This special section explores new foundational and translational research toward enabling exascale computing for emerging scientific and societal challenges. Exascale computing (1,000 times more powerful than petascale) is defined as the capability to perform 1018 operations per second. Productively harnessing such a scale of processing, storage, and networking capabilities for diverse domains—including high-performance computing (HPC) simulations, artificial intelligence (AI), and extreme data-driven computing—relies on not only revitalizing existing parallel and distributed computing technologies but also innovating new solutions. Hence, topics related to applications, programming environments, runtimes, libraries, innovative algorithms, domain-specific frameworks, systems architecture, performance analysis, data processing, and networking technologies, among others, are within the scope of research encompassing parallel, distributed, and heterogeneous systems for exascale.
Paths to exascale have been laid out across various national and international funding programs that are largely characterized by multidisciplinary research spanning from exascale centers of excellence for domain sciences to the development of operational IT infrastructure at scale. Characteristic features of the IT infrastructure technologies include accelerated processors, memory and storage hierarchies, smart networks for workflows, and storage for HPC and AI, among others. Software technologies and applications focus on developing next-generation scalable algorithms and frameworks, enabling performance portability, and enhancing productivity of complex parallel and distributed workflows.
Contributions highlighting fundamental advances in software technologies, hardware, and applications over the current state-of-the art will be considered. Submissions addressing challenges that are unique to exascale computing, from application and algorithm development to reliability and availability of operational IT infrastructure, are within the scope of the special section. Topics of interest include, but are not limited to:
TPDS has recently started a new initiative called “special sections.” Compared with regular submissions to TPDS, special sections have some differences: (1) submissions are focused on special topics of interest (similar to special issues); (2) special sections have fixed deadlines for submission and notifications; and (3) special sections have a standing committee of reviewers similar to conferences.
The timeline for the submission and review process is as follows. All deadlines are 23:59 (11:59pm) anywhere on earth (https://www.worldtimeserver.com/time-zones/aoe/).
Round 1:
Round 2a (only for papers that get a minor revision in Round 1):
Round 2b (only for papers that get a major revision in Round 1):
Round 3 (only for papers that got a minor revision in Round 2b):
Submissions to the special section will be received as TPDS regular papers (survey and comment-style papers are not allowed). Please check submission instructions (including page limit, manuscript format, and author templates) on the TPDS Author Information page.
Authors can submit papers until the deadline through ScholarOne. Once you start the submission process, in Step 1 of the process, you’ll be asked to pick a “Type” for the paper. Please pick “SS on Innovative R&D toward the Exascale Era.”
All papers need to have sufficient new content and contributions (see examples of extension material below) to warrant a separate publication. While the specific amount of acceptable new content is subjective and depends on the reviewer, we estimate that most reviewers expect new material that represents novel research contributions beyond the original publication. Acceptance of the paper is based on this new content and its contributions. Old content from previous conference papers is mainly to help reviewers understand the context. Old content should be clearly cited from the original source. Furthermore, any content used verbatim from previous publications should be appropriately quoted and cited to avoid self-plagiarism.
Authors submitting an extension of a prior publication should clearly respond to the following questions:
1. What are the novel contributions of the submitted paper (beyond the authors’ previous publication(s))?
2. What is the new content and in which sections does this content appear in the submission?
3. How do the contributions (and content) build on the previous published material?
Examples of Extension Material:
Acceptable new content and contributions: new conceptual extensions, experiments that provide new insights, and new theoretical analysis and/or proofs supporting empirical results
Allowable but insufficient content and contributions: extension to background and/or related work; elaboration on the same points in the introduction, observations, and conclusions; additional figures/plots that merely illustrate already-published content; and additional experimental results without new insights
Unacceptable content and contributions: simple union of content from multiple prior publications
Sadaf Alam (Swiss National Supercomputing Centre, ETH Zürich)
Lois Curfman McInnes (Argonne National Laboratory)
Kengo Nakajima (University of Tokyo, RIKEN Center for Computational Science)
