US20160212167A1 - Classification of security policies across multiple security products - Google Patents

Abstract

A management entity connects with multiple security devices across a network. Each security device operates in accordance with one or more security policies. The management entity imports, over the network, data describing the security policies from the multiple security devices. The management entity classifies the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.

Description

TECHNICAL FIELD
• The present disclosure relates to security systems for networks, applications, content, and authentication.
BACKGROUND
• Administering and maintaining security of an enterprise network is critical task to ensure business efficiency and productivity. Security involves numerous tasks, including (without limitation) monitoring for unauthorized operations (intrusions, external accesses, application security, content security, authentication compliance, etc.) which can, among other things, put sensitive data at risk. This can be complicated by the fact that the enterprise network may span numerous geographical regions, nationally an internationally.
• In a typical enterprise network, there are numerous security devices of various types as well as numerous management applications. This can make enforcement of requirements challenging. Each device or type of device has its own set of complex policy definitions. A network administrator needs to be an expert on numerous security products in order to define policies and maintain security.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a diagram of a system including a cloud-based management entity that manages a plurality of security devices in a customer datacenter, according to an example embodiment.
• FIG. 2 is a flow chart depicting a method for managing a plurality of security devices in a customer datacenter, according to an example embodiment.
• FIGS. 3-8 are diagrams illustrating operations of the various steps in the flow chart ofFIG. 2.
• FIG. 9 is a block diagram illustrating an example application of the system ofFIG. 1, according to an example embodiment.
• FIG. 10 is a block diagram illustrating another example application of the system ofFIG. 1 involving use of collective security intelligence, according to an example embodiment.
• FIGS. 11, 12A and 12B are diagrams illustrating a user interface through monitored security activity is presented to a user and from which a user may invoke policy changes across multiple security devices, according to an example embodiment.
• FIG. 13 is a block diagram illustrating a hardware configuration for the cloud-based management entity, according to an example embodiment.
• FIG. 14 is an illustration of a classification operation performed by the management entity, according to an example embodiment.
• FIGS. 15A and 15B collectively represent a flowchart of an overview method of importing and classifying network policies from different security devices, according to an example embodiment.
• FIG. 16 is an illustration of user interface screen displayed by the management entity and through which a user may identify/select, connect to, and then import security policies from one or more security devices, according to an example embodiment.
• FIG. 17 is an illustration of a user interface screen that may used to initiate policy classification of imported security policies and also displaying resulting security policy classifications, according to an example embodiment.
• FIG. 18 is an illustration of a user interface screen that shows an expanded view of a policy classification after an expand icon associated with policy classification has been selected by a user, according to an example embodiment.
• FIG. 19 is an illustration of an example access list in extended rule format in which a “name” referenced by the rule has been resolved to a bound interface, according to an example embodiment.
• FIG. 20A is a flowchart of a method of determining commonality/similarity between different security policies based on matching points of comparison (i.e., corresponding feature parameters of the network security policies), according to an example embodiment.
• FIG. 20B is a flowchart of a high-level method of classifying security policies, according to an example embodiment.
• FIG. 21 illustrates a depiction of a security policy model that follows an “if {Principal} tries to perform an {Action} on {Resource} within {Context} then {Result}” format, referred to herein as a “PARCR” model, according to an example embodiment.
• FIG. 22 is a block diagram of a policy unification module that may be implemented in the management entity to convert or map between native security policies expressed according to native policy models and normalized or generic policies expressed according to the PARCR model, according to an example embodiment.
• FIG. 23 is an illustration of a PARCR mapping performed in part by security device plug-ins of the policy unification module, according to an example embodiment.
• FIG. 24 is an illustration of a PARCR policy model anatomy, according to an example embodiment.
• FIG. 25 is a policy model bridge that may be used to map a simplified Web Security Appliance (WSA) access policy to the PARCR model, according to an example embodiment.
• FIG. 26 is a flowchart of a method of converting between normalized security policies based on the PARCR model and native security policies based on native security policy models, according to an example embodiment.
