Search

In this paper, we describe and experimentally evaluate the performance of our Network Link Outlier Factor (NLOF) for detecting soft-failures in communication networks. The NLOF is computed using the throughput values derived from NetFlow records. The flow throughput values are clustered in two stages, outlier values are determined within each cluster, and the flow outliers are used to compute the outlier factor or score for each network link. When sampling NetFlow records across the full span of a network, NLOF enables the detection of soft-failures across the span of the network; large NLOF scores correlate well with links experiencing failure.

We conducted an extensive set of experiments with an offloading testbed to understand the impact that data marshalling techniques have on computation offloading decisions. We find that the popular JSON format to marshall data between client and server comes at a significant computational expense compared to a minimalistic raw data transfer. The computational time is significant in that it affects computation offloading decisions in a variety of conditions. We outline some of these conditions., Comment: 6 pages, 5 figures

We analyze the conditions in which offloading computation reduces completion time. We extend the existing literature by deriving an inequality (Eq. 4) that relates computation offloading system parameters to the bits per instruction ratio of a computational job. This ratio is the inverse of the arithmetic intensity. We then discuss how this inequality can be used to determine the computations that can benefit from offloading as well as the computation offloading systems required to make offloading beneficial for particular computations.

The emerging Software Defined Networking (SDN) paradigm separates the data plane from the control plane and centralizes network control in an SDN controller. Applications interact with controllers to implement network services, such as network transport with Quality of Service (QoS). SDN facilitates the virtualization of network functions so that multiple virtual networks can operate over a given installed physical network infrastructure. Due to the specific characteristics of optical (photonic) communication components and the high optical transmission capacities, SDN based optical networking poses particular challenges, but holds also great potential. In this article, we comprehensively survey studies that examine the SDN paradigm in optical networks; in brief, we survey the area of Software Defined Optical Networks (SDONs). We mainly organize the SDON studies into studies focused on the infrastructure layer, the control layer, and the application layer. Moreover, we cover SDON studies focused on network virtualization, as well as SDON studies focused on the orchestration of multilayer and multidomain networking. Based on the survey, we identify open challenges for SDONs and outline future directions.

In a hybrid PON/xDSL access network, multiple Customer Premise Equipment (CPE) nodes connect over individual Digital Subscriber Lines (DSLs) to a drop-point device. The drop-point device, which is typically reverse powered from the customer, is co-located with an Optical Network Unit (ONU) of the Passive Optical Network (PON). We demonstrate that the drop-point experiences very high buffer occupancies when no flow control or standard Ethernet PAUSE frame flow control is employed. In order to reduce the buffer occupancies in the drop-point, we introduce two gated flow control protocols that extend the polling-based PON medium access control to the DSL segments between the CPEs and the ONUs. We analyze the timing of the gated flow control mechanisms to specify the latest possible time instant when CPEs can start the DSL upstream transmissions so that the ONU can forward the upstream transmissions at the full PON upstream transmission bit rate. Through extensive simulations for a wide range of bursty traffic models, we find that the gated flow control mechanisms, specifically, the ONU and CPE grant sizing policies, enable effective control of the maximum drop-point buffer occupancies.

Dynamic circuits are well suited for applications that require predictable service with a constant bit rate for a prescribed period of time, such as cloud computing and e-science applications. Past research on upstream transmission in passive optical networks (PONs) has mainly considered packet-switched traffic and has focused on optimizing packet-level performance metrics, such as reducing mean delay. This study proposes and evaluates a dynamic circuit and packet PON (DyCaPPON) that provides dynamic circuits along with packet-switched service. DyCaPPON provides $(i)$ flexible packet-switched service through dynamic bandwidth allocation in periodic polling cycles, and $(ii)$ consistent circuit service by allocating each active circuit a fixed-duration upstream transmission window during each fixed-duration polling cycle. We analyze circuit-level performance metrics, including the blocking probability of dynamic circuit requests in DyCaPPON through a stochastic knapsack-based analysis. Through this analysis we also determine the bandwidth occupied by admitted circuits. The remaining bandwidth is available for packet traffic and we conduct an approximate analysis of the resulting mean delay of packet traffic. Through extensive numerical evaluations and verifying simulations we demonstrate the circuit blocking and packet delay trade-offs in DyCaPPON.

A wide array of dynamic bandwidth allocation (DBA) mechanisms have recently been proposed for improving bandwidth utilization and reducing idle times and packets delays in passive optical networks (PONs). The DBA evaluation studies commonly assumed that the report message for communicating the bandwidth demands of the distributed optical network units (ONUs) to the central optical line terminal (OLT) is scheduled for the end of an ONU's upstream transmission, after the ONU's payload data transmissions. In this article, we conduct a detailed investigation of the impact of the report message scheduling (RMS), either at the beginning (i.e., before the pay load data) or the end of an ONU upstream transmission on PON performance. We analytically characterize the reduction in channel idle time with reporting at the beginning of an upstream transmission compared to reporting at the end. Our extensive simulation experiments consider both the Ethernet Passive Optical Networking (EPON) standard and the Gigabit PON (GPON) standard. We find that for DBAs with offline sizing and scheduling of ONU upstream transmission grants at the end of a polling cycle, which processes requests from all ONUs, reporting at the beginning gives substantial reductions of mean packet delay at high loads. For high-performing DBAs with online grant sizing and scheduling, which immediately processes individual ONU requests, or interleaving of ONUs groups, both reporting at the beginning or end give essentially the same average packet delays.