[#11822] RCR: Input XML support in the base Ruby — Dave Thomas <Dave@...>

15 messages 2001/03/01

[#11960] Not Ruby, for me, for the moment at least — "Michael Kreuzer" <mkreuzer@... (nospam)>

I wrote on this newsgroup last weekend about how I was considering using

11 messages 2001/03/04

[#12023] French RUG ? — "Jerome" <jeromg@...>

Hi fellow rubyers,

16 messages 2001/03/05

[#12103] disassembling and reassembling a hash — raja@... (Raja S.)

Given a hash, h1, will the following always hold?

20 messages 2001/03/06

[#12204] FEATURE REQUEST: 'my' local variables — Leo Razoumov <see_signature@127.0.0.1>

Ruby is, indeed, a very well designed language.

64 messages 2001/03/07
[#12250] Re: FEATURE REQUEST: 'my' local variables — Leo Razoumov <see_signature@127.0.0.1> 2001/03/07

>>>>> "GK" == GOTO Kentaro <gotoken@math.sci.hokudai.ac.jp> writes:

[#12284] Re: FEATURE REQUEST: 'my' local variables — gotoken@... (GOTO Kentaro) 2001/03/08

In message "[ruby-talk:12250] Re: FEATURE REQUEST: 'my' local variables"

[#12289] Re: FEATURE REQUEST: 'my' local variables — matz@... (Yukihiro Matsumoto) 2001/03/08

Hi,

[#12452] Re: FEATURE REQUEST: 'my' local variables — gotoken@... (GOTO Kentaro) 2001/03/12

In message "[ruby-talk:12289] Re: FEATURE REQUEST: 'my' local variables"

[#12553] Re: FEATURE REQUEST: 'my' local variables — Dave Thomas <Dave@...> 2001/03/13

matz@zetabits.com (Yukihiro Matsumoto) writes:

[#12329] Math package — Mathieu Bouchard <matju@...>

18 messages 2001/03/09

[#12330] Haskell goodies, RCR and challenge — Robert Feldt <feldt@...>

Hi,

19 messages 2001/03/09
[#12374] Re: Haskell goodies, RCR and challenge — matz@... (Yukihiro Matsumoto) 2001/03/10

Hi,

[#12349] Can Ruby-GTK display Gif Png or Jpeg files? — Phlip <phlip_cpp@...>

Ruby-san:

20 messages 2001/03/09

[#12444] class variables — Max Ischenko <max@...>

14 messages 2001/03/12

[#12606] Order, chaos, and change requests :) — Dave Thomas <Dave@...>

17 messages 2001/03/14

[#12635] email address regexp — "David Fung" <dfung@...>

i would like to locate probable email addresses in a bunch of text files,

12 messages 2001/03/14

[#12646] police warns you -- Perl is dangerous!! — Leo Razoumov <see_signature@127.0.0.1>

I just read this story on Slashdot

14 messages 2001/03/14
[#12651] Re: police warns you -- Perl is dangerous!! — pete@... (Pete Kernan) 2001/03/14

On 14 Mar 2001 11:46:35 -0800, Leo Razoumov <see_signature@127.0.0.1> wrote:

[#12691] Re: police warns you -- Perl is dangerous!! — "W. Kent Starr" <elderburn@...> 2001/03/15

On Wednesday 14 March 2001 15:40, Pete Kernan wrote:

[#12709] [OFFTOPIC] Re: police warns you -- Perl is dangerous!! — Stephen White <spwhite@...> 2001/03/16

On Fri, 16 Mar 2001, W. Kent Starr wrote:

[#12655] Re: FEATURE REQUEST: 'my' local variables — "Benjamin J. Tilly" <ben_tilly@...>

>===== Original Message From Leo Razoumov <see_signature@127.0.0.1> =====

18 messages 2001/03/14

[#12706] Library packaging — "Nathaniel Talbott" <ntalbott@...>

I have a project that I'm working on that needs to live two different lives,

30 messages 2001/03/16

[#12840] Looking for a decent compression scheme — Dave Thomas <Dave@...>

14 messages 2001/03/19

[#12895] differences between range and array — "Doug Edmunds" <dae_alt3@...>

This code comes from the online code examples for

16 messages 2001/03/20
[#12896] Re: differences between range and array — "Hee-Sob Park" <phasis@...> 2001/03/20

[#12899] Re: differences between range and array — Jim Freeze <jim@...> 2001/03/20

On Tue, 20 Mar 2001, Hee-Sob Park wrote:

[#12960] TextBox ListBox — Ron Jeffries <ronjeffries@...>

Attached is a little Spike that Chet and I are doing. It is a

13 messages 2001/03/20

[#12991] [ANN] Lapidary 0.2.0 — "Nathaniel Talbott" <ntalbott@...>

Well, here's my first major contribution to the Ruby world: Lapidary. It's a

16 messages 2001/03/20

[#13028] mkmf question — Luigi Ballabio <luigi.ballabio@...>

15 messages 2001/03/21

[#13185] Reading a file backwards — "Daniel Berger" <djberg96@...>

Hi all,

21 messages 2001/03/25
[#13197] Re: Reading a file backwards — "Daniel Berger" <djberg96@...> 2001/03/25

> Hi Dan,

[#13203] Re: Reading a file backwards — Mathieu Bouchard <matju@...> 2001/03/25

On Sun, 25 Mar 2001, Daniel Berger wrote:

[#13210] Re: Reading a file backwards — "Daniel Berger" <djberg96@...> 2001/03/25

"Mathieu Bouchard" <matju@sympatico.ca> wrote in message

[#13374] Passing an array to `exec'? — Lloyd Zusman <ljz@...>

I'd like to do the following:

15 messages 2001/03/31

[#13397] Multidimensional arrays and hashes? — Lloyd Zusman <ljz@...>

Is it possible in ruby to make use of constructs that correspond to

14 messages 2001/03/31

[ruby-talk:12830] Re: Floating point performance & Garbage collection

From: ale@...
Date: 2001-03-18 23:27:17 UTC
List: ruby-talk #12830

On Mon, 19 Mar 2001, Jean-Sebastien ROY wrote:

> I recently came across a little performance problem I have difficulties 
> to explain.
...
> So I tried to disable the Garbage Collector (GC.disable) :
> 
> With GC.disable, tmax = 1000 and const = 1, the code take 1.52s to run.
> With GC.disable, tmax = 1000 and const = 1.0, the code take 1.89s to run.
> 
> Could someone explain to me why the floating point addition seems to 
> trigger the GC ?

You almost got it. The reason is just what you thought it to be. It's
about GCing.

First your code created a bunch of objects. Then it looped
and measured the time to sum linearly one million integers. It was
fast. It was fast because integers are kept in the object reference
itself. (And Ruby just pretends to have them in a real objects.)

Calculating with floats was slower for couple of reasons. First each
summation generated a new float object (new object instead of no object
for integer). But the majority of speed decrease came from GC. Once the
memory was filled with once used floats which were abandoned and Ruby ran
out of memory it GCed. Then it had to go through the pre-allocated objects
to see if they're referring to any other objects in order to release the
memory taken by object which are not used anymore. So each normal GC
besides out-of-memory-GC generated a pass through all the objects in
memory and a major slowdown.

The next version of Ruby incorporates a generational GC which should be
much faster regarding cases like this. I suppose it works by moving
"old" objects (like the ones you pre-allocated) after dozen of GC marking
passes to a separate "old object" pool which is checked not so often.

I expect the performance in cases of lots of object creating will be
boosted quite much.

	- Aleksi

In This Thread