Usability is the most biggest problem of Pylons Pyramid.

[KlenPJ hopIJQy]

Usability is the most biggest problem of Pylons Pyramid.

I am  newer to Pylons Pyramid.  i am  care about the usability. 

when we want to built a biggest heavily load website , we care about how could we handler the heavily request.

but via pyramid document, we could not find out the right answer.

what ever the code write , it needs to be running on the real server. 

we should focus on the real world.

i need real tech white paper about how to built high load website.   such as a hot online trade site, a hot forum, a hot web game.

[John Anderson]

Pyramid is built with speed in mind and is really quick for the things it does but it stays out of your way and lets you make your own decisions.   If you need to build a highly scalable application, pyramid is a good choice, but you will have to do a little research and make the right decisions when building it.

Things you should look at:

Template Engine Performance: Jinja2 vs Mako vs Genshi vs Chameleon

Database Performance: MySQL vs Postgres vs NoSQL Options (Mongo, Redis Cassandra)

Caching and Indexing: Beaker, Memcached, Solr, Vagrant

Webservers: Gunicorn, Apache, Gevent

Queueing Long Running Processes: cron, celery, zmq, gevent

All these decisions truly depend on your needs and the application you are building and that is why Pyramid stays out of your way and why one single white paper wouldn't really be that helpful.

If you want to get more detailed with what you are building and the needs of the application we might be able to outline some best practices and help you make the right decisions though.

[Parnell Springmeyer]

Usability is *not* the biggest problem with Pylons OR Pyramid. Pyramid
has a bit of a learning curve but it's because Pyramid doesn't make all
of the decisions FOR YOU. It allows you to mix and match many different
technologies to meet your requirements.

If you have a high-load website you can get down and dirty with Pyramid
(it's EASY! Django is not so easy to do that with). You can use
technologies that are more domain specific, or unusual for your
requirements. You won't be forced into specific paradigm.

I believe it is not Pyramid that is the problem. I am using Pyramid for
my startup, VERY HIGH VOLUME traffic application and Pyramid rocks the
socks off of Pylons, Django, and many other frameworks.

Pyramid, is very "real world" focused!!! Spend some time with the docs,
build a play app so you understand how it all works, get to know it -
you will be pleasantly surprised.

KlenPJ hopIJQy <hufen...@gmail.com> writes:

--
Parnell "ixmatus" Springmeyer (http://ixmat.us)

[Robert Forkel]

I think you meant varnish for caching, Not vagrant. Other than that: this.

--
You received this message because you are subscribed to the Google Groups "pylons-discuss" group.
To post to this group, send email to pylons-...@googlegroups.com.
To unsubscribe from this group, send email to pylons-discus...@googlegroups.com.
For more options, visit this group at http://groups.google.com/group/pylons-discuss?hl=en.

[John Anderson]

On Wed, Nov 30, 2011 at 11:53 PM, Robert Forkel <xrot...@googlemail.com> wrote:

I think you meant varnish for caching, Not vagrant. Other than that: this.

Good catch 

[Robert Forkel]

I never built a high-load website. But doing so based on tech white
papers (real or otherwise) wouldn't inspire much confidence, I guess.
Benchmarking different frameworks for your particular use case and
publishing the results on the other hand would be welcome :) You would
probably get the additional benefit of fans of the respective
frameworks telling you how to do the benchmark right, which is
basically what you are looking for.
regards
robert

On Thu, Dec 1, 2011 at 2:41 AM, KlenPJ hopIJQy <hufen...@gmail.com> wrote:

> i need real tech white paper about how to built high load website.   such as
> a hot online trade site, a hot forum, a hot web game.
>
>

> --
> You received this message because you are subscribed to the Google Groups
> "pylons-discuss" group.

> To view this discussion on the web visit
> https://groups.google.com/d/msg/pylons-discuss/-/mBf7kkp_qLcJ.

[KlenPJ hopIJQy]

1. i am not say pyramid was not good .

2. high load website need a full  complete solution, i know it.

3. i want find some guide in pyramid 's document  for high load website, but not find . 

4. may be take more usefull example , tutorials for  newbie is good .

[Mengu]

well, you won't find such documents on any language's or framework's
web site either. you need an infrastructure in the first place for a
high load web application. it is not something to do with both db and
the web app in 1 server. how high load are we speaking here? 1M daily
users? 100M monthly users? are you or your partners ready to invest
min $100K for your app or in a cloud system such as amazon or
rackspace?

let me guide your way here.

1) go learn apache or nginx.
2) go learn your db of choice.
3) go learn a nosql solution.
4) go learn a caching solution.
5) go learn a messaging system.
6) go learn how to do web cluster
7) go learn how to do db cluster
8) go learn how to do mail cluster
9) go learn a low-level language.
10) go learn python.
11) go learn pyramid.
12) write your web app.

you see how pyramid is a little part of this infrastructure?

[Iain Duncan]

You know a slightly more descriptive version of that list would actually be a really useful document. I've never had to do highload site and I wouldn't even know where to start,  that list is really helpful.

Iain

--
You received this message because you are subscribed to the Google Groups "pylons-discuss" group.

[cd34]

http://highscalability.com/ has a number of case studies as does
http://www.allthingsdistributed.com/ and are pretty good references.

As for a consolidated FAQ, there are so many variables, there is never
a best for every solution. In every case, you have to know what your
stumbling blocks are in order to fix them. And you can't just look at
the benchmark of the day because many are not run properly or aren't
anywhere near testing real world situations.

For one project I'm working on, I thought my use case was rather
specific. After evaluating nine potential solutions and implementing
two test projects in two separate long polling/ajax push scenarios, I
realized that my initial design wouldn't work and that none of the
solutions were optimal for what I needed. So, I had to pull the
decision making back a layer and then I could implement almost any
long polling solution - each worked as long as I had the ability to
access named pipes.

The battles with noSQL versus SQL are almost as religious as which
noSQL solution is best. Do you need streaming map/reduce? do you care
about immediate or eventual consistency?

Every decision brings a dozen solutions and each solution needs to be
evaluated. There are people that will pick their favorites and become
experts in those and in 90% of the cases, you can usually shoehorn an
idea into the tools in your skillset. The fundamental differences
between riak, mongodb, couchdb, cassandra, etc. are there, but, it
would be hard to say that any of them couldn't be engineered to handle
most of the nosql use cases.

While it would be nice to say, a + b + c + d = fastest, you need to
qualify that as 'fastest in this situation'.

Welcome to the world of scaling. :)

[KlenPJ hopIJQy]

Thanks for every body help me , your tips will helpful to me , let me find some way. 

[Thomas G. Willis]

http://code.google.com/appengine/docs/python/overview.html

Because it's more about the infrastructure than the framework. 

I'm not saying it's the only option, but if you don't have time to do your own research, or the capital to hire talent away from google and build your own awesome data center, then this is a pretty good out of the box solution.