NETWORK EVALUATION GRID TECHNIQUES
First Claim
Patent Images
1. A method implemented and residing in a non-transitory computer-readable storage medium, comprising:
- identifying grid services that are distributed over a network for processing on a plurality of devices, wherein the plurality of devices combine to form a grid computing environment and selective collections of the devices combine to form a particular grid of the grid computing environment, each grid service listens for data packets on its grid; and
configuring and distributing a watch grid service with each grid of the grid computing environment, and when initiated on a particular device of the particular grid, a particular watch grid service listens for and reports on traffic in response to enforcement of particular ones of traffic pattern specifications, and when the particular watch grid service is loaded and cannot capture all the traffic another watch grid service cooperates to capture packets associated with the traffic that was missed.
2 Assignments
0 Petitions
Accused Products
Abstract
Network evaluation grid techniques are presented. Local specifications for traffic patterns are evaluated for local traffic on local grids of a grid computing environment and reported out. The local traffic reported out is then evaluated in response to global traffic pattern specifications and a global traffic policy in order to identify one or more remedial actions to take or to recommend in response to that evaluation.
-
Citations
20 Claims
-
1. A method implemented and residing in a non-transitory computer-readable storage medium, comprising:
-
identifying grid services that are distributed over a network for processing on a plurality of devices, wherein the plurality of devices combine to form a grid computing environment and selective collections of the devices combine to form a particular grid of the grid computing environment, each grid service listens for data packets on its grid; and configuring and distributing a watch grid service with each grid of the grid computing environment, and when initiated on a particular device of the particular grid, a particular watch grid service listens for and reports on traffic in response to enforcement of particular ones of traffic pattern specifications, and when the particular watch grid service is loaded and cannot capture all the traffic another watch grid service cooperates to capture packets associated with the traffic that was missed. - View Dependent Claims (2, 3, 5, 6, 7)
-
-
4. The method of claim 4, wherein configuring and distributing further includes determining traffic pattern policies for one or more of the following:
- selective groupings of the traffic pattern specifications and the grid computing environment as a whole.
-
8. A method implemented and residing in a non-transitory computer-readable storage medium for executing on a processing device, comprising:
-
receiving, by the processing device, messages from multiple watch grid services, wherein each watch grid service reports network traffic occurring on grid members associated with a grid computing environment in response to local traffic pattern specifications and local traffic pattern policies; using, by the processing device, global traffic patterns to detect selective traffic being reported, each of the global traffic patterns identifies a particular grid member and a subnet for traffic emanating from that particular grid member to be monitored independent of resources producing that traffic; enforcing, by the processing device, a global traffic policy to correlate the selective traffic; and identifying, by the processing device, actions to take in response to enforcement of the global traffic policy. - View Dependent Claims (9, 10, 11, 12, 13, 14)
-
-
15. A system, comprising:
-
a global grid manager service implemented in a non-transitory computer-readable storage medium and to process on a machine within a network; watch grid services implemented in a non-transitory computer-readable storage medium, each grid service processes on a different machine of the network; and an evaluation grid service implemented in a non-transitory computer-readable storage medium and to process on a machine different from those machines used by the watch grid services, wherein the global grid manager service is to configure each of the watch grid services with local traffic pattern specifications and local traffic pattern policies for detecting local traffic on local machines or local processing environments of the watch grid services, each watch grid service is to report all the detected traffic to the evaluation grid service that emanates from a network interface card of locale machines associated with that watch grid service, and wherein the evaluation grid service is to use global traffic pattern specifications and enforce a global traffic pattern policy to correlate the detected traffic and take one or more actions in response to the correlated traffic. - View Dependent Claims (16, 17, 18, 19, 20)
-
Specification