Saved in:
Bibliographic Details
Main Authors: Beivide, Ramón, Camarero, Cristóbal, Martínez, Carmen, Vallejo, Enrique, Valero, Mateo
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2601.05668
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915718683951104
author Beivide, Ramón
Camarero, Cristóbal
Martínez, Carmen
Vallejo, Enrique
Valero, Mateo
author_facet Beivide, Ramón
Camarero, Cristóbal
Martínez, Carmen
Vallejo, Enrique
Valero, Mateo
contents Several interconnection networks are based on the complete graph topology. Networks with a moderate size can be based on a single complete graph. However, large-scale networks such as Dragonfly and HyperX use, respectively, a hierarchical or a multi-dimensional composition of complete graphs. The number of links in these networks is huge and grows rapidly with their size. This paper introduces LACIN, a set of complete graph implementations that use identically indexed ports to link switches. This way of implementing the network reduces the complexity of its cabling and its routing. LACIN eases the deployment of networks for parallel computers of different scales, from VLSI systems to the largest supercomputers.
format Preprint
id arxiv_https___arxiv_org_abs_2601_05668
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle LACIN: Linearly Arranged Complete Interconnection Networks
Beivide, Ramón
Camarero, Cristóbal
Martínez, Carmen
Vallejo, Enrique
Valero, Mateo
Hardware Architecture
Distributed, Parallel, and Cluster Computing
Networking and Internet Architecture
Several interconnection networks are based on the complete graph topology. Networks with a moderate size can be based on a single complete graph. However, large-scale networks such as Dragonfly and HyperX use, respectively, a hierarchical or a multi-dimensional composition of complete graphs. The number of links in these networks is huge and grows rapidly with their size. This paper introduces LACIN, a set of complete graph implementations that use identically indexed ports to link switches. This way of implementing the network reduces the complexity of its cabling and its routing. LACIN eases the deployment of networks for parallel computers of different scales, from VLSI systems to the largest supercomputers.
title LACIN: Linearly Arranged Complete Interconnection Networks
topic Hardware Architecture
Distributed, Parallel, and Cluster Computing
Networking and Internet Architecture
url https://arxiv.org/abs/2601.05668