Name
Abstract | ||
|---|---|---|
Fog computing is a promising technology that enables users to perform time-sensitive IoT analytics at locations near the clients. Recent studies have shown the improved delivery performance of fog networks by comparing them with traditional cloud-based architectures. In this paper, we focus on further improving the delivery performance of a fog network. We first show that the performance of fog networks in their native form suffers from unfairness among the IoT clients, which results in causing heterogeneous delivery delays. As a consequence, the fog node (which performs analytics) would need to wait for a prolonged time, until all IoT clients send their data. In this paper, we address this critical fairness issue from the network point of view, and propose a framework to achieve fair delivery of clients' data to the fog node. To this end, we identify the transport layer aspects that contribute to the unfairness, and appropriately augment the transport layer in order to enable fairness among the IoT clients. We also develop an experimental prototype of MQTT-based IoT nodes using Raspberry Pi devices, and extensively study the network performance, as well as demonstrate the efficacy of the proposed framework. |
Year | DOI | Venue |
|---|---|---|
2017 | 10.1109/CCNC.2017.7983104 | 2017 14th IEEE Annual Consumer Communications & Networking Conference (CCNC) |
Keywords | Field | DocType |
fair throughput performance,MQTT-based IoT,time-sensitive IoT analytics,fog computing,fog networks,IoT clients,fog node,fair client data delivery,transport layer,MQTT-based IoT nodes,Raspberry Pi devices,network performance | Delivery Performance,Computer science,Server,Network architecture,Computer network,MQTT,Throughput,Analytics,Cloud computing,Network performance,Distributed computing | Conference |
ISSN | ISBN | Citations |
2331-9852 | 978-1-5090-6197-6 | 2 |
PageRank | References | Authors |
0.35 | 2 | 4 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yiming Xu | 1 | 2 | 0.35 |
| Veeramani Mahendran | 2 | 21 | 4.89 |
| Wei Guo | 3 | 4 | 1.41 |
| Sridhar Radhakrishnan | 4 | 341 | 43.65 |