Performance measurement system for large computer network
First Claim
Patent Images
1. A method for measuring the performance of at least a portion of a computer network, the method comprising the steps of:
- providing one or more sampling lists, each sampling list identifying one or more nodes representative of a group of nodes of interest;
sending signals to a relatively large number of nodes identified on the one or more sampling lists in a relatively short time;
receiving signals from responding nodes on the one or more sampling lists;
recording raw data indicative of the corresponding sent and received signals;
reducing the raw data to data streams indicative of the performance of the nodes; and
presenting information derived from the data streams, the information being indicative of the performance of one or more of the groups of interest.
26 Assignments
0 Petitions
Accused Products
Abstract
A method for measuring the performance of a large computer network is defined. The method provides a method of defining and characterizing the population of interest and interesting sub-populations; a method of providing a sampling list of nodes that represent the population; measuring the performance of destinations in the sampling list; recording raw data therefrom; reducing the raw data and providing presentations of information indicative of one or more of the sub-populations in ways that address questions of interest.
-
Citations
47 Claims
-
1. A method for measuring the performance of at least a portion of a computer network, the method comprising the steps of:
-
providing one or more sampling lists, each sampling list identifying one or more nodes representative of a group of nodes of interest;
sending signals to a relatively large number of nodes identified on the one or more sampling lists in a relatively short time;
receiving signals from responding nodes on the one or more sampling lists;
recording raw data indicative of the corresponding sent and received signals;
reducing the raw data to data streams indicative of the performance of the nodes; and
presenting information derived from the data streams, the information being indicative of the performance of one or more of the groups of interest. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 19, 20, 21, 22, 23, 28, 29, 30, 31)
-
-
18. A method for measuring the performance of a computer network, the method comprising the steps of:
-
providing a survey of network nodes;
creating at least one sampling list representative of at least a portion of the nodes;
performing a process comprising the steps of, running a sending procedure which sends signals from a beacon to a plurality of nodes on the sampling list, the signals requesting responses from the plurality of nodes, simultaneously running a receiving procedure which receives return signals from responding nodes, such that the signal being received at any particular time could correspond with any of the signals sent to one of the plurality of nodes, and simultaneously running a recording procedure which records data characterizing the corresponding sent and received signals;
performing data reduction on the data to create data streams comprising information indicative of the performance of at least a portion of the network; and
presenting information derived from the data streams to a user such that the user is able to draw conclusions about the performance of at least a portion of the network.
-
-
24. A method of creating a sampling list for measuring performance of a group of nodes of interest on a large computer network, the method comprising the steps of:
-
obtaining a population survey of the nodes on the network;
gathering data regarding topology of the network;
creating a list of nodes in the group of nodes of interest based on the gathered data;
trimming the list of nodes by excluding from the list those nodes which are not important to the performance of other nodes;
testing the remaining nodes on the list of nodes to determine reachability of each node;
removing unreachable nodes from the list;
categorizing the nodes on the list according to the service type the node provides;
excluding nodes on the list that fall within over-represented service types for the particular emphasis of the group of nodes of interest.
-
-
25. A method of correcting data collected about one or more nodes remotely located from a data collection point, the method comprising the steps of:
-
providing data comprising performance metrics, the data having been collected by a beacon at the collection point by sending signals to and receiving signals from one or more nodes;
determining a great circle distance over which the signals traveled between the collection point and one or more nodes;
calculating the time required for the signal to traverse the distance using the speed of light within the link and the determined distance for one or more nodes to find a geographic distance factor; and
subtracting the appropriate geographic distance factor from the metric values within the data to provide corrected data.
-
-
26. A method of correcting new data collected about one or more nodes remotely located from a beacon, the method comprising the steps of:
-
providing pre-collected data, the pre-collected data having been collected by the beacon by sending signals to and receiving signals from one or more nodes, and the data comprising performance metrics for a predetermined length of time;
determining a baseline metric value for the pre-collected data to find a correction factor;
subtracting the correction factor from the metric values within the new data to provide corrected data.
-
-
27. A general method for evaluating performance data of a computer network system, comprising:
-
receiving performance data measured from a computer network system;
selecting an appropriate event detection method to evaluate the performance data, the selected event detection method adapted to compare the performance data to threshold parameters to determine whether an event has occurred;
sending a signal indicating an event has occurred whenever an event is detected by the appropriate event detection method.
-
-
32. A method of combining data streams comprising:
-
providing two or more input data streams, the data streams representing data collected over a time interval, and each data stream comprising one or more records, each record including a start time indicating the time at which a collection process began collecting the data within the record and a value for one or more metrics;
defining sequential time slices, each time slice comprising a piece of the interval and being the duration of the interval divided by a frequency;
sorting the records from the two or more data streams by the start time;
allocating each record into the time slice in which the start time for each record falls;
calculating one or more statistics for each time slice, each statistic representing the value of one or more metrics for the records allocated to each time slice; and
outputting a data stream comprising records which include the calculated statistics. - View Dependent Claims (33, 34, 35, 37, 39)
-
-
36. A method of selectively notifying interested users of a performance event, the method comprising the steps of:
-
receiving a signal indicating that an event has been detected and the type and magnitude of the event;
composing an email message indicating that an event has been detected and the type and magnitude of the event;
comparing the indications to records within a file comprising conditions in which users desire to be notified of performance events;
automatically sending the email to users whose records included a condition met by the indications.
-
-
38. A method of displaying performance characteristics of at least a portion of a network, the method comprising the steps of:
-
providing values for metrics indicative of the performance of at least a portion of the network; and
producing an image including cells organized in columns and one or more rows, each column representing a performance metric, each row representing at least a portion of the network, each cell including values for a metric as measured over one or more time intervals.
-
-
40. A method of displaying performance characteristics of at least a portion of a network, the method comprising the steps of:
-
providing values for metrics indicative of the performance of at least a portion of the network;
producing an image including a time versus metric graph indicative of the performance of at least a portion of the network;
drawing a shaded region representing a detected performance event on the graph. - View Dependent Claims (41)
-
-
42. A method of collecting data from a large number of nodes on a computer network in a short time, the method comprising:
-
providing a sampling list comprising information indicative of the addresses of nodes about which data is to be collected; and
performing a process comprising the steps of, running a sending procedure which sends signals from a beacon to a plurality of nodes on the sampling list, the signals requesting responses from the plurality of nodes, simultaneously running a receiving procedure which receives return signals from responding nodes, such that the signal being received at any particular time could correspond with any of the signals sent to one of the plurality of nodes, and simultaneously running a recording procedure which records data characterizing the corresponding sent and received signals. - View Dependent Claims (43, 44, 45, 47)
-
-
46. A method of collecting topological information about a large number of nodes on a computer network in a short time, the method comprising:
-
providing a sampling list comprising information indicative of the addresses of example nodes; and
performing a process comprising the steps of, running a sending procedure which sequentially sends a predetermined number of signals from a beacon toward each of a plurality of nodes on the sampling list, the sequentially sent signals being sequentially encoded with a time to live parameter, simultaneously running a receiving procedure which receives return signals from responding hops along the route to each destination, the hops reporting an expired time to live, such that the signal being received at any particular time could correspond with any of the signals sent toward the plurality of nodes, and simultaneously running a recording procedure which records data characterizing the corresponding sent and received signals.
-
Specification