- Our objective is performance, by reducing the amount of read/write to
the swap;
- Compress by pages and write the compressed pages in 4k pages, possibly
WASTING space
in each page.. (it's ok);
- But you can't guarantee that we can compress all cases. Make a map of
the pages
compressed/not-compressed. A 200MB swap will use 6400 bytes of memory
with compressed/
not compressed bits;
- We are not doing ALL, but's is simple.
- We can minimize the time wasted on compression by not compress code
pages, only
data pages. What's the (normal expected) relation of code/data pages?
>Can't optimize this by compressing on a per page basis? I'm not overly
>familiar with the theory, but it seems that there is less to compress
>(yeah, the compression isn't as good) and less for the CPU to worry about.
>Hmmm.... If only there was intellegent hardware that could do this for
>us...
[]'s, Sergio Bruder.
----------------------------------------------+
bruder@centroin.com.br |
http://www.geocities.com/SiliconValley/3123 |
------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu