatting is dictated by the ISO/ANSI C standards and the Single UNIX Specification v3/POSIX 1003.1-2003 standards, and can not be changed without starting by submitting an RFP to The Austin Group? I would suggest you just use the ICU libraries if you want ICU formatting. I imagined the OP's situation was that he had thousands of printf- format strings, and wanted to know if he could automate the task of porting them to ICU. -- Fritz Anderson -- Xcode 3 Unleashed: Now in its second printing -- <http://x3u.manoverboard.org/ > This email sent to site_archiver(a)lists.apple.com

1 0