PPoPP ExHET 2025

ExHET 2025

The 4th International Workshop on Extreme Heterogeneity Solutions

to be held in conjunction with
PPoPP 2025

02 March, 2025
Las Vegas, NV, USA

Introduction

While computing technologies have remained relatively stable for nearly two decades, new architectural features, such as specialized hardware, heterogeneous cores, deep memory hierarchies, and near-memory processing, have emerged as possible solutions to address the concerns of energy efficiency, manufacturability, and cost. However, we expect this ‘golden age’ of architectural change to lead to extreme heterogeneity and will have a major impact on software systems and applications. In this upcoming exascale and extreme heterogeneity era, it will be critical to explore new software approaches that will enable us to effectively exploit this diverse hardware to advance science, the next-generation systems with heterogeneous elements will need to accommodate complex workflows. This is mainly due to the many forms of heterogeneous accelerators (no longer just GPU accelerators) in this heterogeneous era, and the need to map different parts of an application onto elements most appropriate for that application component. In addition, this year we are acknowledging a now clear trend by explicitly encouraging contributions involving the use of trained AI methods -- for example, the use of GPT for code generation.

Objectives, scope and topics of the workshop

This workshop aims to provide a forum to discuss new and emerging solutions to address these important challenges from the upcoming extreme heterogeneity era. Papers are being sought on many aspects of heterogeneous computing including (but not limited to):

Program

TBA

Important Dates

Paper submission deadline : January 10th
Notification of acceptance : January 20th
Camera-ready papers due : TBA
Workshop day: March 02, 2024

Steering Committee

Antonio J. Pena, Barcelona Supercomputing Center, Spain

Hartwig Anzt, Technical University of Munich, Germany

Hyesoon Kim, Georgia Institute of Technology, USA

Ignacio Laguna, Lawrence Livermore National Laboratory, USA

Jeffrey S. Vetter, Oak Ridge National Laboratory, USA

Manuel Prieto, University Complutense of Madrid, Spain

Miwako Tsuji, RIKEN, Japan

Olivier Aumage, INRIA, France

Sunita Chandrasekaran, University of Delaware, USA

Toshiyuki Imamura, RIKEN, Japan

Organizers (Contact us)

Pedro Valero-Lara (co-chair)
Oak Ridge National Laboratory, USA
valerolarap@ornl.gov

Seyong Lee (co-chair)
Oak Ridge National Laboratory, USA
lees2@ornl.gov

Gokcen Kestor (co-chair)
Pacific Northwest National Laboratory, University of California Merced, USA
gokcen.kestor@pnnl.gov

Monil Mohammad Alaul Haque (proceeding chair and program chair)
Oak Ridge National Laboratory, USA
monilm@ornl.gov

Simon Garcia de Gonzalo (publicity and web chair)
Sandia National Laboratory, USA
simgarc@sandia.gov

Programme Committee

Manuscript submission

We invite submissions of original, unpublished research and experiential papers. Full papers should be between 6 to 8 pages in length, formatted according to the standard ACM two-column conference template. Additionally, we are introducing a short paper track for submissions up to 3 pages. ACM templates for both Microsoft Word and LaTeX can be accessed here. All paper submissions will be managed electronically via EasyChair.

Proceedings

All accepted papers will be published in the ExHET-PPoPP Workshops 2024 proceedings by the ACM Digital Library.

Best Paper Award

The Best Full and Short Paper Award will be selected on the basis of explicit recommendations of the reviewers and their scoring towards the paper’s originality and quality.

Special Issue Journal

Selected best papers of ExHET will be considered for publication in a special issue of the international journal Applied Sciences (Heterogeneous Computing Solutions) and FGCS (High-performance Computing Heterogeneous Systems and Subsystems)

Invited Speaker (Catherine Dorothy Schuman, University of Tennessee at Knoxville):

Neuromorphic Computing in a Future Heterogeneous Computing Landscape Dr. Catherine Schuman

Though neuromorphic computing systems have been demonstrated to have extremely low-power operation and are well-suited to edge deployment, there is also tremendous opportunity to leverage neuromorphic computers in future heterogeneous high performance computing environments. In this talk, we will discuss potential use cases of neuromorphic systems in these future HPC environments, with a particular focus on applications in which neuromorphic may provide a key advantage.

Catherine (Katie) Schuman is an Assistant Professor in the Department of Electrical Engineering and Computer Science at the University of Tennessee (UT). She received her Ph.D. in Computer Science from UT in 2015, where she completed her dissertation on the use of evolutionary algorithms to train spiking neural networks for neuromorphic systems. Katie previously served as a research scientist at Oak Ridge National Laboratory, where her research focused on algorithms and applications of neuromorphic systems. Katie co-leads the TENNLab Neuromorphic Computing Research Group at UT. She has over 70 publications as well as seven patents in the field of neuromorphic computing. She received the Department of Energy Early Career Award in 2019. Katie is a senior member of the Association of Computing Machinery and the IEEE.

Invited Speaker (Allen D. Malony, University of Oregon):

Extreme Heterogeneity Demands Extreme Observability Dr. Allen Malony

The extreme heterogeneity era, coupled with strong interests in exascale and AI computing, will significantly increase the complexity of next-generation HPC systems, applications programming/optimization, workflow execution, power/energy management, and more. Post-Moore computer architecture innovations are driving heterogeneity and hardware specialization, resulting in a diversity of processors, accelerators, memory, and interconnects. While there is certainly opportunities to build more powerful HPC systems based on these technology advances, the operational complexity of the heterogeneous system as a whole could make it challenging to achieve its full computational potential. A proposition is presented in the talk that heterogeneity in future HPC systems will necessitate greater support for observability. Indeed, a stronger argument will be made that observability should be an integral, comprehensive aspect of the systems design overall and that specific hardware and software dedicated to observability needs to be developed. The conclusion of this argument is that future HPC systems and the applications that run on them will not be able to deliver their full capabilities unless extreme observability is present.

Allen D. Malony is a Professor in the Department of Computer and Information Science (CIS) at the University of Oregon (UO) and the Director of the Oregon Advanced Computing Institute for Science and Society (OACISS). Malony's research interests are in parallel computing, performance analysis, supercomputing, and computational science. He has extensive experience in performance analysis of parallel systems and has developed performance evaluation tools for a variety of HPC machines. In particular, Malony's research group develops the TAU Performance System (TM), a leading open source parallel performance tool suite in use by many academic, governmental, and industrial projects around the world. He is CEO and Director of ParaTools, Inc., founded with Sameer Shende, who is the President and Director. ParaTools specializes in performance engineering for high-performance computing (HPC).

Panel:

Challenges and Solutions for the upcoming Extreme Heterogeneity Era

During the panel discussion, the panelists and those participants in the workshop will have the opportunity to discuss the fundamentals of extreme heterogeneity: challenges and solutions.

Panelists:

TBA

Registration

Information about registration at PPoPP 2025 website.