1 |
On Sun, Jun 16, 2019 at 6:02 PM Wols Lists <antlists@××××××××××××.uk> wrote: |
2 |
|
3 |
> And those people who wrote your guidelines? Are they the same clueless |
4 |
> people who believe the twice ram rule is pure fiction? (As I said, it is |
5 |
> *historical* *fact*). And why should I believe people who tell me the |
6 |
> rule no longer applies, if they can't tell me WHY it no longer applies? |
7 |
> I'd love to be enlightened - why can't anybody do that? |
8 |
> |
9 |
> |
10 |
What we call UNIX today is an API, not an implementation, and different |
11 |
UNIX implementations have completely different internals. You'd be |
12 |
hard-pressed to find any original UNIX code in any modern UNIX. Linux was |
13 |
written from scratch to implement the UNIX API, but other than some header |
14 |
files (the subject of the SCO lawsuits) there is no UNIX code in Linux. |
15 |
|
16 |
Modern UNIXes don't map the swap pages 1:1 to RAM pages like the original |
17 |
UNIX code did. Instead they only map the pages that actually need to be |
18 |
swapped. And some pages, such as executable code, don't get swapped at all |
19 |
- instead the existing on-disk executable file or shared library file is |
20 |
used as the "swap". Code and static data pages are also shared between |
21 |
processes, saving even more RAM. |
22 |
|
23 |
For example, per IBM for AIX ( |
24 |
https://developer.ibm.com/articles/au-aix7memoryoptimize3/): |
25 |
|
26 |
"A more sensible rule is to configure the paging space to be half the size |
27 |
of RAM plus 4GB with an upper limit of 32GB. In systems with more than 32GB |
28 |
of RAM, or on systems where you are using LPAR and WPAR to help split your |
29 |
workload, you can be significantly more selective and specific about your |
30 |
memory requirements." |
31 |
|
32 |
-- |
33 |
Manuel A. McLure WW1FA <manuel@××××××.org> <http://www.mclure.org> |
34 |
...for in Ulthar, according to an ancient and significant law, |
35 |
no man may kill a cat. -- H.P. Lovecraft |