On Thu, 9 Mar 2017 09:52:27 +0100
Andr? Warnier <aw@ice-sa.com> wrote:
> If I understand the general idea correctly, it consists of no longer running complicated
> and memory-hungry things directly in Apache through mod_perl, but to use Apache as a
> front-end reverse proxy, and proxy such calls to one or more back-end processes having
> their own persistent perl (or other) interpreter. Is that correct ?
>
What is your use case? If this is for a high traffic site serving lots of static pages, or your databases are huge and require long running queries, it may be worth it.
My situation is the opposite, no static content, low traffic, all of it hitting small databases with 5 or six queries on average per page (for CRM type apps).
This is what top looks like on a dedibox serving about 20 users :
top - 16:25:21 up 11 days, 44 min, 1 user, load average: 0,00, 0,00, 0,00
Tasks: 128 total, 1 running, 127 sleeping, 0 stopped, 0 zombie
%Cpu(s): 0,0 us, 0,1 sy, 0,0 ni, 99,9 id, 0,0 wa, 0,0 hi, 0,0 si, 0,0 st
KiB Mem: 4041248 total, 1112512 used, 2928736 free, 134504 buffers
KiB Swap: 4094972 total, 0 used, 4094972 free. 661944 cached Mem
Tests with ab show 50 requests/second, and I don't see a memory problem that would warrant the overhead of a front-end reverse proxy.
--
Bien ? vous, Vincent Veyron
https://marica.fr/ Gestion des sinistres assurances, des dossiers contentieux et des contrats pour le service juridique