• FIG. 27 is an illustration of a user interface screen displayed by the management entity and through which the user has entered a policy sub-class name “Branch Allow Web Traffic” into a sub-class name field/option, and which also allows for selection of identical network security policy (NSP) classifications, according to an example embodiment.
• FIG. 28 is an illustration of a user interface screen that may be used to create a security policy template (referred to as a “policy template”) based on previously created policy sub-classes or sub-templates, according to an example embodiment.
• FIG. 29 is an illustration of a network environment including existing geographically distributed data centers or branch sites (referred to as “branches”) operating under control of the management entity, according to an example embodiment.
• FIG. 30 is an illustration of a user interface screen displayed by the management entity that may be used to create a new security policy based on one or more previously created policy templates, according to an example embodiment.
• FIG. 31 is an illustration of a user interface screen displayed by the management system in response to a user selection of a “Validate” icon in the user interface screen ofFIG. 30, according to an example embodiment.
• FIG. 32 is an illustration of a user interface screen displayed by the management system after a user has selected an “add rule” icon and also inserted a new rule via the user interface screen to be added to a new branch in the network environment shown inFIG. 29, according to an example embodiment.
• FIG. 33 is a flowchart of an example method of creating and deploying security policies based on policy templates, performed by the management entity through interactions with a user, according to an example embodiment.
• FIG. 34 is a diagram illustrating a user interface screen through which a user may view security topology, according to an example embodiment.
• FIG. 35 is a diagram illustrating a user interface screen through which a user may view security topology overlaid on a geographical map, according to an example embodiment.
• FIG. 36 is a diagram illustrating another user interface screen which a user may view security topology and invoke security policy changes, according to an example embodiment.
• FIG. 37 is a diagram illustrating still another user interface screen through which a user may invoke security policy changes, according to an example embodiment.
• FIG. 38 is a flow chart of a process for performing user interface operations in connection with the user interface screen ofFIG. 37, according to an example embodiment.
DESCRIPTION OF EXAMPLE EMBODIMENTSOverview
• A management entity connects with multiple security devices across a network. Each security device operates in accordance with one or more security policies. The management entity imports, over the network, data describing the security policies from the multiple security devices. The management entity classifies the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.
Example Embodiments
• Presented herein are a system and methods which simplify, unify and enhance policy management for network security products. A centralized management entity, that may take the form of a cloud-based application, communicates with the network security products (e.g., devices, applications, etc., but generally referred to as “security devices” herein) in a given network environment (e.g., a customer datacenter or customer enterprise network, etc.). The system can perform analytics and obey shared best practices to provide enhanced insight into network security threats in order to make prompt mitigation. The system presented herein can achieve real-time integration between threats and policy enforcement in a way not heretofore possible.
• Examples of network security devices/products that may be integrated into the system presented herein include, but are not limited to including, firewalls, intrusion prevention systems (IPSs), Next Generation firewalls (NGFWs), Next Generation IPSs (NGIPSs), web security appliances (WSAs), identity services engines (ISEs), application security appliances (ASAs), cloud web security (CWS) products, security manager products, content security management appliances, cloud firewalls, intrusion detection systems (IDSs), etc.
• As used herein, a security policy is a set of (one or more) rules that governs what is and what is not allowed through security devices/products. Security policies include network security policies, application security policies, and authentication security policies. A policy typically includes multiple attributes, such as source, destination, applications, port, universal resource locator (URL), on which to take a network security operation or action (e.g., permit or deny). The embodiments presented below are directed to network security policies for illustrative purposes only, and may be used for application security policies and authentication security polices, as would be appreciated by one of ordinary skill in the relevant arts with access to present description.
• Thus, the term “network security device” as used herein is not meant to be limited to network devices, and may include other security devices and applications, such as application security, content security, authentication, etc. Thus, more generally these devices are referred to as “security devices” and are meant to include physical devices as well as applications or tools. Likewise, the term “network security policy” is not limited to network policies, and may include other security policies, such as application security policies, content security policies, authentication policies, etc., and thus, more generally these policies are referred to as “security policies.”
• A business policy is typically a statement in writing of how a company plans to protect the company's physical and information technology (IT) assets. A role of a security architect or security operator is to apply the business policy into enforceable security policy(ies), monitor the enforcement, and make changes as needed.
Overall System
• Referring now toFIG. 1, a cloud-basedmanagement system100 is shown that connects to and communicates with network security devices in a customer datacenter.FIG. 1 shows the details of customer datacenter120(1), but it should be understood that the cloud-basedmanagement system100 may connect and communicate with multiple customer datacenters120(1)-120(N) as shown inFIG. 1.
• The cloud-basedmanagement system100 includes amanagement entity110 that consists of one or more computer servers112(1)-112(M) that execute software to perform the operations described throughout this disclosure. An example of a hardware configuration for themanagement entity110 is described in more detail below in connection withFIG. 13.
• The customer datacenters120(1)-120(N) each includes a plurality of network security devices or products, shown at reference numerals130(1)-130(P). Within a customer datacenter there are one ormore resources140 and one ormore actors150. Theresources140 may include servers, databases, and theactors150 are users or processes using a computing device (personal computer, SmartPhone, etc.) that may seek access to one or more of theresources140. The resources and actors may also reside outside the customer datacenter itself, e.g., in the Internet. The network security devices130(1)-130(P) control access of theactors150 to theresources140 according to policies, i.e., a set of one or more rules configured on the respective network security devices.
• As explained above, the network security devices130(1)-130(P) may be of different types from the same or different vendors of network security products. Themanagement entity110 centralizes and unifies the management of network security policies across the plurality of network security devices130(1)-130(P) to greatly simplify network security management in a customer datacenter.
• Turning now toFIG. 2, a flow chart is shown for aprocess200 according to an example embodiment. Thisprocess200 is described in connection withFIGS. 3-8. Theprocess200 begins atstep210 where a customer (e.g., a business or enterprise) is on-boarded to a cloud-based management system. This on-board operation is shown inFIG. 3. This involves a network administrator shown atreference numeral300 logging on to a log-onweb page310 served by one of the servers112(1)-112(M) of themanagement entity110. The log-onweb page310 allows the network administrator to set up privileges to permit the management entity to communicate, over the Internet, into the customer datacenter120(1) in order to connect to the network security devices130(1)-130(P). In addition, during the initial log-in and setup phase, the network administrator provides names and address (e.g., Internet Protocol (IP) addresses) for each of the network security devices in the customer datacenter. Other types of set-up processes may be used other than use of a log-on web page.
• Next, atstep220, themanagement entity110 discovers the network security devices and imports the policies from each network security device. This operation is depicted inFIG. 4. The discovery step is described in more detail hereinafter. Briefly, this involves sending a connection string and device type tag to each network security device. Each network security device responds with device descriptor and policy data for each network security rule configured on the respective network security device. This data is shown at reference numerals400(1)-400(P) from network security devices130(1)-130(P), respectively. An example of the policy data imported form a security device may be:
• • Protocol: HTTP S
  • Network: All
  • Destination: 132.180.0.0/24
  • Description: Web
  • Policy: On
  • Logging: On
• Themanagement entity110 stores the discovered data describing the discovered security devices and their native policies, as shown atreference numeral410. Each native network security policy may be one or more native network security rules associated with a named network security device and formatted according to a corresponding native policy model for a network security device. Each native network security rule may in turn include a set of rule parameters to permit or deny network access for the named network security device based on a network protocol, source and destination addresses, and a device port.
• Next, atstep230, and as an optional step, the imported policies are classified. Specifically, the imported policies are compared against each other to determine whether they can be grouped into one of several categories. These categories include, but are not limited to including: (1) identical, (2) similar, (3) unique and (4) further investigation required. Theclassification step230 is described in further detail below in under the heading “Security Policy Classification.” Generally speaking, classifying may involve classifying imported native network security policies into network security policy classifications each including one or more of the imported native network security policies based on commonality between security rules included in the native network security policies across the multiple network security devices. Thus, classifying may involve automatically classifying the network security policies into the classifications based on commonality between the network security rules across the named devices associated with the network security policies.
• Next, atstep240, data describing the native network security policies received from each of the network security devices is normalized in accordance with a generic policy model to produce normalized policy data. The normalized policy data is shown atreference numeral500 inFIG. 5. Corresponding normalized policy data510(1)-510(P) is sent to each of the network security devices, and as explained further hereinafter in the section under the heading “Security Policy Unification”, a translation is made from the normalizedpolicy data500 in accordance with the generic policy model to rules in accordance with the native rule set for the respective network security devices. Generally, each native network security policy imported from a network security device may be a set of one or more native network security rules, each native network security rule including native rule parameters expressed according to the corresponding native model. The imported native network security policies are normalized by, for each imported native network security policy, mapping the native rule parameters expressed according to the corresponding native model to corresponding components of a generic rule defined according to the generic policy model. The mapping may include mapping the native rule parameters to the corresponding components {a principal or actor}, {action}, {a resource}, {a context}, and {perform a result} as used in the generic rule: “if {a principal or actor} tries to perform an {action) on {a resource} within {a context} then {perform a result}.”
• Next, atstep250, additional network security devices may be added and in so doing a network policy template may be created that is deployed to any additional network devices.FIG. 6 illustrates the additional network security device at reference numeral130(P+1). When a new network security device is added, its device descriptor and native security policy data is imported as shown atreference numeral600. This policy data is normalized against the generic policy model, as described above in connection withFIG. 5, and normalized policy data610(P+1) is sent to the network security device130(P+1). Creating a network security policy template (from one or more existing normalized network security policies) and deploying it to a network security device is described in more detail hereinafter under the heading “Network Security Policy Template Creation and Deployment.”
• Next, atstep260, themanagement entity110 receives network security events from the network security devices130(1)-130(P) in the customer datacenter. Data describing network security events from respective network security devices is shown at reference numerals700(1)-700(P) inFIG. 7. Examples of network security events include, but are not limited to including, how many times a device wants to access a range of Internet Protocol (IP) addresses (that may or may not the addresses are allowed in a WSA device, for example), how many times an outside device is trying to get through a firewall over a certain range IP addresses, access attempts to certain destination IP addresses beyond a firewall, etc. As explained above, these security events include application security, content security, authentication related, etc., and thus they are “security events” in general.
• Next, atstep270, the management entity processes (orchestrates and automates) the received network security events. As shown inFIG. 8, themanagement entity110 performs event processing at800. This is important because it may be necessary to make changes to rules to permit certain activity that is revealed by numerous attempts to certain IP addresses outside the customer datacenter from within the customer datacenter as described herein, or to prevent certain activity that appears to be malicious. Thus, as shown inFIG. 8, themanagement entity110 may generate one or more event triggered controls810(1)-810(P) to one or more of the network security devices130(1)-130(P), causing a change to the native rules applied by respective ones of the network security devices130(1)-130(P). In addition, the management entity may send event triggered alerts, shown at820, to anetwork administrator830 or some specified destination associated with the customer datacenter.
• To summarize, at a system level, a method is provided comprising: at a management entity, discovering multiple security devices connected to a network, each security network device to control network access by devices associated therewith according to a corresponding native security policy that is based on a corresponding native policy model associated with the security device; importing the native security policies from the corresponding security devices over the network; normalizing the imported native security policies across the security devices based on a generic policy model, to produce normalized security policies that are based on the generic policy model and representative of the native security polices; and processing (orchestrating and automating) received network security events among the security devices based on the normalized security policies. Processing may involve reporting the received network security events to a desired destination (e.g., a network administrator), and making changes to security policies as needed based on the received network security events.
• Reference is now made toFIG. 9 for an example of a business policy and a network security policy.FIG. 9 depicts a use case on how security policy may need to run on multiple devices to be complete, from the gateway controlling access of Bring Your Own Devices (BYODs) to a cloud firewall requiring a policy that is synchronized across multiple devices by definition. In this example, there a network security device130(1) in the form of cloud firewall, which may reside outside of the physical premises of a customer facility. There is another network security device130(2) in the form of an identity services engine (ISE). The network security device130(1) is useful to control access to certain web services, such asWeb Service 1 at reference numeral140(1) andWeb Service 2 at reference number1402(2). Network security device130(2) connects to arouter900. A user, such as an employee in a corporation, is shown at910, who may be using a Bring Your Own Device (BYOD), such as a tablet computer on thecorporate network920. The corporate network connects to the outside world viaWAN930.
