3 ms·
I have the same question. This benchmark seem to be comparing apples and oranges. I'm just moving from sysops to dev, so I hope someone can explain, but here is
by 18pfsmt 16y ago
I have the same question. This benchmark seem to be comparing apples and oranges. I'm just moving from sysops to dev, so I hope someone can explain, but here is how I see the three items being compared:
Django- a python web framework
Node.js- a javascript server framework
Tornado- a non-blocking webserver
- Swizec 16y agoThe thing I was going for is that all three of these get used to handle concurrent requests so it pays to know how they compare in this task. All of them act as a server, often behind nginx or something, all of them take a request and pump out a response. They're not that different.
- lamby 16y ago> All of them act as a server Django is not a server. How did you even make it serve HTTP requests for your benchmark? (Please don't say "runserver")
- Swizec 16y ago./manage.py runfcgi The django docs recommended way of running it behind nginx.
- mrud 16y agoSo you compared nginx with django via fastcgi to node and tornado? JFTR, this benchmarks feels strange. Please describe next time how you setup django/tested it and describe the whole software stack. This is important while comparing and understanding benchmarks. You never provided details about your django setup.
- ericmoritz 16y agonginx+uwsgi FTW