TCSC SCALE Challenge
The CCGRID International Scalable Computing Challenge (SCALE 2024) objective is to highlight and showcase real-world problem solving using computing that scales.
Effective solutions to many scientific and engineering problems require applications that can scale. There are different dimensions to application scalability. For example, applications can scale-up to a large number of cores on a compute unit, scale-out to utilize multiple distinct compute units, or exhibit elastic scaling to acquire and release resources on-demand, based on current need. Scaling can also span edge, cloud, cluster and accelerated resources. The result may be an application that can solve a larger problem, increase throughput, and/or reduce execution time. In order to scale, applications need to be supported by tools, middleware, software cyber-infrastructure, programming frameworks, computing infrastructure, etc. The SCALE Challenge is concerned with advances in application development and their supporting infrastructure to enable scaling.
Participants in the challenge will be expected to identify significant current real-world problems where scalable computing techniques can be effectively used, and design, implement, evaluate and demonstrate solutions. Problems that have a social impact are of particular interest.
TCSC SCALE Challenge Award
Authors should submit technical papers that outline the nature of the problem being solved, the solution technique, and offer preliminary evidence of the scalability achieved. The technical committee will shortlist finalists from these proposals, and invite them to present their work and offer a live demonstration of their system and/or software solution at the CCGRID 2024 conference.
A panel of experts will select a first prize winner at the conference, who will be presented the TCSC SCALE Challenge award for 2024 with a cash prize of US$1000 sponsored by IEEE TCSC.
Important Dates
All deadlines are in Anywhere on Earth (AoE), UTC-12 timezone.
Event | Date |
---|---|
Abstract Submission: | |
Paper Submission: | |
Acceptance notification: | |
Camera ready submission: | March 4, 2024 |
Chairs
Dragi Kimovski, University of Klagenfurt, Austria (dragi.kimovski@aau.at)
Jean Luca Bez, Lawrence Berkeley National Laboratory, USA (jlbez@lbl.gov)
Submission Guidelines
Proposals may be submitted as a technical paper of up to 6 pages, including figures and tables but excluding references, using the IEEE 2-column format for conference proceedings. Submissions must be done in PDF format.
The proposals should clearly outline:
- An overview of the problem being solved and the techniques employed.
- The motivating application scenario and its scaling requirements.
- The proposed solution, including system architecture, underlying concepts and technologies used, and the innovative aspects of the solution.
- Performance metrics (number of cores, computational throughput, I/O bandwidth, data/metadata operations, etc.), preliminary empirical evidence, and a qualitative description of how the application scales – scale-up, scale-out or any other type of scaling
- Impact of the solution, including its extensibility, the uniqueness of the solutions and the results, and the extent to which the proposed solution pushes the envelope in scalable computing
- An analysis of the solution and technology employed in contrast to related and alternative approaches from research and technology.
The submissions will undergo a single-blinded peer-review by a program committee based on the above metrics. Submitted papers must represent original unpublished research that is not currently under review for any other conference or journal. Up to five proposals will be shortlisted for presentation at the conference.
Shortlisted papers will be published through the IEEE Xplore digital library as part of the CCGRIDW companion proceedings.
The conference will be held in person physically. One author of each accepted proposal must register at the conference, and present their work and demonstration in-person at a special SCALE Challenge session. Participation by students and early-career researchers, especially in leadership roles, is strongly encouraged.
Please use the following Easychair link to submit your paper(s) for the SCALE Challenge: https://easychair.org/conferences/?conf=ccgrid2024