The Chapel Parallel Programming Language


Chapel Presentations

Featured Presentations

Making Parallel Computing as Easy as Py(thon), from Laptops to Supercomputers [video], HPE Dev Munch & Learn, online, April 20, 2022.
This talk provides background on Chapel, such as how it compares to other mainstream language and HPC programming models, along with some of its benefits in the Arkouda and CHAMPS applications.
Achieving Productivity at Scale with Chapel in User Applications, a SIAM PP22 Minisymposium, online, February 24, 2022
This session featured a series of talks about uses of Chapel in application codes developed in the community:
Generating GPU Kernels from Chapel's Features for Parallelism and Locality, Engin Kayraklioglu, SIAM PP22 minisymposium on Code Generation and Transformation in HPC on Heterogeneous Platforms, online, February 26, 2022.
This talk describes Chapel's recently added support for GPU programming, detailing the programming model and code generation strategy.
What's New With Chapel? Applications, Aggregators, and Accelerators [video], Brad Chamberlain, Northwest C++ Users' Group, online, January 19, 2022.
This talk provides an introduction to Chapel through some recent highlights from the project, including the Arkouda and CHAMPS applications, high-level support for aggregated communication, and nascent GPU programming capabilities.

Timeless Talks

HPC Lessons from 30 Years of Practice in CFD Towards Aircraft Design and Analysis, [video], Éric Laurendeau, CHIUW 2021, online, June 4, 2021.
This CHIUW keynote describes CHAMPS, a ~48k-line framework written in Chapel for 3D unstructured computational fluid dynamics (CFD), while also providing an introduction to the role of HPC in Aerodynamics. The productivity benefits that Chapel brings to the CHAMPS team's work are made clear.
Arkouda: Chapel-Powered, Interactive Supercomputing for Data Science [video|Q&A], William Reus, CHIUW 2020, online, May 22, 2020.
This CHIUW keynote describes Arkouda, a Python package that provides a NumPy-like interface implemented using a Chapel server that scales to dozens of Terabytes of data at interactive rates.
Chapel’s Home in the New Landscape of Scientific Frameworks (and what it can learn from the neighbours) [video | PDF slides], Jonathan Dursi, CHIUW 2017 keynote, Orlando FL, June 2, 2017.
This keynote by Jonathan Dursi presents a survey of modern parallel computing frameworks as seen through the filter of the speaker's applications background, and describes Chapel's unique position within that landscape.
Chapel in the (Cosmological) Wild [video], Nikhil Padmanabhan (Yale University), CHIUW 2016, May 27, 2016.
This was the keynote talk at CHIUW 2016, reporting on the personal experiences of an Astrophysics Professor who's been looking at using Chapel in his research.

Other Recent Presentations

Rapid Prototyping by Example: Arkouda Argsort in Chapel, Brad Chamberlain, Rapid Prototyping for Exascale, ECP BoF Days, online, May 12, 2022.
This is a 5-minute opening statement about Chapel's role in supporting the rapid prototyping and scalability of Arkouda's Argsort operation, capable of sorting 72 TiB of data in half the time of the talk itself.
Chapel: Recent Successes, Ongoing Challenges, Brad Chamberlain, DOE Programming Systems Research Forum, online, February 28, 2022.
This talk provides an update to the DOE community about recent Chapel progress, along with a retrospective about how we got here and some research challenges going forward.
Parallel Programming in Chapel: Overview and Ookami [video], Brad Chamberlain, Ookami User Community Webinar, online, January 13, 2022.
This talk covers Chapel's motivation and some key benchmarking and application results before doing a deep-dive overview of the language and how it compares to SPMD programming. Initial baseline results of Chapel on Ookami are presented, with possible avenues for improving them given.
Separating Parallel Performance Concerns using Chapel [video], Michelle Strout, LCPC 2021, online, October 13, 2021.
This invited talk discusses the importance of separating concerns in High-Performance Computing, and the role of Chapel's multiresolution programming model in supporting such separation.
CHIUW 2021 talks, June 4, 2021
CHIUW 2021 was the latest instance of our annual Chapel workshop featuring talks on the Chapel language, its implementation, and its use in a variety of applications. See the program page for slides and videos from the talks.

Chapel Overviews

Chapel 101 [video], Brad Chamberlain, CHIUW 2020, online, May 22, 2020.
This is a fairly standard introduction to Chapel that was prepared for our annual Chapel workshop.
Chapel: An Example of Language-Level PGAS Support, Lydia Duncan, Practical and Efficient Partitioned Global Address Space Support for Data Intensive Applications minisymposium, SIAM AN21, online, July 22, 2021.
This talk describes Chapel's use for data-intensive computing within the context of CHAMPS and Arkouda, highlighting some of the benefits that a language-based approach to PGAS can yield.

