Alright, I'm soliciting opinions here (despite it being a biased source).
I'm looking to make a personal AIM transport, which only purpose is so I can use AIM over Google Talk.
What's the most lightweight way to do this? jabberd14, jabberd2, Openfire?
Also, any recommendations for transports? PyAIM-t requires Python & Twisted, while AIM-t is C/C++, so it *should* require less resources, right?
Lemme know your opinions. Thanks!
Better to try yourself
Alright, I'm soliciting opinions here (despite it being a biased source).
You can also ask in the JADMIN mailing list inJabber.org .
I'm looking to make a personal AIM transport, which only purpose is so I can use AIM over Google Talk.
What's the most lightweight way to do this? jabberd14, jabberd2, Openfire?
Don't forget djabberd, pretzel, xmppd.py, Tigase... and ejabberd :)
I guess the only way to know is to experiment. Some years ago I wrote this How-to:Benchmarking Jabber/XMPP Servers with Jabsimul . Note that the results are not usable nowadays, since they were obtained using very old software versions.
Also, any recommendations for transports? PyAIM-t requires Python & Twisted, while AIM-t is C/C++, so it *should* require less resources, right?
I'd say: it *may* or *may not* require less computational resources. It was also supposed that ejabberd (Erlang) shouldn't outperform jabberd14 (C) or jabberd2 (C). But it did in 2003.
After seeing those results, I consider that the only way to know which Jabber server consumes less CPU and RAM with a given load (be it real or synthetically generated) is to try it with that load.
Thanks for the reply. I
Thanks for the reply. I guess I'll have to try them all... :)