Packet loss

About: Packet loss is a research topic. Over the lifetime, 21235 publications have been published within this topic receiving 302453 citations.

LiveNet: Using Passive Monitoring to Reconstruct Sensor Network Dynamics

Abstract: We describe LiveNet, a set of tools and analysis methods for reconstructing the complex behavior of a deployed sensor network LiveNet is based on the use of multiple passive packet sniffers co-located with the network, which collect packet traces that are merged to form a global picture of the network's operation The merged trace can be used to reconstruct critical aspects of the network's operation that cannot be observed from a single vantage point or with simple application-level instrumentation We address several challenges: merging multiple sniffer traces, determining sniffer coverage, and inference of missing information for routing path reconstruction We perform a detailed validation of LiveNet's accuracy and coverage using a 184-node sensor network testbed, and present results from a real-world deployment involving physiological monitoring of patients during a disaster drill Our results show that LiveNet is able to accurately reconstruct network topology, determine bandwidth usage and routing paths, identify hot-spot nodes, and disambiguate sources of packet loss observed at the application level

Impact of Network Dynamics on User's Video Quality: Analytical Framework and QoS Provision

Abstract: We develop an analytical framework to investigate the impacts of network dynamics on the user perceived video quality. Our investigation stands from the end user's perspective by analyzing the receiver playout buffer. In specific, we model the playback buffer at the receiver by a G/G/1/? and G/G/1/N queue, respectively, with arbitrary patterns of packet arrival and playback. We then examine the transient queue length of the buffer using the diffusion approximation. We obtain the closed-form expressions of the video quality in terms of the start-up delay, fluency of video playback and packet loss, and represent them by the network statistics, i.e., the average network throughput and delay jitter. Based on the analytical framework, we propose adaptive playout buffer management schemes to optimally manage the threshold of video playback towards the maximal user utility, according to different quality-of-service requirements of end users. The proposed framework is validated by extensive simulations.

Joint Optimization of Chain Placement and Request Scheduling for Network Function Virtualization

Abstract: Compared with executing Network Functions (NFs) on dedicated hardwares, the recent trend of Network Function Virtualization (NFV) holds the promise for operators to flexibly deploy software-based NFs on commodity servers. However, virtual NFs (VNFs) are normally "chained" together to provide a specific network service. Thus, an efficient scheme is needed to place the VNF chains across the network and effectively schedule requests to service instances, which can maximize the average resource utilization of each node in service and simultaneously minimize the average response latency of each request. To this end, we formulate first VNF chains placement problem as a variant of bin-packing problem, which is NP-hard, and we model request scheduling problem based on the key concepts from open Jackson network. To jointly optimize the performance of NFV, we propose a priority-driven weighted algorithm to improve resource utilization and a heuristic algorithm to reduce response latency. Through extensive trace-driven simulations, we show that our methods can indeed enhance performance in diverse scenarios. In particular, we can improve the average resource utilization by 33.4% and can reduce the average total latency by 19.9% as compared with the state-of-the-art methods.

Throughput analysis of TCP on channels with memory

Abstract: The focus of this paper is to analyze the relative sensitivity of the bulk throughput performance of different versions of TCP, viz., OldTahoe, Tahoe, Reno, and New Reno, to channel errors that are correlated. We investigate the performance of a single wireless TCP connection in a local environment by modeling the correlated packet loss/error process (e.g., as induced by a multipath fading channel) as a first-order Markov chain. A major contribution of the paper is a unified analytical approach which allows the evaluation of the throughput performance of various versions of TCP. The main findings of this study are that 1) error correlations significantly affect the performance of TCP, and in particular may result in considerably better performance for Tahoe and NewReno; and 2) over slowly fading channels which are characterized by significant channel memory, Tahoe performs as well as NewReno. This leads us to conclude that a clever design of the lower layers that preserve error correlations, naturally present on wireless links because of the fading behavior, could be an attractive alternative to the development or the use of more complex versions of TCP.

System and method for monitoring and optimizing network performance with user datagram protocol network performance information packets

Abstract: A method for monitoring and optimizing performance of a packet network utilizing user datagram protocol including determining network performance information between a sending user datagram protocol network element and a receiving user datagram protocol network element; concatenating the network performance information into data packets; transmitting the data packets to at least one non-user datagram protocol network device having transport layer capabilities; and controlling bandwidth of the at least one non-user datagram protocol network device to optimize the network utilizing the user datagram protocol based on the network performance information in the data packets. A system for monitoring and optimizing performance of a packet network utilizing user datagram protocol is also included.