Databases
StackOverflow2013
Postsdbo
POClojure number crunching performance
 

Note that there are some explanatory texts on larger screens.

plurals
  1. POClojure number crunching performance
    primarykey
    data
    text
    <p>I'm not sure whether this belongs on StackOverflow or in the Clojure Google group. But the group seems to be busy discussing <a href="http://groups.google.com/group/clojure/browse_thread/thread/c8c850595c91cc11/171cacba292a0583?q=primitive+support&amp;lnk=ol&amp;" rel="noreferrer">numeric improvements for Clojure 1.2</a>, so I'll try here:</p> <p><a href="http://shootout.alioth.debian.org/" rel="noreferrer">http://shootout.alioth.debian.org/</a> has a number of performance benchmarks for various languages. </p> <p>I noticed that Clojure was missing, so I made a Clojure version of the <a href="http://shootout.alioth.debian.org/u32q/benchmark.php?test=nbody&amp;lang=all" rel="noreferrer">n-body problem</a>.</p> <p>The fastest code I was able to produce can be <a href="http://github.com/j-g-faustus/Clojure-test-code/blob/master/shootout/nbody.clj" rel="noreferrer">found here</a>, and benchmarking it seems to be saying that for number crunching Clojure is</p> <ul> <li>factor ~10 quicker than Python/Ruby/Perl</li> <li>factor ~4 slower than C/Java/Scala/Ada</li> <li>approximately on par with OCaml, Erlang and Go</li> </ul> <p>I'm quite happy with that level of performance.</p> <p>My question to the Clojure gurus is</p> <ul> <li>Are there obvious improvements I have missed, either in terms of speed or in terms of code brevity or readability (without sacrificing speed)?</li> <li>Do you consider this to be representative of Clojure performance vs Python/Ruby/Perl on one hand and Java/C on the other?</li> </ul> <p><strong>Update</strong></p> <p>More Clojure 1.1 benchmark programs for the shootout <a href="http://github.com/jafingerhut/clojure-benchmarks" rel="noreferrer">here</a>, including the n-body problem.</p>
    singulars
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. PTQuestion
    1. USj-g-faustus
    1. USj-g-faustus
    plurals
    1. This table or related slice is empty.
    1. PL
      singulars
      1. LTLinked
    1. This table or related slice is empty.
    1. PO
      singulars
      1. PTAnswer
    2. PO
      singulars
      1. PTAnswer
    3. PO
      singulars
      1. PTAnswer
    1. VO
      singulars
      1. POClojure number crunching performance
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. POClojure number crunching performance
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. POClojure number crunching performance
      1. USJulien Lebosquain
      1. VTFavorite
    1. COI have to imagine the JVM is going to play a big part here. What JVM are you using? Are you using the same one as the shootout?
      singulars
      1. POClojure number crunching performance
      1. USAdam Gent
    2. COJava 1.6.0, the standard java that ships with OS X 10.6. It's approximately the same JVM (me: 1.6.0_20, shootout: 1.6.0_18) but different computer than the shootout. I ran both Clojure and the shootout Java implementation locally. I estimated relative performance by using Java as the baseline and scaling the shootout results accordingly.
      singulars
      1. POClojure number crunching performance
      1. USj-g-faustus
    3. COYour code looks pretty good. With the primitive improvements in 1.2 I would expect you to be able to get pretty close to the Java time. Have you tried running it through a profiler? My suspicion is that something somewhere is adding a boxing or function call overhead that is hurting you.
      singulars
      1. POClojure number crunching performance
      1. USmikera
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

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