StackOverflow2013

Note that there are some explanatory texts on larger screens.

plurals

POIs MATLAB OOP slow or am I doing something wrong?
text
Body
copied!<p>I'm experimenting with <a href="http://en.wikipedia.org/wiki/MATLAB" rel="noreferrer">MATLAB</a> <a href="http://en.wikipedia.org/wiki/Object-oriented_programming" rel="noreferrer">OOP</a>, as a start I mimicked my C++'s Logger classes and I'm putting all my string helper functions in a String class, thinking it would be great to be able to do things like <code>a + b</code>, <code>a == b</code>, <code>a.find( b )</code> instead of <code>strcat( a b )</code>, <code>strcmp( a, b )</code>, retrieve first element of <code>strfind( a, b )</code>, etc.</p> <p><strong>The problem: slowdown</strong></p> <p>I put the above things to use and immediately noticed a <em>drastic</em> slowdown. Am I doing it wrong (which is certainly possible as I have rather limited MATLAB experience), or does MATLAB's OOP just introduce a lot of overhead?</p> <p><strong>My test case</strong></p> <p>Here's the simple test I did for string, basically just appending a string and removing the appended part again:</p> <pre><code>classdef String < handle .... properties stringobj = ''; end function o = plus( o, b ) o.stringobj = [ o.stringobj b ]; end function n = Length( o ) n = length( o.stringobj ); end function o = SetLength( o, n ) o.stringobj = o.stringobj( 1 : n ); end end function atest( a, b ) %plain functions n = length( a ); a = [ a b ]; a = a( 1 : n ); function btest( a, b ) %OOP n = a.Length(); a = a + b; a.SetLength( n ); function RunProfilerLoop( nLoop, fun, varargin ) profile on; for i = 1 : nLoop fun( varargin{ : } ); end profile off; profile report; a = 'test'; aString = String( 'test' ); RunProfilerLoop( 1000, @(x,y)atest(x,y), a, 'appendme' ); RunProfilerLoop( 1000, @(x,y)btest(x,y), aString, 'appendme' ); </code></pre> <p><strong>The results</strong></p> <p>Total time in seconds, for 1000 iterations:</p> <blockquote> <p>btest 0.550 (with String.SetLength 0.138, String.plus 0.065, String.Length 0.057)</p> <p>atest 0.015</p> </blockquote> <p>Results for the logger system are likewise: 0.1 seconds for 1000 calls to <code>frpintf( 1, 'test\n' )</code>, 7 (!) seconds for 1000 calls to my system when using the String class internally (OK, it has a lot more logic in it, but to compare with C++: the overhead of my system that uses <code>std::string( "blah" )</code> and <code>std::cout</code> at the output side vs plain <code>std::cout << "blah"</code> is on the order of 1 millisecond.)</p> <p><strong>Is it just overhead when looking up class/package functions?</strong></p> <p>Since MATLAB is interpreted, it has to look up the definition of a function/object at run time. So I was wondering that maybe much more overhead is involved in looking up class or package function vs functions that are in the path. I tried to test this, and it just gets stranger. To rule out the influence of classes/objects, I compared calling a function in the path vs a function in a package:</p> <pre><code>function n = atest( x, y ) n = ctest( x, y ); % ctest is in matlab path function n = btest( x, y ) n = util.ctest( x, y ); % ctest is in +util directory, parent directory is in path </code></pre> <p>Results, gathered same way as above:</p> <blockquote> <p>atest 0.004 sec, 0.001 sec in ctest</p> <p>btest 0.060 sec, 0.014 sec in util.ctest</p> </blockquote> <p>So, is all this overhead just coming from MATLAB spending time looking up definitions for it's OOP implementation, whereas this overhead is not there for functions that are directly in the path?</p>

Querying!

Guidance

An individual column

Larger individual text columns get their own page to allow for proper reading.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload