[hpc-announce] CFP: Workshop on Asynchronous Many-Task Systems and Applications 2023

Patrick Diehl pdiehl at cct.lsu.edu
Wed Sep 7 10:57:16 CDT 2022

Dear colleagues,

find below the call for abstracts and papers for the hybrid Workshop on 
Asynchronous Many-Task Systems and Applications 2023 held from February 
15th to 17th, 2023 in Baton Rouge, LA, USA.

As our compute capacity grows, science simulations are not only becoming 
bigger, but more complex. Simulations are carried out at multiple scales 
and using multiple kinds of physics at once. Boundaries are irregular, 
grids are irregular, computational domains can be dynamic and complex. 
In such scenarios, the ideal way to parallelize often cannot be 
statically determined. At the same time, hardware is becoming more 
heterogeneous and difficult to program. Increasingly, scientists are 
turning to asynchronous, dynamic parallelism in order to make the best 
use of increasingly challenging hardware. As a result, numerous 
frameworks, platforms, and specialized languages have sprung up to 
answer this need.

The objectives of this workshop are to bring together experts in 
asynchronous many-task frameworks, developers of science codes, 
performance experts, and hardware vendors to discuss the 
state-of-the-art techniques needed to program, analyze, benchmark, and 
profile these codes to achieve the maximum performance possible from 
modern machines. This workshop will promote a dialogue between these 
communities, and help identify challenges and opportunities for 
advancement in all the disciplines they represent.

The topics of interest include, but are by no means limited to:

Novel task-based runtime environments

Experiences of using task-based runtime environments for large 

Experiences comparing task-based runtime environments

Experiences gathered from porting one large-scale parallel solution to 
another, e.g., MPI to Charm++, etc.

Profiling and performance monitoring of task-based environments

Benchmarks for task-based runtimes

Tools for debugging programs using task-based runtimes

Challenges to task-based runtimes in scaling to large clusters

Hardware challenges and solutions in using task-based environments

We are looking forward to your abstract and paper submission.

Kind regards,

Patrick Diehl (on behalf of the organizers)

Call for Abstracts

Submitted talks are 20 minutes (15 minutes for presentation and 5 
minutes for questions). Please use EasyChair for your submission. The 
submission should not exceed 500 words and should not include 
references. Please indicate by the topic if you are submitting a poster 
or talk.

To submit your abstract, please follow the below link to the submission 


      Deadline for abstract and poster submission: October 15, 2022.
      Author notification for poster and talks: November 1, 2022.

Note that at least one of the authors of each abstract accepted for 
presentation in WAMTA 23 must be registered.

Call for Papers

Authors need to submit their work through EasyChair. Submissions must be 
in Springer LNCS format and between ten and twelve pages, including 
text, the references section, appendices, and figures.

At least three program committee members will review each submission. 
Reviews will be double-blind, and papers get assessed for quality, 
relevance, and presentation of contributions. Springer LNCS will publish 
the accepted papers. All submitted manuscripts will be checked for 
originality by iThenticate (papers that show an insufficient originality 
might be rejected without a review).


      Deadline for proceedings paper submission: January 15, 2023.
      Author notification for proceedings papers: February 8, 2023.
      Deadline for camera-ready paper submission: February 19, 2023.

Patrick Diehl
Center for Computation and Technology
Louisiana State University
Digital Media Center
340 E Parker Blvd, Baton Rouge, LA 70803, USA
Cell: 225-284-1710

More information about the hpc-announce mailing list