In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side a...In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side application. Towards this goal, we first develop an analytical model that formulates the incast probability as a function of connection variables and network environment settings. We combine the model with the optimization theory and derive some insights into minimizing the incast probability through tuning connection variables related to applications. Then,enlightened by the analytical results, we propose an adaptive application-layer solution to the TCP incast.The solution equally allocates advertised windows to concurrent connections, and dynamically adapts the number of concurrent connections to the varying conditions. Simulation results show that our solution consistently eludes incast and achieves high goodput in various scenarios including the ones with multiple bottleneck links and background TCP traffic.展开更多
In a translucent network scenario, development of an optical control plane (OCP) that is aware of the location and number of available regenerators and all-optical wavelength converters (AOWCs) is of paramount importa...In a translucent network scenario, development of an optical control plane (OCP) that is aware of the location and number of available regenerators and all-optical wavelength converters (AOWCs) is of paramount importance. However, current generalized multiprotocol label switching (GMPLS) protocol suite does not consider the distribution of regenerator and AOWC availability information to all the network nodes. In this paper, we propose a novel optical control plane (OCP) architecture that 1) disseminates information about network components (i.e. regenerators and AOWCs) to all the network nodes, and 2) evaluates candidate routes which use fewest amounts of network components. Performance of the proposed OCP is compared with a recently proposed hybrid OCP approach in terms of blocking performance, number of deployed components and lightpath establishment setup times. The obtained simulation results show that the proposed OCP approach demonstrates low connection blocking and establishes lightpaths by 1) minimizing the overall network cost owing to the deployment of minimum total number of network components, and 2) demonstrating acceptable lightpath establishment setup times at all traffic loads. Further, the proposed OCP methodology is compatible and suitable for controlling the operations of a novel electro-optical hybrid translucent node which is a latency efficient technology capable of delivering a cost effective implementation suitable for large scale deployment.展开更多
Transmission Control Protocol (TCP) is the most important transport layer protocol being used nowadays. It suffers from many problems over mobile networks especially over Cognitive Radio (CR). CR is one of the latest ...Transmission Control Protocol (TCP) is the most important transport layer protocol being used nowadays. It suffers from many problems over mobile networks especially over Cognitive Radio (CR). CR is one of the latest mobile technologies that brings its own share of problems for TCP. The buffer overflow for CR secondary network relay node can affect the performance of TCP. The contribution of this paper is the novel cross-layer model being used to evaluate the effect of the TCP congestion control on the secondary relay node buffer size in Cognitive Radio Network (CRN). The performance has been assessed by buffer overflow probability.展开更多
基金supported by the Fundamental Research Fundsfor the Central Universities under Grant No.ZYGX2015J009the Sichuan Province Scientific and Technological Support Project under Grants No.2014GZ0017 and No.2016GZ0093
文摘In data centers, the transmission control protocol(TCP) incast causes catastrophic goodput degradation to applications with a many-to-one traffic pattern. In this paper, we intend to tame incast at the receiver-side application. Towards this goal, we first develop an analytical model that formulates the incast probability as a function of connection variables and network environment settings. We combine the model with the optimization theory and derive some insights into minimizing the incast probability through tuning connection variables related to applications. Then,enlightened by the analytical results, we propose an adaptive application-layer solution to the TCP incast.The solution equally allocates advertised windows to concurrent connections, and dynamically adapts the number of concurrent connections to the varying conditions. Simulation results show that our solution consistently eludes incast and achieves high goodput in various scenarios including the ones with multiple bottleneck links and background TCP traffic.
文摘In a translucent network scenario, development of an optical control plane (OCP) that is aware of the location and number of available regenerators and all-optical wavelength converters (AOWCs) is of paramount importance. However, current generalized multiprotocol label switching (GMPLS) protocol suite does not consider the distribution of regenerator and AOWC availability information to all the network nodes. In this paper, we propose a novel optical control plane (OCP) architecture that 1) disseminates information about network components (i.e. regenerators and AOWCs) to all the network nodes, and 2) evaluates candidate routes which use fewest amounts of network components. Performance of the proposed OCP is compared with a recently proposed hybrid OCP approach in terms of blocking performance, number of deployed components and lightpath establishment setup times. The obtained simulation results show that the proposed OCP approach demonstrates low connection blocking and establishes lightpaths by 1) minimizing the overall network cost owing to the deployment of minimum total number of network components, and 2) demonstrating acceptable lightpath establishment setup times at all traffic loads. Further, the proposed OCP methodology is compatible and suitable for controlling the operations of a novel electro-optical hybrid translucent node which is a latency efficient technology capable of delivering a cost effective implementation suitable for large scale deployment.
文摘Transmission Control Protocol (TCP) is the most important transport layer protocol being used nowadays. It suffers from many problems over mobile networks especially over Cognitive Radio (CR). CR is one of the latest mobile technologies that brings its own share of problems for TCP. The buffer overflow for CR secondary network relay node can affect the performance of TCP. The contribution of this paper is the novel cross-layer model being used to evaluate the effect of the TCP congestion control on the secondary relay node buffer size in Cognitive Radio Network (CRN). The performance has been assessed by buffer overflow probability.