• There is a business policy set by the corporation that says access toWeb Service 1 is to be denied, butWeb Service 2 is to have access to employees in the corporate network. The rules on the network security devices130(1) and130(2) are as follows.
• Network Security Device130(1):
• • Src=Any User=Employees Dst=Any Application=Web Service 1 Action=Deny
  • Src=Internal network User=Mktg Dst=Any Application=Web Service 2 Action=Allow
• Network Security Device (ISE)130(2):
• • Rule Name: Employee
  • Identity Group: Any
  • Other Condition: AD1:ExternalGroups EQUALS corporate.com
    • /Users/Domain Users AND Session: Posture
    • Status EQUALS Compliant
  • Permissions: Allow
• Themanagement entity110 receives network security events from the network security devices130(1) and130(2) which indicate that a desired business goal is not being met with currently running security policies. Themanagement entity110 thus makes a change of those security policies with a single normalized policy as described above in connection withFIG. 8, to return operations of the network security devices130(1) and130(2) so that the business policy is achieved.
• Reference is now made toFIG. 10.FIG. 10 illustrates a further capability of the cloud-basedmanagement system100. Thesystem100 can be integrated with various security intelligence feeds. For example, a collectivesecurity intelligence entity1000 receives events from the customer data center via the cloud-basedmanagement system100, and receives security intelligence feeds from multiplesecurity intelligence resources1010 and1020. This allows the collectivesecurity intelligence entity1000 to correlate security intelligence data and events from a customer datacenter's security devices to determine threats. The collectivesecurity intelligence entity1000 can generate reports provide insight into a security state of a customer's network. The data in these reports may be linked with the policy generation capabilities in themanagement entity110 to quickly create business policy rules that will mitigate risks.
• FIG. 11 illustrates an examplegraphical user interface1100 for a threat summary report that themanagement entity110 may generate and present to a network administrator for a customer datacenter. The report includes a column for different categories of risks (Malware, Applications, Users) as shown at1110. Next to each category type name there is a expandicon1112. There is also a column for risk level shown at1120 and a graphical element, such as aslider bar1122, may be used to indicate risk level for each category.Columns1130 indicates the number of block packets for each category, and column1140 indicates the number of allowed packets for each category. The slider bars1122 represent accepted levels of risk and changes can be made by a user by moving aslider bar1122, which will result on change in network security policies and updates across devices.
• FIGS. 12A and 12B illustrate further details of the summaryreport user interface1100. In particular, if a user selects the expandelement1112 for the Applications category as shown inFIG. 12A, additional details are displayed for types of applications. Specifically, a table1200 is displayed contains a column for Block, Name, Type, Popularity, Risk and # Blocked. In this example, there are three web applications ongoing,Web Application1,Web Application2 andWeb Application3. The Blocked column indicates whether a policy has been set to block traffic for that application. Thus,FIG. 12A shows that policies have already been set to block/deny traffic forWeb Application2 and forWeb Application3, and the number of packets that have been blocked (the box in the Block column has been checked as shown at reference numerals1210 and1220) for those two applications are indicated in the column # Blocked. At this point in time, traffic forWeb Application1 has not been blocked because the box in the Block column has not been checked as shown at reference numeral1230.
• FIG. 12B shows theuser interface1100 similar to that shown inFIG. 12A, now as shown at1210, a new policy is set by clocking the box in the Block column forWeb Application1 as shown at reference numeral1240, which will start blocking traffic forWeb Application1. Up to this point, traffic forWeb Application1 has not been blocked so the field in # Blocked forWeb Application1 is still “0” but over time that number will increment upwards.
