摘要:
A system for monitoring traffic on a network first discovers the network so as to map the various devices and links in the network. Statistics are then gathered from various points in the network relating to quality of service, and especially loads on the network devices. Synthetic calls are generated at selected points of the network while monitoring the network. This data is then stored and displayed in a manner that is easy for the operator to analyze, with more detailed displays being available through the use of a mouse or keystrokes.
摘要:
Techniques are disclosed for improved monitoring and analysis of VoIP communications, multimedia communications or other types of network traffic in a network-based communication system. In accordance with one aspect of the invention, an endpoint device of the system includes an integrated call synthesis capability. More particularly, the endpoint device is configurable into at least a first mode of operation in which the device responds to one or more commands from a testing server of the system directing the device to participate in a synthetic call or other test communication within the system, and a second mode of operation in which the device is utilizable by a system user to establish an actual communication. Other aspects of the invention relating to kernel timestamping and timestamp post-processing provide improved accuracy for measurements such as jitter, loss and delay.
摘要:
A system discovers the topology of devices in a network, such as all of the layer-2 and layer-3 devices, including routers and switches. The devices are mapped, as well as the links that join them. The devices are polled using SNMP requests. Route tables are derived from each router to determine addresses of the devices. A spanning tree process is used to discover layer-2 devices. Connections between layers are then discovered. Where information is incomplete, an undiscovered router cloud is recognized.
摘要:
A system visualizes and generates reports from stored data, concerning various measured parameters related to a network carrying call traffic. The network is displayed in an easily recognized arrangement, with various parameters being indicated by different shapes and colors for each device on the network. The parameters are changed to indicate the status of the network at different times. More detailed information about a parameter, a device or a path between devices may be obtained simply using a mouse or keystrokes. A variety of different types of detailed displays allows the operator to troubleshoot problems in the network.
摘要:
A network monitoring and analysis system is configured to include a root cause analysis function that permits the automatic determination of performance problem locations in a network of a communication system comprising a plurality of endpoint devices. Test communications are generated in the system in accordance with a selected pattern, which may be based on a flow matrix, and end-to-end path measurement data is collected utilizing the generated test communications. The end-to-end path measurement data is transformed to produce a plurality of performance indicators comprising a performance indicator for each of a plurality of non-end-to-end paths defined at least in part by the selected pattern.
摘要:
A network monitoring and analysis system is configured to include a root cause analysis function that permits the automatic determination of performance problem locations in a network of a communication system comprising a plurality of endpoint devices. Test communications are generated in the system in accordance with a selected pattern, which may be based on a flow matrix, and end-to-end path measurement data is collected utilizing the generated test communications. The end-to-end path measurement data is transformed to produce a plurality of performance indicators comprising a performance indicator for each of a plurality of non-end-to-end paths defined at least in part by the selected pattern.
摘要:
Techniques are disclosed for improved monitoring and analysis of VoIP communications, multimedia communications or other types of network traffic in a network-based communication system. In accordance with one aspect of the invention, endpoint devices of the network-based communication system are configurable so as to collectively implement a distributed monitoring and analysis system which does not require a centralized testing server or other centralized controller. Distributed test units associated with the endpoint devices may be utilized in implementing the distributed monitoring and analysis system, and are preferably configured to support a web-based user interface providing access to measurement data. The endpoint devices may be advantageously organized into a hierarchy comprising a plurality of zones, with each of the endpoint devices belonging to at least one zone. For each zone, one of the endpoint devices may be designated as a zone leader for controlling the periodic generation of communications between selected endpoint devices that belong to subzones of that zone in the hierarchy.
摘要:
A method is disclosed for adaptively and intelligently partitioning traffic among a plurality of communications channels, without injecting probe traffic into any of the channels. In accordance with the illustrative embodiment, a source node transmits traffic to a receiving node via K communications channels in accordance with partition coefficients α1, . . . , αK, and the source node receives information regarding the quality of service (QoS) provided by each of the communications channels. The source node adjusts the values of the partition coefficients α1, . . . , αK adaptively based on: the current channel QoS information, prior channel QoS information, the prior values of α1, . . . , αK, and a measure of channel independence for one or more pairs of communications channels.
摘要:
A technique is disclosed that evaluates a network path between (i) a first node in a first subnetwork of endpoint nodes, such as IP phones, and (ii) a second node in a second subnetwork. A “ricochet” node in the network path evaluates the path by probing one or both subnetworks, where the ricochet node acts as relay for traffic packets being transmitted between the two subnetworks. A given relay has only to probe a single, representative node within a subnetwork at any given time in order to obtain performance data that is representative of the subnetwork overall. By probing the representative node, the relay is able to acquire an assessment of network conditions that is valid for the path between the relay and any endpoint in the subnetwork. As a result, the disclosed technique reduces the probing overhead when many endpoint nodes on a given subnetwork are simultaneously active and experiencing adverse network conditions.
摘要:
A method is disclosed for adaptively and intelligently partitioning traffic among a plurality of communications channels, without injecting probe traffic into any of the channels. In accordance with the illustrative embodiment, a source node transmits traffic to a receiving node via K communications channels in accordance with partition coefficients α1, . . . , αK, and the source node receives information regarding the quality of service (QoS) provided by each of the communications channels. The source node adjusts the values of the partition coefficients α1, . . . , αK adaptively based on: the current channel QoS information, prior channel QoS information, the prior values of α1, . . . , αK, and a measure of channel independence for one or more pairs of communications channels.