Special Issue of Journal: Engineering Applications of Artificial Intelligence
on Artificial Intelligence for High-performance Computing systems
Description:
The need for High-performance computing (HPC) systems has boomed in the past years. The
latter finds its roots in the recent advances in technology, ranging from billions of connected
IoT devices generating and transmitting data to the massive usage of Artificial Intelligence
(AI) techniques requiring unprecedented processing power.
AI and HPC are thus strongly intertwined and increasingly being used together to develop
new and innovative solutions to a wide range of challenges.
One of the main ways that AI is used in HPC is to accelerate the training and deployment of
machine learning (ML) models. HPC systems can be used to train ML models much faster
and often provide access to specialized hardware, such as GPUs, well-suited for accelerating
machine learning workloads. However designing and implementing algorithms efficiently
exploiting such complex and heterogeneous systems opens numerous research challenges.
On the other hand, AI can also be used in HPC to optimize the performance of HPC systems
themselves. AI can help develop new algorithms and scheduling strategies to improve the
efficiency of HPC systems, making them better able to handle complex workloads or more
energy efficient. AI can also be used to monitor HPC systems and identify potential problems
before they cause outages or performance degradation.
Topics of interest:
Topics of interest include, but are not limited to:
● Design and deployment of AI algorithms on HPC systems, e.g.: deep learning,
reinforcement learning, evolutionary algorithms, swarm intelligence, ...);
● Distributed and federated learning algorithms on HPC systems;
● AI algorithms on GPU, TPU, FPGA, …;
● AI for energy efficient HPC and Cloud systems;
● AI for scheduling of HPC systems;
● Real world applications of AI on HPC systems: e.g., cloud computing, planning,
logistics, manufacturing, finance, telecommunications, bioinformatics, etc.
The organization of this special issue is linked to the 13th IEEE Workshop Parallel /
Distributed Combinatorics and Optimization (PDCO 2023), that took place together with the
37th IEEE International Parallel and Distributed Symposium (IPDPS 2023) in Saint
Petersburg, Florida, USA, May 2023. The most outstanding papers from the workshop will
be considered for possible publication in the special issue, after a peer review process.
The special issue is open to any submissions, and not restricted to papers from the PDCO
workshop.
Important Dates:
- First Submission Date: November 15th, 2023.
- Final Submission Deadline: June 15th, 2024.
- Notifications of Acceptance: October 15th, 2024.
Submission website:
https://www2.cloud.editorialmanager.com/eaai/default2.aspx
When submitting your paper to our Special Issue of the Journal please select:
“VSI: AI for HPC Systems”when you reach the "Article Type" step in the submission process.
Guest Editor(s):
Dr. Grégoire Danoy (Executive Guest Editor),
University of Luxembourg, Luxembourg
Email: gregoire.danoy@uni.lu
Areas of expertise: artificial intelligence, optimisation,
metaheuristics, machine learning,
Dr. Eng. Didier El Baz,
Team CDA, LAAS-CNRS, France
Email: elbaz@laas.fr
Areas of expertise: HPC, distributed
computing, optimization, scheduling and machine learning
