The Chapel Parallel Programming Language

Fortress (<a href="http://en.wikipedia.org/wiki/Fortress_(programming_language)" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Fortress_(programming_language)</a>) and X10 (<a href="http://en.wikipedia.org/wiki/X10_(programming_language)" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;X10_(programming_language)</a>) were the other two languages that came out of the DOE HPCS program (<a href="http://en.wikipedia.org/wiki/High_Productivity_Computing_Systems" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;High_Productivity_Computing_Sys...</a>) that might be interesting if you liked Chapel.<p>Other aiming-for-HPC languages include co-array Fortran (<a href="http://en.wikipedia.org/wiki/Coarray_Fortran" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Coarray_Fortran</a>) and Unified Parallel C (<a href="http://en.wikipedia.org/wiki/Unified_Parallel_C" rel="nofollow">http:&#x2F;&#x2F;en.wikipedia.org&#x2F;wiki&#x2F;Unified_Parallel_C</a>).<p>I never really saw any of them while working in HPC, though. It was just Fortran, C, and sometimes Python. The Python would really just call out via SWIG to a C function for the numeric kernel.

The overview paper [1] is a reminder of the pace of change in computing:<p>&gt; DARPA’s HPCS program was launched in 2002 with five teams, each led by a hardware vendor: Cray Inc., Hewlett-Packard, IBM, SGI, and Sun.<p>It also has one of the more amusing euphemisms I&#x27;ve heard in a while:<p>&gt; In 2003, the HPCS program transitioned to phase II, and a <i>programmatic downselect</i> occurred, enabling the Cray, IBM, and Sun teams to pursue their proposed research plans.<p>[1] <a href="http://chapel.cray.com/papers/BriefOverviewChapel.pdf" rel="nofollow">http:&#x2F;&#x2F;chapel.cray.com&#x2F;papers&#x2F;BriefOverviewChapel.pdf</a>

This looks seriously cool. I need to spend a day or so really diving into the documentation, but if they can actually get this completely built and optimized it looks like it has the right features to be a systems language for the age of massively parallel and cloud computing.<p>I&#x27;m especially interested in the domain maps - I need to read more, but it looks like a system that separates the semantic view of a data structure (this is an array of integers, you can perform these functions on them) from the implementation details of in-memory layout and distribution across systems. Essentially, you seem to be able to adjust the algorithms on your data structures and the (extremely performance-critical) representation of them somewhat orthogonally.

Chapel emerged a while back from DARPA&#x27;s High Productivity Computing Systems project along with the Fortress (now defunct) and X10 programming languages. Out of curiosity, is there a reason Chapel is a getting attention now?

How not to promote a language: hide almost all examples of source code for the language behind at least two clicks in PDFs. I was finally able to find an example of defining a variable after visiting the Tutorials page and reading through the first two and a half powerpoints, the first two not containing a single line of Chapel code.<p>Seriously, if I need to spend ten minutes just to find out what the language looks like where an example is somewhere between the extremes of the ten lines of code in the &quot;quick reference&quot; chart and the 313-page formal language definition, I&#x27;m probably not going to dive into the setup process to get the thing running when I can just go learn Rust or whatever.<p><a href="https://twitter.com/MozElevator/status/368467593805328384" rel="nofollow">https:&#x2F;&#x2F;twitter.com&#x2F;MozElevator&#x2F;status&#x2F;368467593805328384</a><p>To someone on this project: you&#x27;ll get a heck of a lot more community behind your code if you have at least one or more pages that look like the following:<p><a href="http://dlang.org/comparison.html" rel="nofollow">http:&#x2F;&#x2F;dlang.org&#x2F;comparison.html</a> <a href="http://coffeescript.org/#overview" rel="nofollow">http:&#x2F;&#x2F;coffeescript.org&#x2F;#overview</a> <a href="https://www.python.org/doc/" rel="nofollow">https:&#x2F;&#x2F;www.python.org&#x2F;doc&#x2F;</a> <a href="http://doc.rust-lang.org/tutorial.html#syntax-basics" rel="nofollow">http:&#x2F;&#x2F;doc.rust-lang.org&#x2F;tutorial.html#syntax-basics</a> <a href="http://learnyouahaskell.com/starting-out" rel="nofollow">http:&#x2F;&#x2F;learnyouahaskell.com&#x2F;starting-out</a> <a href="http://tour.golang.org/#1" rel="nofollow">http:&#x2F;&#x2F;tour.golang.org&#x2F;#1</a> <a href="http://www.scala-lang.org/documentation/getting-started.html" rel="nofollow">http:&#x2F;&#x2F;www.scala-lang.org&#x2F;documentation&#x2F;getting-started.html</a> <a href="http://clojure.org/getting_started" rel="nofollow">http:&#x2F;&#x2F;clojure.org&#x2F;getting_started</a> <a href="http://www.php.net/manual/en/intro-whatis.php" rel="nofollow">http:&#x2F;&#x2F;www.php.net&#x2F;manual&#x2F;en&#x2F;intro-whatis.php</a>

I worked on the Chapel team in MN about eight years ago. I&#x27;m glad to see they&#x27;re still actively working on the project. It seemed very promising then.

This looks definitely interesting. I&#x27;ll have a closer look and play around with it a little to see how it goes.

From the little I saw, it seems to be a pretty nice language.