• Turning now toFIG. 13, a block diagram is shown of an example hardware implementation for themanagement entity110. In one example, themanagement entity110 includes one or more servers112(1)-112(M). Each server includes one ormore processors1310, one or morenetwork interface units1312 andmemory1314. Thememory1314 stores controlsoftware1316, that when executed by the processor(s)1310, cause the server to perform the various operations described herein for themanagement entity110.
• The processor(s)1310 may be a microprocessor or microcontroller (or multiple instances of such components). The network interface unit(s)1312 may include one or more network interface cards that enable network connectivity.
• Thememory1314 may include read only memory (ROM), random access memory (RAM), magnetic disk storage media devices, optical storage media devices, flash memory devices, electrical, optical, or other physically tangible (i.e., non-transitory) memory storage devices. Thus, in general, thememory1314 may comprise one or more tangible (non-transitory) computer readable storage media (e.g., memory device(s)) encoded with software or firmware that comprises computer executable instructions. For example,control software1316 includes logic to implement (i) network security classification, (ii) network policy unification, (iii) network policy template creation and deployment, (iii) a generalized network policy user interface, as described below, and (iv) processing (orchestration and automation) of network policy management functions in connection with collections of reported network security events. Such logic also includes logic to implement Graphical User Interfaces (GUIs) as necessary in connection with the classification, unification, template creation and deployment, network policy user interface, and orchestration and automation.
• A user, such as a network administrator, may interact with themanagement entity110, to receive reports, change policy, etc., through GUIs by way of auser device1320 that connects by way of a network (local area network (LAN) and/or wide area network (WAN)) with themanagement entity110. Theuser device1320 may be a personal computer (laptop, desktop), tablet computer, SmartPhone, etc.
Security Policy Classification
• As mentioned above,management entity110 imports network security policies from multiple distributednetwork security devices130. Each network security policy may include a combination of policy features that collectively define the policy and control the security behavior of the network security device from which the policy was imported. The policy features include network security rules, and may also include one or more of configuration, interface, object, and object content features. According to an embodiment presented below,management entity110 classifies or categorizes the imported network security policies into different network security policy classifications based on commonality between the policy features included in the network security policies. The different network security policy classifications or categories include (1) identical, (2) similar, (3) unique, and (4) investigate further.
• With reference toFIG. 14, there is an illustration of anexample classification operation1400 performed bymanagement entity110. In the example ofFIG. 14,management entity110 determines commonality between policies A (multiple instances of policy A associated with multiple network security devices), B1, B2, B3, C, D, and E imported from different network security devices based on comparisons between the respective policy features of the imported policies. Such commonality may indicate imported policies which are related, linked, and the like, across security devices, and indications of commonality to a user bymanagement entity110 may help the user to manage such policies across the devices. The commonality determination indicates that the instances of policy A are identical, policies B1, B2 and B3 are similar to each other, policies C and D are unique (i.e., different from each other and all of the other imported policies), and policy E needs further investigation. Accordingly,management entity110 groups the instances of policy A into an identical bucket orclassification1410, policies B1, B2 and B3 into asimilar classification1412, policies C and D into aunique classification1414, and policy E into an investigate (further)classification1416.
• With reference toFIGS. 15A and 15B, there is a flowchart of anexample overview method1500 of importing and classifying network policies fromnetwork security devices130.
• At1502,1504, and1506,management entity110 connects to a network, and then to networksecurity devices130 over the network, using URLs that direct the management entity to the devices, for example.
• At1510,1512, and1514,management entity110 respectively imports network security policies fromsecurity devices130 over the network, detects any failures that may occur while importing the policies, and whether the importing should continue if any such failures are detected. The importation of policies in1510-1514 results in a policy importation report, a number of network policy rules that are imported, basic definitions associated with the imported policies such as a number of Network Address Translation (NAT) objects and other metrics, and a number of failed connections and imports.
• At1518,management entity110 determines which of the imported network security polices are identical to each other based on a comparison between the imported policies, and classifies those policies determined to be identical into an identical classification.
• At1520,management entity110 determines which of the imported network security policies are similar to each other based on a comparison between the imported policies and classifies those policies determined to be similar into a similar classification.
