Centralized network control
A method and system of network control is disclosed. The method and system includes collecting real-time operation information on one or more first network elements of a network, selecting a policy to be implemented by at least one second network element, different from the first network element, responsive to the collected real time information from the one or more first network elements, the at least one second element including an end-point element of the network and enforcing the selected policy on an agent hosted by the at least one second network element.
1 . A method for controlling a network, said network comprising:
a first set of two or more network elements and a second set of one or more network elements;
wherein two or more of said network elements of said first set and one or more of said network elements of second set comprise an end point element of said network which hosts an agent; and
a policy controller;
said method comprising:
collecting real-time operational information relating to attribute values at two or more agents from said first set of two or more network elements which host said agents;
receiving said real-time operational information relating to said attribute values at said policy controller from said two or more agents from said first set;
selecting a policy, by said policy controller, without human intervention, based on said real time information relating to said attribute values to be implemented by at least one of said second set of network elements hosting said at least one or more agents; and
enforcing said selected policy on said at least one of said second set of network elements through said agent hosted thereby.
2 . A method according to claim 1 , wherein collecting real-time operation information comprises collecting information on operation problems.
3 . A method according to claim 2 , wherein collecting real-time operation information comprises collecting information on applications that do not operate or operate slowly.
4 . A method according to claim 2 , wherein collecting real-time operation information comprises collecting information on system or application crashes.
5 . A method according to claim 1 , wherein collecting real-time operation information comprises collecting information on software applications installed or running on the network elements.
6 . A method according to claim 1 , wherein collecting real-time operation information comprises collecting information on the communications between elements of the network.
7 . A method according to claim 1 , wherein selecting the policy to be implemented comprises selecting a policy relating to a software to be installed on the second network element.
8 . A method according to claim 1 , wherein selecting the policy to be implemented comprises selecting a policy relating to a software to be uninstalled from the second network element.
9 . A method according to claim 1 , wherein selecting the policy to be implemented comprises selecting a policy relating to preventing installation of a software on the second network element.
10 . A method according to claim 1 , wherein selecting the policy to be implemented comprises selecting responsive to a determination that a group of network elements having a common problem have installed thereon a specific software application or combination of software applications.
11 . A method according to claim 1 , wherein selecting the policy to be implemented comprises selecting a policy relating to allocation of network resources.
12 . A method according to claim 1 , wherein the policy is selected within less than 60 minutes from the collecting of the information.
13 . A method according to claim 1 , wherein collecting the operation information is performed repeatedly.
14 . A method according to claim 1 , wherein the processor is adapted to find, for a group of network elements having a problem, a combination of attribute values that correlate with the problem to at least a predetermined degree.
15 . A method according to claim 1 , wherein the processor is adapted to collect for at least one network element, a plurality of snapshot records of the network element at different times.
16 . A method according to claim 15 which includes referring to one or more snapshots to provide an additional amount of data which can be analyzed to determine correlations between problems and cause attribute values.
17 . A system for controlling a network, said network comprising:
a first set of two or more network elements and a second set of one or more network elements, wherein each network element of said first set and said second set comprises an end point element of said network which hosts an agent;
wherein said agents hosted by said first set of two or more network elements are configured to collect real-time operational information relating to attribute values;
wherein two or more of said agents hosted by said first set of two or more network elements are configured to enforce a selected policy, said enforcing comprising:
repeatedly collecting, by at least two or more of said agents of said first set, updated real-time operational information relating to the attribute values,
receiving instructions for performing the enforcement of a selected policy, and
executing said instructions for performing the enforcement of said selected policy; and
a policy controller,
wherein said policy controller is configured to:
receive said real-time operational information,
select said selected policy, without human intervention, based on said real time information relating to said attribute values to be implemented by said agents hosted by said second set of one or more network elements
transmit, to each agent hosted by network elements in said second set, said instructions for performing the enforcement of said selected policy.
18 . A system according to claim 17 , wherein the processor is adapted to find, for a group of network elements having a problem, a combination of attribute values that correlate with the problem to at least a predetermined degree.
19 . A system according to claim 17 , wherein the processor is adapted to find, for a group of network elements having a problem, a combination of attribute values that appears only on the network elements having the problem.
20 . A system according to claim 17 , wherein the processor is adapted to collect for at least one network element, a plurality of snapshot records of the network element at different times.
21 . A system according to claim 17 including referring to one or more snapshots to provide an additional amount of data which can be analyzed to determine correlations between problems and cause attribute values
22 . A system according to claim 17 , wherein the processor is adapted to verify that each network element belongs to the network before collecting information from the network element.
23 . A system according to claim 17 , wherein the processor is adapted to find groups using a k-clustering or hierarchy clustering method.