Computations in Chapel

Arkouda: Terascale Data Science at Interactive Rates, [video], Ben Albrecht, SciPy 2020, online, July 7, 2020.
This is an introductory talk describing Arkouda with some recent performance comparisons to NumPy.
Simulating Ultralight Dark Matter with Chapel: An Experience Report [paper], Nikhil Padmanabhan, PAW-ATM 2019, Denver CO, November 17, 2019.
This talk describes a use of Chapel to explore dark matter in cosmological models.
Arkouda: NumPy-like arrays at massive scale backed by Chapel [paper], Mike Merrill, PAW-ATM 2019, Denver CO, November 17, 2019.
This talk describes the role of Chapel in supporting Exploratory Data Analysis (EDA) in Arkouda.

Implementing and Optimizing Chapel

Compiling Chapel: Keys to Making Parallel Programming Productive at Scale [video], Brad Chamberlain, PACT'20, online, October 7, 2020.
This talk gives a peek into what's required to compile some of Chapel's key features, and describes a pair of optimizations that are made possible through its unique features.
Multiresolution Support for Aggregated Communication in Chapel, Brad Chamberlain, OpenSHMEM 2021 keynote, online, September 16, 2021.
This keynote describes various forms of optimized and aggregated communications in Chapel for sparse communication patterns as exhibited by HPCC RA, Bale IndexGather, or Arkouda. Approaches include asynchronous fine-grain communications, manual copies expressed using Chapel's global namespace, and aggregation via user-level abstractions or compiler transformations.
Communication Optimizations for the Chapel Programming Language, Michael Ferguson, University of Maryland, March 24, 2016.
This talk describes the Chapel memory consistency model and how it enables two communication optimizations that have been implemented for Chapel.
Chapel: The Design and Implementation of a Multiresolution Language, Brad Chamberlain, Keynotes on HPC Languages, Lyon, France, June 30, 2013.
This talk is a fairly comprehensive overview of Chapel's themes, features, and status, with a bit more emphasis on the implementation and multiresolution design of the language than a typical talk allows for.

Exascale Programming and Chapel

Five Things You Should Do to Create a Future-Proof Exascale Language, Brad Chamberlain, PGAS 2015, Washington DC, September 17, 2015.
This talk was an editorial to the PGAS language community about things that should be done to create a successful language for exascale (or at all), combined with a summary of how we think Chapel is already achieving them.
Chapel Hierarchical Locales: Adaptable Portability for Exascale Node Architectures [poster], Greg Titus, SC14 Emerging Technologies Presentations, November 18, 2014.
This talk and poster provide an introduction to Chapel's hierarchical locales, a Chapel concept for making the language and user codes future-proof against future changes in node architecture.
Exascale: Your Opportunity to Create a Decent HPC Language, Brad Chamberlain, PPME workshop, Portland OR, August 14th, 2012
This talk was created as a call to arms for the DOE Exascale community, arguing that rather than simply being a time of challenges, Exascale can also be a time to break free of traditional lower-level programming models and create the first truly decent HPC programming language.
Exascale: An Opportunity to Atone for the Parallel Programming Models of the Past?, Brad Chamberlain, Punctuated Equilibrium at Exascale Panel/BoF, November 17th, 2011.
This is a panel talk arguing for programming models that are further abstracted from machine architecture and exascale as being an appropriate time for this change.

Chapel Design and Philosophy

If the HPC Community were to create a truly productive language...[how] would we ever know?, Brad Chamberlain, keynote at Dagstuhl Seminar on Performance Portability in Extreme Scale Computing: Metrics, Challenges, Solutions, Wadern Germany, October 23-27, 2017.
This keynote provided a review of some of the productivity metrics that were pursued under the DARPA HPCS program, but then argued that productivity seems like a very personal/social decision and that it therefore should be studied in forums supporting personal/social decisions. Two specific proposals are made.
A Language Designer's Perspective on Benchmarking Suites and Competitions, Brad Chamberlain, EMBRACE 2017 invited talk, Orlando FL, June 2, 2017
This talk surveys past approaches to benchmarking from a language designer's perspective, rating them along various axes of importance. It wraps up by advocating for an HPC equivalent to the Computer Language Benchmarks game.
Lessons Learned in Array Programming: from ZPL to Chapel, Brad Chamberlain, ARRAY 2016 keynote, June 14, 2016.
This keynote talk reflects on some of the successes of ZPL's support for data-parallel array-based programming, lists reasons that ZPL was ultimately limited, and how we addressed those limitations in Chapel's design.
Reflections on Programming Environments and Productivity (based on experiences with HPCS and Chapel), Brad Chamberlain, ASCR Exascale Computing Systems Productivity Workshop, Gaithersburg MD, June 3rd, 2014.
This talk briefly summarizes productivity-oriented metrics work undertaken by the Cray Cascade project during the HPCS program, along with a few anecdotal instances of Chapel productivity. It also provides some of Brad's personal takeaways from the experience.
Five Things About HPC Programming Models I Can Live Without, Sung-Eun Choi, DOE Workshop on Exascale Programming Challenges, July 27, 2011.
This talk lists some of the things that we think make HPC programming non-productive today and gives examples of how we are trying to address them in Chapel.
Five Key Parallel Design Decisions (for Multicore, Petascale, and Beyond), Brad Chamberlain, Barcelona Multicore Workshop, October 22, 2010.
This talk considers five design decisions that parallel language designers should wrestle with and how Chapel's design deals with them.

Archived Presentations