1 |
maillog: 01/12/2004-16:30:40(-0700): Rob Baxter types |
2 |
> i think the point is that the data that gets cached can and does change |
3 |
> every 30 minutes, so caching doesn't really help. |
4 |
|
5 |
It does help. If data changes, it is first updated in the cache. New |
6 |
data is served directly out of the cache, while the hard disk is synced |
7 |
in the background. That's unless I am majorly mistaken about how caching |
8 |
works. |
9 |
|
10 |
> all 5 servers in the rsync.gentoo.org rotation are currently running |
11 |
> rsync out of ram. i think i can safely say they wouldn't be nearly as |
12 |
> fast as they are, running off a hard disk. |
13 |
|
14 |
Do you have some real numbers? I am not trying to doubt you too much. I |
15 |
am genuinely curious how big the speedup is and if it is worth the |
16 |
effort. All I wanted to point in my post is that the speedup is probably |
17 |
not that great, but I'd really like to see benchmark numbers if someone |
18 |
went to the trouble of doing it. |
19 |
|
20 |
> besides, ram is cheap |
21 |
|
22 |
That's pretty relative. $70 for 512MB is not cheap for me. One reason |
23 |
why my poor "server" is still running with 2x512MB *PC133*. |
24 |
|
25 |
> and has a small footprint (smaller blocks) using |
26 |
> ramdrive, why not use it. |
27 |
|
28 |
Well, you may be right. I guess I could try serving two trees and do |
29 |
some tests myself. |
30 |
|
31 |
-- |
32 |
*) Georgi Georgiev *) What fools these morals be! *) |
33 |
(* chutz@×××.net (* (* |
34 |
*) +81(90)6266-1163 *) *) |
35 |
|
36 |
-- |
37 |
gentoo-mirrors@g.o mailing list |