Related papers: Enabling Work-conserving Bandwidth Guarantees for …
The single-chip crosspoint-queued (CQ) switch is a compact switching architecture that has all its buffers placed at the crosspoints of input and output lines. Scheduling is also performed inside the switching core, and does not rely on…
Pipeline parallelism has achieved great success in deploying large-scale transformer models in cloud environments, but has received less attention in edge environments. Unlike in cloud scenarios with high-speed and stable network…
This article defines the QoS-guaranteed efficient cloudlet deploy problem in wireless metropolitan area network, which aims to minimize the average access delay of mobile users i.e. the average delay when service requests are successfully…
Today, network devices share buffer across priority queues to avoid drops during transient congestion. While cost-effective most of the time, this sharing can cause undesired interference among seemingly independent traffic. As a result,…
Systems for processing big data---e.g., Hadoop, Spark, and massively parallel databases---need to run workloads on behalf of multiple tenants simultaneously. The abundant disk-based storage in these systems is usually complemented by a…
Quantum key distribution (QKD) provides an attractive means for securing communications in optical fibre networks. However, deployment of the technology has been hampered by the frequent need for dedicated dark fibres to segregate the very…
Quantum key distribution (QKD), which promises secure key exchange between two remote parties, is now moving toward the realization of scalable and secure QKD networks (QNs). Fully connected, trusted node-free QNs have been realized based…
Streaming algorithms are fundamental in the analysis of large and online datasets. A key component of many such analytic tasks is $q$-MAX, which finds the largest $q$ values in a number stream. Modern approaches attain a constant runtime by…
In mobile wireless networks, livestreaming in high user density areas presents two typical challenges: the wireless bandwidth is depleted and the number of users is limited. In this study, a media distribution model utilizing peer to peer…
In this paper, we consider a system model in conjunction with two major technologies in 5G communications, i.e., mobile edge computing and spectrum sharing. An IoT network, which does not have access to any licensed spectrum, carries its…
The rapid advancement of communication technologies has established cellular networks as the backbone for diverse applications, each with distinct quality of service requirements. Meeting these varying demands within a unified…
Cloud computing is a new paradigm where data and services of Information Technology are provided via the Internet by using remote servers. It represents a new way of delivering computing resources allowing access to the network on demand.…
The volume of data moving through a network increases with new scientific experiments and simulations. Network bandwidth requirements also increase proportionally to deliver data within a certain time frame. We observe that a significant…
Current practice of shaping subscriber traffic based on token bucket by Internet service provider (ISP) allows short-term fluctuations in its shaped rate and thereby enables a subscriber to transmit traffic at a higher rate than a…
A number of production deep learning clusters have attempted to explore inference hardware for DNN training, at the off-peak serving hours with many inference GPUs idling. Conducting DNN training with a combination of heterogeneous training…
QKD technology is being increasingly adopted inside the network core for protecting information transport against any form of computational attacks. However, the use of QKD for wide-area internetworking is still challenging and costly, due…
This paper presents a systematic approach for implementing a class of nonlinear signal processing systems as a distributed web service, which in turn is used to solve optimization problems in a distributed, asynchronous fashion. As opposed…
Peer-to-peer distributed storage systems provide reliable access to data through redundancy spread over nodes across the Internet. A key goal is to minimize the amount of bandwidth used to maintain that redundancy. Storing a file using an…
The Cloud Computing paradigm is providing system architects with a new powerful tool for building scalable applications. Clouds allow allocation of resources on a "pay-as-you-go" model, so that additional resources can be requested during…
Network slicing allows 5G network operators to provide service to multiple tenants with diverging service requirements. This paper considers network slicing aware optimal resource allocation in terms of throughput and energy efficiency. We…