English

Scalable and Efficient Intra- and Inter-node Interconnection Networks for Post-Exascale Supercomputers and Data centers

Hardware Architecture 2025-11-07 v1

Abstract

The rapid growth of data-intensive applications such as generative AI, scientific simulations, and large-scale analytics is driving modern supercomputers and data centers toward increasingly heterogeneous and tightly integrated architectures. These systems combine powerful CPUs and accelerators with emerging high-bandwidth memory and storage technologies to reduce data movement and improve computational efficiency. However, as the number of accelerators per node increases, communication bottlenecks emerge both within and between nodes, particularly when network resources are shared among heterogeneous components.

Keywords

Cite

@article{arxiv.2511.04677,
  title  = {Scalable and Efficient Intra- and Inter-node Interconnection Networks for Post-Exascale Supercomputers and Data centers},
  author = {Joaquin Tarraga-Moreno and Daniel Barley and Francisco J. Andujar Munoz and Jesus Escudero-Sahuquillo and Holger Froning and Pedro Javier Garcia and Francisco J. Quiles and Jose Duato},
  journal= {arXiv preprint arXiv:2511.04677},
  year   = {2025}
}
R2 v1 2026-07-01T07:25:06.903Z