Compare Tornado and Pyramid
Tornado is a Python web framework and asynchronous networking library developed at FriendFeed. By using non-blocking network I/O, Tornado can scale to tens of thousands of open connections, making it ideal for long polling, WebSockets, and other applications that require a long-lived connection to each user. It is great for use cases that are I/O intensive (e.g., proxies) but not ideal for compute-intensive use cases.
Pyramid is a web framework that is written in Python and is based on WSGI. It is minimalistic and inspired by Zope, Pylons, and Django. Before 2010, Pyramid was known as repoze.bfg.
Let's see how Tornado and Pyramid compare on various factors and features and which to choose when.
Type
Python web framework and asynchronous networking library, which does non-blocking I/O.
Type
Python "microframework" for building web applications.
Used by 145,000 projects.
Used by
FriendFeed, Turntable.fm, Fantamaster.it
Used by
Reddit, Yelp, Mozilla, SurveyMoney, NewCars.com.
300 job openings which list Tornado as a requirement.
80 job openings which list Pyramid as a requirement.
Claims to have top-notch performance but virtue of being a microframework but
benchmarks prove it to be slower than even Django beyond simple use cases.
Flexibility
Tornado is simple and flexible.
Flexibility
Pyramid is flexible.
Ease of Learning
Tornado is not difficult to learn if the user is familiar with asynchronous and non-blocking I/O. The online resources, courses, and tutorials are not as plentiful compared to Flask or Django.
Ease of Learning
Being a microframework, the learning curve isn't very steep. However, it doesn't have the same amount of tutorials, courses, and resources available for leaning compared to Django and Flask
RDBMS Support
Through Plugins or Extensions
RDBMS Support
Through Plugins or Extensions
No built-in ORM framework. Leaves it up to developers to choose a library like
SQLAlchemy
NoSQL Support
No built-in support for NoSQL databases, but 3rd party libraries support asynchronous, non-blocking access.
Motor is one such library from the makers of MongoDB. However, a lot of client libraries like
DynamoDB appear to have been deprecated or not actively maintained.
NoSQL Support
Doesn't get in the way of using NoSQL, and they are supported just as well as a relational database. Read more
here.
Admin Dashboard
No built-in admin panel. No well-known 3rd party tools either.
Admin Dashboard
Through Plugins or Extensions
REST Support
No built-in support for REST API, but users can implement REST APIs manually.
REST Support
Because it is a minimal framework, users can build a REST API themselves easily, or use a 3rd party library like
Cornice.
Security
Built-in
security mechanisms such as secure cookies, XSRF, DNS Rebinding and etc. Has been used in production for many years so security is generally decent.
Security
Built-in protection against CSRF, but
XSS vulnerabilities have been discovered in older versions.
Templating Library
Tornado uses
custom templating library out of the box.
Templating Library
Doesn't ship with a default templating language and leaves it up to developers to add one.
Jinja2 can be added to Pyramid. Another engine popular amongst Pyramid users is
Chameleon .
Web Forms
No built-in support.
Web Forms
No built-in support. Prefers 3rd party
Deform library.
Authentication
Provides user authentication and also supports 3rd party authentication and authorization systems like Google, Twitter, Facebook, etc.
Testing
Built-in support using Python's
unittest framework.
How is performance rating determined?
Performance rating is determined using reputable online benchmarks listed below.
Where is job data coming from?
Job data is collected from Indeed, Google Jobs and Stack Overflow jobs.
How is popularity calculated?
Popularity is calculated using a formula which looks at weighted score on the following publicly available data points:
- Popularity per Google Trends
- Number of GitHub Users
- Number of GitHub Stars
How is this calculated?
Ease of learning is calculated using the following data:
- Number of features and depth of tool.
- Number of online resources: articles, blogs, tutorials and YouTube videos.
- Number of courses
- Freshness of online material
For example, a microframework may not have a lot of online resources but still get a high-rating because it's minamalistic and easy to learn just by following official documentation.
If you found this useful, please help us grow by sharing this article with your followers using the sharing icons. Every share or call out will help. Thank you.
Similar Comparisons
