Related papers: Vermilion: A Traffic-Aware Reconfigurable Optical …
Distributed deep learning (DDL) systems strongly depend on network performance. Current electronic packet switched (EPS) network architectures and technologies suffer from variable diameter topologies, low-bisection bandwidth and…
Interactive data visualization and exploration (DVE) applications are often network-bottlenecked due to bursty request patterns, large response sizes, and heterogeneous deployments over a range of networks and devices. This makes it…
Generative models have achieved remarkable success across various applications, driving the demand for multi-GPU computing. Inter-GPU communication becomes a bottleneck in multi-GPU computing systems, particularly on consumer-grade GPUs. By…
In digital circuit designs, sequential components such as flip-flops are used to synchronize signal propagations. Logic computations are aligned at and thus isolated by flip-flop stages. Although this fully synchronous style can reduce…
Cloud computing and AI workloads are driving unprecedented demand for efficient communication within and across datacenters. However, the coexistence of intra- and inter-datacenter traffic within datacenters plus the disparity between the…
Transformers process tokens in parallel but are temporally shallow: at position $t$, each layer attends to key-value pairs computed based on the previous layer, yielding a depth capped by the number of layers. Recurrent models offer…
Datacenters provide the infrastructure for cloud computing services used by millions of users everyday. Many such services are distributed over multiple datacenters at geographically distant locations possibly in different continents. These…
The implementation of 5G and the future deployment of 6G necessitate the utilization of optical networks that possess substantial capacity and exhibit minimal latency. The dynamic arrival and departure of connection requests in optical…
Reinforcement learning (RL) constitutes a promising solution for alleviating the problem of traffic congestion. In particular, deep RL algorithms have been shown to produce adaptive traffic signal controllers that outperform conventional…
The recent amalgamation of transformer and convolutional designs has led to steady improvements in accuracy and efficiency of the models. In this work, we introduce FastViT, a hybrid vision transformer architecture that obtains the…
The rapid growth of AI training has dramatically increased datacenter traffic demand and energy consumption, which has motivated renewed interest in optical circuit switches (OCSes) as a high-bandwidth, energy-efficient alternative for AI…
Vehicle-to-Everything (V2X) communications enable the exchange of information among vehicles to improve road safety and traffic efficiency. As V2X deployments progress, vehicles are expected to support an increasing number of V2X services,…
Financial time-series forecasting in real-world high-frequency markets is often hindered by delayed or partially stale observations caused by asynchronous data acquisition and transmission latency. To better reflect such practical…
The highly dynamic nature of Low-Earth Orbit (LEO) satellite networks introduces challenges that existing transport protocols fail to address, including non-congestive latency variation and loss, transient congestion hotspots, and frequent…
Realizing delay-capacity in intermittently connected mobile networks remains a largely open question, with state-of-the-art routing schemes typically focusing either on delay or on capacity. We show the feasibility of routing with both high…
Future Industrial Internet-of-Things (IIoT) systems will require wireless solutions to connect sensors, actuators, and controllers as part of high data rate feedback-control loops over real-time flows. A key challenge in such networks is to…
The automotive industry is transitioning from federated, homogeneous, interconnected devices to integrated, heterogeneous, mixed-criticality systems (MCS). This leads to challenges in achieving timing predictability techniques due to access…
This paper studies the design of self-adjusting networks whose topology dynamically adapts to the workload, in an online and demand-aware manner. This problem is motivated by emerging optical technologies which allow to reconfigure the…
In this paper we carry out a computational study of a novel microscopic follow-the-leader model for traffic flow on road networks. We assume that each driver has its own origin and destination, and wants to complete its journey in minimal…
Datacenters need networks that support both low-latency and high-bandwidth packet delivery to meet the stringent requirements of modern applications. We present Opera, a dynamic network that delivers latency-sensitive traffic quickly by…