Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Niphlod
there's the usual ton on optimizations to be made. BTW: if sessions aren't needed at all, the statement to use is session.forget(response) On Tuesday, June 9, 2015 at 12:55:52 AM UTC+2, Derek wrote: > > I created two pull requests, so it should be a bit faster now. I see it's > also storing sess

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Alfonso de la Guarda
Hi, Right now i am using pandas to format the data in a short time i will post an ipython file Saludos, Alfonso de la Guarda Twitter: @alfonsodg Redes sociales: alfonsodg Telef. 991935157 1024D/B23B24A4 5469 ED92 75A3 BBDB FD6B 58A5 54A1 851D B23B 24A4 On Mo

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Derek
I created two pull requests, so it should be a bit faster now. I see it's also storing sessions in the database and logs, no need for that... On Monday, June 8, 2015 at 3:44:04 PM UTC-7, Alfonso de la Guarda Reyes wrote: > > Massimo, > > As Derek shows we can change anything and rerun the test

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Alfonso de la Guarda
Massimo, As Derek shows we can change anything and rerun the tests Saludos, Alfonso de la Guarda Twitter: @alfonsodg Redes sociales: alfonsodg Telef. 991935157 1024D/B23B24A4 5469 ED92 75A3 BBDB FD6B 58A5 54A1 851D B23B 24A4 On Mon, Jun 8, 2015 at 5:41 PM, M

[web2py] Re: Python framework benchmarks

2015-06-08 Thread Derek
https://github.com/TechEmpower/FrameworkBenchmarks/tree/master/frameworks/Python/web2py It's open source, easy to see, perhaps we can create pull requests for our changes... On Monday, June 8, 2015 at 3:32:27 PM UTC-7, Niphlod wrote: > > it's the usual weirdo-test. I bet all has been done with w

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Massimo Di Pierro
Where is the code you run? Did you session.forget() in web2py or did you create a session at every request? Also, did you disable migrations? Remember web2py defaults to everything on, Flask instead, for example, defaults to everything off. On Monday, 8 June 2015 17:30:30 UTC-5, Alfonso de la G

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Derek
It's very difficult to compare the performances using that json file you posted. Is there another way to review the data? On Monday, June 8, 2015 at 3:30:30 PM UTC-7, Alfonso de la Guarda Reyes wrote: > > Hi, > > The idea was not to compare web2py itself, as I said I use it in the > frontend b

[web2py] Re: Python framework benchmarks

2015-06-08 Thread Niphlod
it's the usual weirdo-test. I bet all has been done with web2py defaults, which means migrations, sessions on disk, etc etc etc ?! BTW there's no news in knowing that a "minimal-something" performs better than a "batteries-included" one, let alone the "I was built specifically towards something

Re: [web2py] Re: Python framework benchmarks

2015-06-08 Thread Alfonso de la Guarda
Hi, The idea was not to compare web2py itself, as I said I use it in the frontend but I found it interesting how they will review other frameworks, but also lead to reflection on how to increase the performance (requests per second) for web2py Saludos, Alfonso de

[web2py] Re: Python framework benchmarks

2015-06-08 Thread Derek
well, that's not a fair bet, falcon will compile itself with cython if it's available. On Monday, June 8, 2015 at 3:05:05 PM UTC-7, Alfonso de la Guarda Reyes wrote: > > Hello, > > As I'm finishing a project that requires processing information very > quickly (machine learning) and lately i am