CoCalc offers collaborative calculation in the cloud. This includes working with the full (scientific) Python stack, SageMath, Julia, R, Octave, and more. It also offers capabilities to author documents in LaTeX, R/knitr or Markdown, storing and organizing files, a web-based Linux Terminal, communication tools like a chat, course management and more.
- CoCalc -- the online service
- Github
- Mailing List
You can easily use CoCalc on your own computer for free by running a Docker image.
CoCalc was formerly called SageMathCloud. It started to offer way more than just SageMath and hence outgrew itself. The name was coined in fall 2016 and changed around spring 2017.
- John Jeng
- Harald Schilly
- Hal Snyder
- William Stein
- Greg Bard
- Rob Beezer
- Keith Clawson
- Tim Clemans
- Andy Huchala
- Jon Lee
- Simon Luu
- Nicholas Ruhland
- Todd Zimmerman
... and many others: See https://github.com/sagemathinc/cocalc/graphs/contributors
The copyright of CoCalc is owned by SageMath, Inc., and the source code here is released under the GNU Affero General Public License version 3+. See the included file LICENSE.md.
None of the frontend or server dependencies of SMC are themselves GPL licensed; they all have non-viral liberal licenses. If want to host your own SMC at a company, and need a different AGPL-free license, please contact help@sagemath.com.
"CoCalc" is a registered trademark.
- Client -- javascript client library that runs in web browser
- Load balancer/ssl -- HAproxy
- Database -- PostgreSQL
- Compute -- VM's running TCP servers (e.g., sage, console, projects, python3, R, etc.)
- Hub -- written in Node.js; primus server; connects with everything -- compute servers, database, other hubs, and clients.
- Storage -- Snapshots of project data
- HTTP server -- Nginx
Client Client Client Client ...
/|\
|
https://cocalc.com (primus)
|
|
\|/
HAproxy (load balancing...)HAproxy Admin (monitor and control system)
/|\ /|\ /|\ /|\
| | | |
|http1.1 | | |
| | | |
\|/ \|/ \|/ \|/
Hub<----> Hub<---->Hub<---> Hub <-----------> PostgreSQL <--> PostgreSQL <--> PostgreSQL ...
/|\ /|\ /|\
| | |
---------| | | (tcp)
| | |
| | |
\|/ \|/ \|/
Compute<-------->Compute<-->Compute <--- rsync replication to Storage Server, which has ZFS snapshots
We are grateful to BrowserStack for providing infrastructure to test CoCalc.
- node
- Postgres
pip install pyyaml
The following instruction don't install SMC. They're for development purposes only!
git clone https://github.com/sagemathinc/cocalc
-- copy repocd cocalc/src
npm run install-all
-- buildnpm test
-- run test suite (expected failures if your clock is not UTC)install.py all --compute --web
-- build and install some parts system-wide for development use- See
INSTALL.md
for more details.
For further options please go here.