Offloading machine learning to programmable data planes: A systematic survey
The demand for machine learning (ML) has increased significantly in recent decades,
enabling several applications, such as speech recognition, computer vision, and …
enabling several applications, such as speech recognition, computer vision, and …
In-network aggregation for data center networks: A survey
A Feng, D Dong, F Lei, J Ma, E Yu, R Wang - Computer Communications, 2023 - Elsevier
Aggregation applications are widely deployed in data centers, such as distributed machine
learning and MapReduce-like framework. These applications typically have large …
learning and MapReduce-like framework. These applications typically have large …
Using trio: juniper networks' programmable chipset-for emerging in-network applications
M Yang, A Baban, V Kugel, J Libby, S Mackie… - Proceedings of the …, 2022 - dl.acm.org
This paper describes Trio, a programmable chipset used in Juniper Networks' MX-series
routers and switches. Trio's architecture is based on a multi-threaded programmable packet …
routers and switches. Trio's architecture is based on a multi-threaded programmable packet …
In-network aggregation with transport transparency for distributed training
Recent In-Network Aggregation (INA) solutions offload the all-reduce operation onto network
switches to accelerate and scale distributed training (DT). On end hosts, these solutions …
switches to accelerate and scale distributed training (DT). On end hosts, these solutions …
Sifter: An {Inversion-Free} and {Large-Capacity} Programmable Packet Scheduler
Packet schedulers play a crucial role in determining the order in which packets are served.
They achieve this by assigning a rank to each packet and sorting them based on these …
They achieve this by assigning a rank to each packet and sorting them based on these …
On-fiber photonic computing
In the 1800s, Charles Babbage envisioned computers as analog devices. However, it was
not until 150 years later that a Mechanical Analog Computer was constructed for the US …
not until 150 years later that a Mechanical Analog Computer was constructed for the US …
Introducing packet-level analysis in programmable data planes to advance Network Intrusion Detection
R Doriguzzi-Corin, LAD Knob, L Mendozzi… - Computer Networks, 2024 - Elsevier
Programmable data planes offer precise control over the low-level processing steps applied
to network packets, serving as a valuable tool for analysing malicious flows in the field of …
to network packets, serving as a valuable tool for analysing malicious flows in the field of …
[HTML][HTML] Canary: Congestion-aware in-network allreduce using dynamic trees
The allreduce operation is an essential building block for many distributed applications,
ranging from the training of deep learning models to scientific computing. In an allreduce …
ranging from the training of deep learning models to scientific computing. In an allreduce …
Leo: Online {ML-based} Traffic Classification at {Multi-Terabit} Line Rate
Online traffic classification enables critical applications such as network intrusion detection
and prevention, providing Quality-of-Service, and real-time IoT analytics. However, with …
and prevention, providing Quality-of-Service, and real-time IoT analytics. However, with …
{THC}: Accelerating Distributed Deep Learning Using Tensor Homomorphic Compression
Deep neural networks (DNNs) are the de facto standard for essential use cases, such as
image classification, computer vision, and natural language processing. As DNNs and …
image classification, computer vision, and natural language processing. As DNNs and …