On Wed, 2003-06-18 at 16:56, Sascha Schumann wrote: > > That's good, but that's only one step. Some system mallocs are very > > inefficient on small blocks so it might be very worthwhile to grab a > > "chunk" of, say, 64k, instead of many small mallocs of 20 bytes. > How do you know that this is a significant problem?
Because I have written my own operating system, including allocator, and I know that it can be hard to do it right. Also, I have read that at least previous versions of SunOS and BSD have this problem (over-consumption of memory with many small mallocs). I would think that this also applies to Linux a little bit. Prove me wrong, but I am fairly certain that doing 100 000 allocations of 20 bytes is a whole lot 1) slower and 2) more memory consuming than allocating 2 megs at once. This is what all research I have read comes down to: the bigger the blocks that you allocate from the system, the better your performance. You are very welcome to show me research that shows the opposite, but I doubt that anyone would even consider researching this since it is a fairly accepted computer science thesis really. -- Best regards, Per Lundberg / Capio ApS Phone: +46-18-4186040 Fax: +46-18-4186049 Web: http://www.nobolt.com
signature.asc
Description: This is a digitally signed message part
