Advances in High-Performance Computing and Artificial Intelligence

The field of high-performance computing (HPC) and artificial intelligence (AI) is witnessing significant advancements, driven by the need for efficient and scalable solutions. Research is focused on developing innovative frameworks, libraries, and algorithms that can leverage the capabilities of emerging hardware and software architectures. One key area of focus is the integration of HPC and AI, with the development of seamless and scalable frameworks that can enable the efficient collaboration of these two disciplines. Another area of research is the development of efficient communication libraries and interfaces that can support the growing demands of multithreaded and asynchronous communication. Additionally, there is a growing interest in the application of machine learning and AI techniques to scientific workflows, with the development of modular and extensible middleware designed to deploy autonomous agents across federated research ecosystems. Noteworthy papers include: An Empirical Study on the Performance and Energy Usage of Compiled Python Code, which investigates the impact of compilation on Python code performance and energy efficiency. MojoFrame: Dataframe Library in Mojo Language, which introduces a novel dataframe library for the emerging Mojo programming language, offering significant performance improvements for relational operations. Accelerating Triangle Counting with Real Processing-in-Memory Systems, which presents a new algorithm for triangle counting that leverages the capabilities of processing-in-memory architectures, achieving significant performance gains over traditional CPU-based implementations.

Sources

Heterogeneous Memory Benchmarking Toolkit

LCI: a Lightweight Communication Interface for Efficient Asynchronous Multithreaded Communication

Scalable Genomic Context Analysis with GCsnap2 on HPC Clusters

An Empirical Study on the Performance and Energy Usage of Compiled Python Code

A Unifying Framework to Enable Artificial Intelligence in High Performance Computing Workflows

Automotive Middleware Performance: Comparison of FastDDS, Zenoh and vSomeIP

Benchmark-based Study of CPU/GPU Power-Related Features through JAX and TensorFlow

MojoFrame: Dataframe Library in Mojo Language

Accelerating Triangle Counting with Real Processing-in-Memory Systems

An Asynchronous Distributed-Memory Parallel Algorithm for k-mer Counting

Testing Message-Passing Concurrency

Empowering Scientific Workflows with Federated Agents

Built with on top of