[hpc-announce] CFP: ModSim 2022 Workshop on Modeling & Simulation of Systems and Architectures

[Hoisie, Adolfy]

Call for Abstracts

ModSim 2022: Workshop on Modeling & Simulation of Systems and Applications
August 10-12, 2022, University of Washington Botanic Gardens
Center for Urban Horticulture, Seattle

Workshop URL: https://www.bnl.gov/modsim2022/
Submission URL: https://easychair.org/conferences/?conf=modsim2022
EasyChair Submission Deadline: Sunday, May 22, 2022 (11:59 PM, Pacific Daylight Time [PDT])
Notification of Acceptance: Monday, June 06, 2022 (via e-mail)

To promote advancements in modeling and simulation (ModSim) research, we are soliciting community input in the form of abstracts. If accepted, author(s) will be invited to offer a poster and short presentation at the annual gathering of our community, the ModSim 2022 Workshop.

The overarching theme for this year's workshop is "Reflections on ModSim: Successes, Failures, and the Future." The emphasis will be on methodologies, tools, best practices, projects, and initiatives that aim to address the challenges and achieve the goal of modeling performance, power, and reliability of high-performance systems under a realistic application workload. Submissions related to this year's workshop theme, imparting lessons learned from specific projects, methods, tools, and use cases are highly encouraged.

Abstract Submission Guidelines
There is no set word limit for abstract submissions. However, please limit the submission to one page with no smaller than 11-point font type. The abstract should provide an overview that adequately summarizes the topic(s) presented and any proposed impact on ModSim research or techniques, especially any relevant to the workshop theme. The following details a proposed abstract layout and points to consider:

Abstract Title
Primary research areas:

- ModSim of architectures on the range of scale, from computing at the edge to extreme
- ModSim of applications and complex workflows
- ModSim of subsystems (memory, I/O, microprocessors, etc.)
- ModSim of distributed data analytics infrastructures
- Advances in ModSim implementation

The abstract should include specific aspects of the work and answer questions such as:

What is being modeled (e.g., performance, reliability, power, other)?
What is the target application?
What modeling techniques are being used?
What is novel about the approach versus current state of the art?
Are preliminary results or any notable lessons learned available?

All abstracts must be submitted through EasyChair (https://easychair.org/conferences/?conf=modsim2022) no later than Sunday May 22, 2022 (11:59 PM, PDT). Those with accepted abstracts will be notified via e-mail on Monday, June 06, 2022.

Dr. Sudhakar Yalamanchili Award
Submissions will be eligible for the Dr. Sudhakar Yalamanchili Award, which is intended to recognize young researchers for their outstanding contribution to the field of performance modeling and simulation. Presenters, who must be a student, postdoctoral researcher, or otherwise within six years of their highest awarded degree at the time of the ModSim conference, will be evaluated during the Contributed Presentation/Poster Session at the ModSim 2022 Workshop (https://www.bnl.gov/modsim2022/sudhaaward.php).


Topic Areas

Abstract contributions should relate to the workshop theme (as applicable) and focus on the following topical areas of interest:

Modeling and Simulation of Architectures Ranging from Computing at the Edge to the Extreme Scale
This thrust area will address new methods, tools, and frameworks for modeling, simulation, and emulation, as well as new areas for applying ModSim of performance, power, and reliability at all scales of computing. Emphasis should be placed on novel approaches and using ModSim rather than specific successes of existing methodologies.

Modeling and Simulation of Applications and Complex Computational Workflows
Here, the emphasis is on ModSim methods for describing workflows for complex computations consisting of multiple stages, such as setup, data collection and preparation, analysis, and numerical simulation. Both ModSim methods for workflows designed for performance prediction (static) and performance optimization (dynamic/at runtime) that capture the mapping of the workflows to heterogeneous architectures are of interest.

Modeling and Simulation of Subsystems
Memory device, storage, and input/output (I/O) manufacturing and design, as well as how these technologies are used by application software, profoundly impact current and future computing systems in terms of their performance, function, reliability, predictability, power dissipation, and cost. Meanwhile, emerging approaches offer the potential to overcome both technology- and design-related limitations to answer the requirements of many different applications. As such, modeling these subsystems is tremendously important, especially in affording the ability to characterize and quantify data movement behavior in large-scale systems. Submissions should relate to subsystems technology areas and their characterization or provide use cases that describe how ModSim can help overcome significant challenges.

ModSim of Distributed Data Analytics Infrastructures
Advances in sensor and detector technologies have expanded the HPC workload beyond one of just solving large batch-oriented, computationally intensive jobs. New, possibly interactive, data-intensive analysis jobs benefit from using a range of computing resources from edge to HPC. Capturing the performance capabilities of these complete data collection and analysis infrastructures requires new ModSim capabilities. The emphasis is on combining or linking models and simulation capabilities of individual subsystems to capture the performance and behavior of the complete end-to-end infrastructure.

Recent Advances in Modeling and Simulation Implementation
The rapidly increasing complexity of systems and application workloads - along with the blending of compute, memory devices, storage, and interconnect then further combined with application software — translates into unprecedented challenges within the ModSim field. Submissions in this are expected to highlight recent developments that can help overcome these significant challenges. Possible topics include, but are not limited to, novel ModSim methodologies, emerging areas of research and development, new projects or advances in existing efforts, and new applications of ModSim tools to real-life problems.