Messages in this thread |  | | | From | "Michael Herf" <> | | Subject | RE: Can Kernel Compression Code make Sense? | | Date | Sun, 22 Feb 1998 06:08:56 -0800 |
| |
> Hm... I'm able to write with a spedd of about 3 MB/sec to my HD > with only a few cpucycles lost (you know, It's the think called SCSI :)) > Compressing my cache hast to be at least that fast but will always > take much more cpucycles... that's why I think it will be always a lost > of performance...
A RAID-10 array might keep up, but hmm...
Any fast RLE-like compression algorithm will run at memory-read bandwidth on modern machines. (About 80 MB/sec.) Even LZO (a PD fast compression technique) claims a significant fraction of memcpy() -- like 1/2, with much better compression than RLE.
On another level, I wonder how important CPU usage is in this case. i.e. If the system is getting no work done because it's waiting for pages to be swapped, saving CPU cycles may not accomplish much, but that's probably a faulty line of reasoning.
mike
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu
|  |