Hey Raphaël,
Hey, I'm not too sure that 2 full time Google engineers working on this can be labeled as "artisanal engineering" :)). I'm new to Python (I'm a mostly Java guy) and the first think I wanted to know when I got interrested in OpenERP (after reading your paper, BTW)was: How good is Python? Apparently it's pretty good given that big names have bet their (server) farms on it. But Python is notoriously slow (i.e. compared to Java) and there is an infamous global lock issue that seems to get in the way of scalability (I don't know actually, I never verified, but that seems to be a recurring subject in the Python community). So when somebody is willing to fix those issues and that somebody's name is google, it can only be a good thing. BTW, how is OpenERP-on-Jython-Round-2 doing? sraps, You raise very good points. It's true that most of the time, the DB is the bottleneck. But having Postgrsql has a backend you know that if you want to fine tune it for performance and 'distribute' it for scalability, you can. Is it the case with python ? I don't know. In a sense, in terms of scalability, it doesn't matter because, as you pointed out, you can balance multiple share nothing OpenERP servers and scale. > One of the main bottlenecks are that OpenERP is single threaded (at least I > have not got any proof of other - I am speaking for 4.2) so when one user > fires up some time consuming request, other have to wait. You are thus confirming that an OpenERP server can accept only 1 '''concurrent''' user at a time. Can you elaborate further on that particular point? Raphaël, does this also happen with Jython? A+ Rgds, Dom. -------------------- m2f -------------------- -- http://www.openobject.com/forum/viewtopic.php?p=34353#34353 -------------------- m2f --------------------
_______________________________________________ Tinyerp-users mailing list http://tiny.be/mailman/listinfo/tinyerp-users
