<div dir="ltr">hash-update is defined in Racket and, besides the error checking, is basically the composition of those two functions. I expect it was written as something to abstract over a common pattern, not speed things up.<div> <br></div><div>Is this a bottleneck in some real program? Perhaps you'd share that (or some inner loop of it)? I wonder if there is a better way to speed it up than this.<br><div><br></div><div>Robby</div></div></div> <div class="gmail_extra"><br><br><div class="gmail_quote">On Thu, Feb 14, 2013 at 1:09 PM, J. Ian Johnson <span dir="ltr"><<a href="mailto:ianj@ccs.neu.edu" target="_blank">ianj@ccs.neu.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"> If I am right, then it would make sense to use hash-update when the hash is big enough that a log-time traversal takes longer than a general procedure call's set-up and tear-down.<br> <div class="im HOEnZb">-Ian<br> ----- Original Message -----<br> From: James Bergstra <<a href="mailto:james.bergstra@gmail.com">james.bergstra@gmail.com</a>><br> </div><div class="HOEnZb"><div class="h5">To: J. Ian Johnson <<a href="mailto:ianj@ccs.neu.edu">ianj@ccs.neu.edu</a>><br> Cc: <a href="mailto:users@racket-lang.org">users@racket-lang.org</a><br> Sent: Thu, 14 Feb 2013 13:47:51 -0500 (EST)<br> Subject: Re: [racket] Question about hash table efficiency<br> <br> Thanks, I suppose I just had my hopes up. I thought the more compact<br> form might involve fewer expression evaluations and require just one<br> hash lookup instead of two, and therefore might run something like<br> twice as fast, rather than slightly slower.<br> <br> On Thu, Feb 14, 2013 at 1:36 PM, J. Ian Johnson <<a href="mailto:ianj@ccs.neu.edu">ianj@ccs.neu.edu</a>> wrote:<br> > My guess would be the higher-orderness on hash-update causes the (minor) slowdown.<br> > -Ian<br> > ----- Original Message -----<br> > From: James Bergstra <<a href="mailto:james.bergstra@gmail.com">james.bergstra@gmail.com</a>><br> > To: <a href="mailto:users@racket-lang.org">users@racket-lang.org</a><br> > Sent: Thu, 14 Feb 2013 13:16:47 -0500 (EST)<br> > Subject: [racket] Question about hash table efficiency<br> ><br> > Hi list, just discovered racket and having fun learning about scheme<br> > again. Haven't used it since undergrad. Thanks!<br> ><br> > I was wondering about the efficiency of hash-update vs. hash-set and<br> > hash-ref. I thought the former would be faster, otherwise why have it?<br> > A little benchmark script reveals the same surprising result as my<br> > actual application though:<br> ><br> > """<br> > #lang racket<br> ><br> > ;; warm-up<br> > (time (void<br> > (for/fold ([table (make-immutable-hasheq '())])<br> > ([ii (in-range 100000)])<br> > (hash-update table (modulo ii 100) add1 ii))))<br> ><br> > ;; update<br> > (newline)<br> > (time (void<br> > (for/fold ([table (make-immutable-hasheq '())])<br> > ([ii (in-range 100000)])<br> > (hash-update table (modulo ii 100) add1 ii))))<br> ><br> > ;; set/ref<br> > (newline)<br> > (time (void<br> > (for/fold ([table (make-immutable-hasheq '())])<br> > ([ii (in-range 100000)])<br> > (hash-set table (modulo ii 100) (add1 (hash-ref table (modulo ii 100) ii))))))<br> ><br> > """<br> ><br> > This prints out:<br> ><br> > cpu time: 112 real time: 114 gc time: 32<br> ><br> > cpu time: 80 real time: 82 gc time: 0<br> ><br> > cpu time: 76 real time: 75 gc time: 4<br> ><br> ><br> > My original application used mutable hash tables and the same effect<br> > was observed (if not more dramatically). Any thoughts on why the<br> > update form is slower? I'm running this in Racket v5.1.1.<br> ><br> > - James<br> > ____________________<br> > Racket Users list:<br> > <a href="http://lists.racket-lang.org/users" target="_blank">http://lists.racket-lang.org/users</a><br> ><br> <br> ____________________<br> Racket Users list:<br> <a href="http://lists.racket-lang.org/users" target="_blank">http://lists.racket-lang.org/users</a><br> </div></div></blockquote></div><br></div>