On Aug 23, 1:58 pm, William Stein <wst...@gmail.com> wrote:
> To whet our appetite though, might you do some benchmarks that compare
> the speed of adding 2+2 via the library and via pexpect?
OK, here's a little experiment. I thought adding 2 and 2 would be too
hard for now, so I stuck with 1+2. Good news is that all routines
arrived at the correct answer. I did count the conversions back and
forth every time as well, so times are not so very low. The
conversions are completely string based, so I think performance can be
quite a bit better still. Anyway, results:
ecl/library (+ 2 2): 27.6 µs per loop
maxima/library(+ 2 2): 134 µs per loop
maxima/pexpect(+ 2 2): 4.97 ms per loop
python/strings(2+2): 14 µs per loop
So yes, library access does work a little better than pexpect.
code used:
sage: from sage.libs.ecl import *
sage: init_ecl()
sage: ecl_eval("(require 'asdf)")
;;; Loading #P"/home/wstein/build/sage-4.1.1/local/lib/ecl-9.4.1/
asdf.fas"
;;; Loading #P"/home/wstein/build/sage-4.1.1/local/lib/ecl-9.4.1/
cmp.fas"
;;; Loading #P"/home/wstein/build/sage-4.1.1/local/lib/ecl-9.4.1/
sysfun.lsp"
<ECL: ("ASDF" "CMP") >
sage: ecl_eval('(load "%s")'%(SAGE_ROOT+"/local/lib/maxima/
maxima.fasb"))
;;; Loading "/scratch/nbruin/sage-4.1.1-ecl-x86_64-Linux/local/lib/
maxima/maxima.fasb"
<ECL: #P"/scratch/nbruin/sage-4.1.1-ecl-x86_64-Linux/local/lib/maxima/
maxima.fasb" >
sage: ecl_eval('(in-package :maxima)')
<ECL: #<"MAXIMA" package> >
sage:
sage: def ecladd(a,b):
....: return eval(str(ecl_eval('(+ %s %s)'%(a,b))));
....:
sage: def libadd(a,b):
....: return eval(str(ecl_eval('(MEVAL #$ %s + %s$)'%(a,b))))
....:
sage: def maxadd(a,b):
....: return (maxima(a)+maxima(b)).sage()
....:
sage: def pythadd(a,b):
....: return eval("%s + %s"%(a,b))
....:
sage: timeit("ecladd(1,2)")
625 loops, best of 3: 27.6 µs per loop
sage: timeit("libadd(1,2)")
625 loops, best of 3: 134 µs per loop
sage: timeit("maxadd(1,2)")
5 loops, best of 3: 4.97 ms per loop
sage: timeit("pythadd(1,2)")
625 loops, best of 3: 14 µs per loop
--~--~---------~--~----~------------~-------~--~----~
To post to this group, send an email to sage-devel@googlegroups.com
To unsubscribe from this group, send an email to
sage-devel-unsubscr...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/sage-devel
URLs: http://www.sagemath.org
-~----------~----~----~----~------~----~------~--~---
