| 1 |
Volker Armin Hemmann <volker.armin.hemmann@××××××××××××.de> posted |
| 2 |
200802131346.26316.volker.armin.hemmann@××××××××××××.de, excerpted below, |
| 3 |
on Wed, 13 Feb 2008 13:46:26 +0100: |
| 4 |
|
| 5 |
> On Mittwoch, 13. Februar 2008, Duncan wrote: |
| 6 |
>>>removed lots of irrelevant 'my hardware is so cool' stuff'. |
| 7 |
> |
| 8 |
> You forget some (little) things. Not everything can be swapped out. Swap |
| 9 |
> is extremly slow AND it is much worse to swapout/swapin programm code |
| 10 |
> that should be run, instead of fetching some files from disk while the |
| 11 |
> programm runs. |
| 12 |
|
| 13 |
It's not always much worse, because as I explained, in my case, swap is 4- |
| 14 |
way striped while most of the main system is only two-way striped. Thus, |
| 15 |
that "irrelevant" stuff is relevant after all, because it alters the |
| 16 |
conditions of the case in debate, because swap reads in at ~2x the speed |
| 17 |
of most data read off disk including apps (which is itself ~2x what a |
| 18 |
single-disk system might reasonably expect). |
| 19 |
|
| 20 |
I've a feeling not appreciating this, not appreciating that your "test" |
| 21 |
case example of compiling with 2 gigs RAM vs only 1 has little to do with |
| 22 |
what might occur with PORTAGE_TMPDIR on tmpfs vs on disk, and not |
| 23 |
appreciating the point RF and I are both trying to make, is due to the |
| 24 |
same logic flaw. |
| 25 |
|
| 26 |
-- |
| 27 |
Duncan - List replies preferred. No HTML msgs. |
| 28 |
"Every nonfree program has a lord, a master -- |
| 29 |
and if you use the program, he is your master." Richard Stallman |
| 30 |
|
| 31 |
-- |
| 32 |
gentoo-amd64@l.g.o mailing list |
Archives