Maximizing throughput on a dragonfly network
Interconnection networks are a critical resource for large supercomputers. The dragonfly
topology, which provides a low network diameter and large bisection bandwidth, is being …
topology, which provides a low network diameter and large bisection bandwidth, is being …
Barter-based cooperation in delay-tolerant personal wireless networks
In this paper, we consider the application of delay-tolerant networks to personal wireless
communications. In these networks, selfish nodes can exploit the services provided by other …
communications. In these networks, selfish nodes can exploit the services provided by other …
Comparing global link arrangements for dragonfly networks
High-performance computing systems are shifting away from traditional interconnect
topologies to exploit new technologies and to reduce interconnect power consumption. The …
topologies to exploit new technologies and to reduce interconnect power consumption. The …
Unveiling the interplay between global link arrangements and network management algorithms on dragonfly networks
Network messaging delay historically constitutes a large portion of the wall-clock time for
High Performance Computing (HPC) applications, as these applications run on many nodes …
High Performance Computing (HPC) applications, as these applications run on many nodes …
Level-spread: A new job allocation policy for dragonfly networks
The dragonfly network topology has attracted attention in recent years owing to its high radix
and constant diameter. However, the influence of job allocation on communication time in …
and constant diameter. However, the influence of job allocation on communication time in …
New link arrangements for dragonfly networks
M Belka, M Doubet, S Meyers, R Momoh… - 2017 IEEE 3rd …, 2017 - ieeexplore.ieee.org
Dragonfly networks have been proposed to exploit high-radix routers and optical links for
high performance computing (HPC) systems. Such networks divide the switches into groups …
high performance computing (HPC) systems. Such networks divide the switches into groups …
[PDF][PDF] Task Map** on a Dragonfly Supercomputer.
The dragonfly network topology has recently gained traction in the design of high
performance computing (HPC) systems and has been implemented in large-scale …
performance computing (HPC) systems and has been implemented in large-scale …
Brief Announcement: Coloring-based Task Map** for Dragonfly Systems
I Chinavinijkul, J Newcomb, L ** is the assignment of job tasks to nodes. Traditionally, the goal of task
map** is to maximize locality, reducing the number of network hops needed to deliver …
map** is to maximize locality, reducing the number of network hops needed to deliver …
Intelligent middleware for HPC systems to improve performance and energy cost efficiency
Y Zhang - 2021 - search.proquest.com
High-performance computing (HPC) systems play an essential role in large-scale scientific
computations. As the number of nodes in HPC systems continues to increase, their power …
computations. As the number of nodes in HPC systems continues to increase, their power …
Graph Constructions Derived from Interconnection Networks
R Draper - … Conference on Combinatorics, Graph Theory, and …, 2021 - Springer
A class of interconnection networks for massively parallel processors are designed by taking
copies of a building block network and wiring them together. For Dragonfly networks, the …
copies of a building block network and wiring them together. For Dragonfly networks, the …