High-performance routing with multipathing and path diversity in ethernet and HPC networks

M Besta, J Domke, M Schneider… - … on Parallel and …, 2020 - ieeexplore.ieee.org
The recent line of research into topology design focuses on lowering network diameter.
Many low-diameter topologies such as Slim Fly or Jellyfish that substantially reduce cost …

Mitigating inter-job interference using adaptive flow-aware routing

SA Smith, CE Cromey, DK Lowenthal… - … Conference for High …, 2018 - ieeexplore.ieee.org
On most high performance computing platforms, concurrently executing jobs share network
resources. This sharing can lead to inter-job network interference, which can have a …

HyperX topology: First at-scale implementation and comparison to the fat-tree

J Domke, S Matsuoka, IR Ivanov, Y Tsushima… - Proceedings of the …, 2019 - dl.acm.org
The de-facto standard topology for modern HPC systems and data-centers are Folded Clos
networks, commonly known as Fat-Trees. The number of network endpoints in these …

[HTML][HTML] Speeding up the communications on a cluster using MPI by means of Software Defined Networks

P Gomariz-Martínez, FMD Martínez… - Future Generation …, 2024 - Elsevier
The Open MPI library is widely employed for implementing the message-passing
programming model on parallel applications running on distributed memory computer …

A self-adaptive network for HPC clouds: Architecture, framework, and implementation

F Zahid, A Taherkordi, EG Gran, T Skeie… - … on Parallel and …, 2018 - ieeexplore.ieee.org
Clouds offer flexible and economically attractive compute and storage solutions for
enterprises. However, the effectiveness of cloud computing for high-performance computing …

Interactive Investigation of Traffic Congestion on Fat‐Tree Networks Using TreeScope

H Bhatia, N Jain, A Bhatele, Y Livnat… - Computer Graphics …, 2018 - Wiley Online Library
Parallel simulation codes often suffer from performance bottlenecks due to network
congestion, leaving millions of dollars of investments underutilized. Given a network …

Jigsaw: a high-utilization, interference-free job scheduler for fat-tree clusters

SA Smith, DK Lowenthal - … of the 30th International Symposium on High …, 2021 - dl.acm.org
Jobs on HPC clusters can suffer significant performance degradation due to inter-job
network interference. Approaches to mitigating this interference primarily focus on reactive …

Improved power of two choices for fat-tree routing

S Wang, J Luo, WS Wong - IEEE Transactions on Network and …, 2018 - ieeexplore.ieee.org
The fat-tree networking topology have gained prominence in various parallel and distributed
systems such as high-performance computing clusters and data centers. To support high …

Network optimization for high performance cloud computing

F Zahid - 2017 - duo.uio.no
Cloud Computing has seen a tremendous popularity in last several years. A scalable and
efficient data center network is essential for a performance capable cloud computing …

The first supercomputer with hyperx topology: A viable alternative to fat-trees?

J Domke, S Matsuoka, I Radanov… - … IEEE Symposium on …, 2019 - ieeexplore.ieee.org
The state-of-the-art topology for modern supercomputers are Folded Clos networks, aka Fat-
Trees. The node count in these massively parallel systems is steadily increasing. This forces …