The Chapel Parallel Programming Language

What is Chapel?

Note: this page is no longer being maintained.
Please refer to this replacement page instead

Chapel is a programming language designed for productive parallel computing at scale.

Why Chapel?  Because it simplifies parallel programming through elegant support for:

• data parallelism to trivially use the cores of a laptop, cluster, or supercomputer
• task parallelism to create concurrency within a node or across the system
• a global namespace supporting direct access to local or remote variables
• GPU programming in a vendor-neutral manner using the same features as above
• distributed arrays that can leverage thousands of nodes' memories and cores

• productive: code tends to be similarly readable/writable as Python
• scalable: runs on laptops, clusters, the cloud, and HPC systems
• fast: performance competes with or beats conventional HPC programming models
• portable: compiles and runs in virtually any *nix environment
• open-source: hosted on GitHub, permissively licensed
  
• production-ready: used in real-world applications spanning diverse fields

New to Chapel?

• watch an overview talk or browse its slides
• read a chapter-length introduction to Chapel
• learn about projects powered by Chapel
• read about GPU programming in Chapel, or watch a recent talk or demo about it
• check out performance highlights like these:
  
• browse sample programs or learn how to write distributed programs like this one:

use CyclicDist;         // use the Cyclic distribution library
config const n = 100;   // use --n=<val> when executing to override this default

forall i in Cyclic.createDomain(1..n) do
  writeln("Hello from iteration ", i, " of ", n, " running on node ", here.id);

What's Hot?

• Chapel 2.3 is out now! Read its release announcement and download a copy
• Our fall quarterly newsletter covers recent highlights in the community

• This user interview blog series explores Chapel's use in scientific applications

• Check out this interview about Chapel published by HPCwire

• Learn to program GPUs in Chapel from this HPE Dev Meetup talk and demos

• If you missed ChapelCon '24 be sure to check out:
  • Paul Sathre's excellent keynote talk [slides | video]
  • Engin Kayrakloglu's blog article summarizing the event
  • Brad Chamberlain's annual State of the Project talk
  • Tutorial videos for Chapel and Arkouda
  • Slide and video links for all of the submitted talks

• Doing science in Python and needing more speed/scale? Maybe we can help?

• Also see: What's New?