[antlr-interest] Re: Java memory mapped IO is slow for big files :(

lgcraymer lgc at mail1.jpl.nasa.gov
Wed Nov 17 16:31:55 PST 2004



--- In antlr-interest at yahoogroups.com, "John D. Mitchell"
<johnm-antlr at n...> wrote:
> >>>>> "Terence" == Terence Parr <parrt at c...> writes:
> [...]
> 
> > In typical fashion, your expectations are not always met with java
> > libraries.  I'm using jdk 1.4.2 on my os x box.
> 
> Well, that's part of the problem.  Apple hasn't been the best at making
> their port of the JDK great.  Also, are you completely up to date
with the
> Java version?  Early versions had lots of bugs.  I.e., I'd be
careful about
> extrapolating generally from those numbers without actually
verifying the
> numbers on at least MS Windows and Linux.
> 
> Also, are you testing this on your laptop?  The disks on the Apple
laptops
> suck ass in terms of speed.

John--

If you take a closer look at Ter's data, it looks like the conclusion
is that
a = c[n] // c is a character array
is noticeably slower than
a = X.read().

That might be due to bounds checking, array indexing, and/or something
else; it's not a disk issue because the three tests are more-or-less
equivalent in terms of disk usage.

--Loring

> Take care,
> 	John





 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/antlr-interest/

<*> To unsubscribe from this group, send an email to:
    antlr-interest-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 





More information about the antlr-interest mailing list