We propose Themis, a novel collective scheduling scheme that dynamically schedules collectives (divided into chunks) to balance the communication loads across all dimensions...
We propose Themis, a novel collective scheduling scheme that dynamically schedules collectives (divided into chunks) to balance the communication loads across all dimensions...
In this work we review a number of existing proposals for distributed payment systems that offer some form of auditability for regulators.
Our work provides insights into the shortcomings of capacity volume driven deployments. We provide a methodology to rank the contribution of each entity pending deployment towards...
The Third Generation Partnership Project (3GPP) has been adding support for Non-Terrestrial Networks (NTN) into the 5G standards, with increasing NTN features included between...
With new research focusing on developing wireless diffractive backhaul technologies to provide Non-Line-Of-Sight (NLOS) links, this paper evaluates the engineering-economic implications.
In this paper we introduce a novel link failure detection and resolution method that improves on the previous method across three dimensions: faster resolution, reliable troubleshooting and scalable implementation. We introduce BER Illusion Methodology (BIM) that is a highly scalable and resource efficient solution that significantly reduces the time taken to troubleshoot pluggable optical interconnects.
In this paper, we port the previously proposed ASTRASIM simulation platform on top of the versatile NS3 network simulator by introducing a portable network interface for ASTRA-SIM.
Our core design involves heuristic algorithms to select Hose-compliant traffic matrices and cross-layer optimization between the optical and IP networks. We evaluate the system performance in production and share insights from years of production experience.
In this paper, we propose NeuroPlan, a deep reinforcement learning (RL) approach to solve the network planning problem. This problem involves multi-step decision making and cost minimization, which can be naturally cast as a deep RL problem. We develop two important domain-specific techniques.
This poses a cross-layer challenge from the Traffic Engineering (TE) perspective that has not been considered before: “Which IP links should be restored and by how much to best match the TE objective?”