CUBIC is a congestion control protocol for TCP (transmission protocol) and the current default algorithm in Linux. The modifies linear window growth function of existing standards to be cubic order improve scalability over fast long distance networks. It also achieves more equitable bandwidth allocations among flows with different RTTs (round trip times) by making independent RTT -- thus those grow their at same rate. During steady state, increases size aggressively when far from saturation...

High-speed networks with large delays present a unique environment where TCP may have problem utilizing the full bandwidth. Several congestion control proposals been suggested to remedy this problem. The existing protocols consider mainly two properties: friendliness and bandwidth scalability. That is, protocol should not take away too much from standard flows while of high-speed networks. This work presents another important constraint, namely, RTT (round trip time) unfairness competing...

This paper presents a distributed implementation of RAND, randomized time slot scheduling algorithm, called DRAND. DRAND runs in O(\delta ) and message complexity where \delta is the maximum size two-hop neighborhood wireless network while remains ), assuming that delays can be bounded by an unknown constant. first fully version RAND. The algorithm suitable for most nodes do not move, such as mesh networks sensor networks. We implement TinyOS demonstrate its performance real testbed Mica2...

Wavelength-division multiplexing appears to be the solution of choice for providing a faster networking infrastructure that can meet explosive growth Internet. Several different technologies have been developed so far transfer data over WDM. We survey two new which are still in experimental stage-optical packet switching and optical burst switching-and comment on their suitability transporting IP traffic.

This paper presents a distributed implementation of RAND, randomized time slot scheduling algorithm, called DRAND. DRAND runs in O(δ) and message complexity where δ is the maximum size two-hop neighborhood wire-less network while remains O(δ), assuming that delays can be bounded by an unknown constant.DRAND first fully version RAND. The algorithm suitable for wireless most nodes do not move,such as mesh networks sensor networks.We implement TinyOS demonstrate its performance real testbed...

The Internet has recently been evolving from homogeneous congestion control to heterogeneous control. Several years ago, traffic was mainly controlled by the traditional RENO, whereas it is now multiple different TCP algorithms, such as CUBIC, and Compound (CTCP). However, there very little work on performance stability study of with One fundamental reason lack deployment information algorithms. In this paper, we first propose a tool called Congestion Avoidance Algorithm Identification...

We investigate the application of network coding to all-optical networks from both algorithmic and infrastructural perspectives. study effectiveness using for optical-layer dedicated protection multicast traffic that provides robustness against link failures in network. present a heuristic solving this problem compare it with inefficient optimal methods non-network-coding approaches. Our experiments show our near-optimal performance while significantly outperforming existing approaches...

The Internet has recently been evolving from homogeneous congestion control to heterogeneous control. Several years ago, traffic was mainly controlled by the traditional AIMD algorithm, whereas is now many different TCP algorithms, such as AIMD, BIC, CUBIC, and CTCP. However, there very little work on performance stability study of with One fundamental reason lack deployment information algorithms. In this paper, we first propose a tool called Congestion Avoidance Algorithm Identification...

We study limitations of an equation-based congestion control protocol, called TFRC (TCP Friendly Rate Control). It examines how the three main factors that determine throughput, namely, TCP friendly equation, loss event rate estimation and delay estimation, can influence long-term throughput imbalance between TCP. Especially, we show different sending rates competing flows cause these to experience rates. There are several fundamental reasons why have average rates, from first place. Earlier...

Most of the commercial P2P video streaming deployments support hundreds channels and are referred to as multichannel systems. Measurement studies show that bandwidth resources different highly unbalanced thus recent research have proposed various protocols improve qualities for all by enabling cross-channel cooperation among multiple channels. However, there is no general framework comparing existing potential designs multi-channel The goal this paper establish tractable models answering...

We study limitations of an equation-based congestion control protocol, called TCP-friendly rate (TFRC). It examines how the three main factors that determine TFRC throughput, namely, equation, loss event estimation, and delay can influence long-term throughput imbalance between TCP. Especially, we show different sending rates competing flows cause these to experience rates. There are several fundamental reasons why TCP have average rates, from first place. Earlier work shows convexity...

TCP-Friendly Rate Control (TFRC) is being adopted in Internet standards for congestion control of various streaming media applications. In this paper, we consider the transmission pre-recorded from a server to client by using TFRC, and analytically study impact TFRC on user-perceived quality, which roughly measured calculating rebuffering probability. A probability defined be that total duration all events experienced user longer than certain threshold. Two approaches are presented help an...

Multi-view peer-to-peer (P2P) live streaming systems have recently emerged, where a user can simultaneously watch multiple channels. Previous work on multi-view P2P solves the fundamental inter-channel bandwidth competition problem at individual peer level, and thus be used with very limited types of protocols. In this paper, we propose new protocol for streaming, called divide-and-conquer (DAC), which efficiently using divide-and conquer strategy channel is flexible to various This makes...

Data Center Network architectures (DCN) are evolving for increased scalability, performance, and low network complexity. In this paper, we propose HyScale, a switch-centric DCN architecture using hybrid optical networks. HyScale employs Optical Burst Switching Circuit technologies transmitting high volumes of data respectively in center. The proposed is highly scalable, recursively defined, fault-tolerant, has It also multitude desirable graph-theoretic properties like bisection width,...