Abstract
The Internet of Things (IoT) has gained popularity in recent years. Today’s IoT applications are now increasingly deployed in cloud platforms to perform Big Data analytics. In cloud data center networks (DCN), TCP incast usually happens when multiple senders simultaneously communicate with a single receiver. However, when TCP incast happens, DCN may suffer from both throughput collapse for TCP burst flows and temporary starvation for TCP background flows. In this paper, we propose a software defined network (SDN)-based TCP congestion control mechanism, referred to as SDTCP, to leverage the features, e.g., centralized control methods and the global view of the network, in order to solve the TCP incast problems. When we detect network congestion on an OpenFlow switch, our controller can select the background flows and reduce their bandwidth by adjusting the advertised window of TCP ACK packets of the corresponding background flows so as to reserve more bandwidth for burst flows. SDTCP is transparent to the end systems and can accurately decelerate the rate of background flows by leveraging the global view of the network gained via SDN. The experiments demonstrate that our SDTCP can provide high tolerance for burst flows and achieve better flow completion time for short flows. Therefore, SDTCP is an effective and scalable solution for the TCP incast problem.
Highlights
The Internet of Things (IoT) is getting popular all over the world
The data center networks (DCN), which are the primary infrastructures for the delivery of cloud services, play an important role in data transmission to satisfy the diverse IoT applications
Weincast set up usually the following experiment scenario: there are N flows competing a 1experiment, Gbps link, and of them is the background flow with infinite data and the others are burst flows
Summary
The Internet of Things (IoT) is getting popular all over the world. Currently, IoT devices are used to sense/collect data and send the data to cloud s for further processing and analysis, such asBig Data analytics on real-time sensor streams [1]. IoT devices are used to sense/collect data and send the data to cloud s for further processing and analysis, such as. The data center networks (DCN), which are the primary infrastructures for the delivery of cloud services, play an important role in data transmission to satisfy the diverse IoT applications. In today’s DCN, TCP has been used as the de facto transport layer protocol to ensure reliable data delivery. The unique workloads scale and environments of the data center violate the WAN (wide area network) assumptions on which TCP was originally designed. A well-known reported open problem, i.e., TCP incast [2,3], was initially identified in distributed storage clusters [2] and has nowadays become a practical issue in DCN
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call