• At1524-1532,management entity110 determines whether imported network security policies not previously classified as either identical or similar should be classified into either unique or investigate further classifications. Based on that determination,management entity110 classifies the remaining network security policies as either unique or investigate further. More specifically, at1528,management entity110 determines whether any remaining network security policies should be classified as unique and, if it is determined that any remaining network security policies should be classified as unique, classifies those policies into the unique classification. At1530,management entity110 determines which of the network security policies require further investigation and classifies those policies, if any, into the investigate further classification. At1532, management entity determines if any remaining network security policies should be ignored, i.e., not classified as described above.Operations1528 and1530 will be described further below in connection withFIGS. 18 and 20B.
• With reference toFIG. 16, there is an illustration of an example Graphical User Interface (GUI)1600 displayed onmanagement entity110 and through which a user may identify/select, connect to, and then import network security policies from one or more network security devices. At1605 and1610, the user selects a type of network security device, e.g., ASA, in a “device” portion ofGUI1600 and selects a tag or label to be associated with a network security policy imported from the selected device type in a “device group” portion of the GUI. At1615, the user enters a URL for the selected device type into a “device URL” portion ofGUI1600. At1620, the use selects a “connect” icon to connect with the selected device type using the entered URL. At1625, the user selects an “import” icon to import network security policies from the connected device type.
• With reference toFIG. 17, there is an illustration of anexample GUI1700 displayed bymanagement entity110 that may be used to initiate policy classification of imported security policies and shows resulting security policy classifications.GUI1700 includes a “classify” icon that, when selected by a user, initiates policy classification of imported security policies bymanagement entity110.GUI1700 also shows first and second example policy classifications atrows1705 and1710 that result from the policy classification. In the example ofFIG. 17,policy classifications1705 and1710 each represent an identical classification, as described below.
• Policy classification1705 shows a network security rule that is formatted according to an access list (ACL) extended rule model. The ACL extended rule includes the following rule parameters: service/app (e.g., HTTP); a protocol (e.g., TCP); a port (e.g.,80); a source (address) 192.168.0.0; a destination (address) 0.0.0.0; and an access (e.g., allow).Policy classification1705 also includes a count (e.g.,4) that indicates that 4 network security policies imported from four security devices share the indicated security rule (the ACL extended rule). In other words,policy classification1705 represents an identical policy classification for the indicated security rule, across the 4 security policies/devices.Policy classification1710 similarly represents an identical policy classification across four security policies/devices.
• GUI1700 includes a “filter” option/field1715 through which the user may specify a policy/rule parameter that, once specified, causes the GUI to shows only those classified policies (e.g., rules therein) that include the specified parameter. Thus, the specified parameter operates as a filter of what classified policies (e.g., rules therein)GUI1700 shows. In an example, the filter parameter may be HTTP, TCP, or any other parameter.
• GUI1700 also includes a “sub-class name” field/option1720 through which the user may select and/or enter a sub-class name that is to be assigned to a combination of selected ones of the policy classifications presented by the GUI (e.g.,policy classifications1705 and1710). This feature is described below in connection withFIGS. 27-33.
• In the example ofFIG. 17,GUI1700 showspolicy classifications1705 and1710 in unexpanded views.GUI1700 also includes expand icons “+” associated with each ofpolicy classifications1705 and1710 that, when selected, causes the GUI to expand on the associated policy classification (e.g., identical rule) to reveal identities/names of the different security devices from which the identical rules were imported, as shown inFIG. 18.
• With reference toFIG. 18, there is an illustration ofGUI1700 with an expandedview1805 after the expand icon “+” associated withpolicy classification1705 has been selected.Expanded view1805 reveals identities of 4 security devices from which the identical rules inpolicy classification1705 were imported.
• Returning tomethod1500 ofFIG. 15,management entity110 generates identical, similar, and unique/investigate policy classifications of imported policies inoperations1518,1520, and1524-1532, respectively, based on comparisons between corresponding features or “points of comparison” of the imported security policies. For example,management entity110 compares security rules from different policies against each other as a point of comparison.Management entity110 may also compare one or more corresponding ones of the other policy features against each other, such as the configurations, interfaces, objects, and object content of the different security policies.
• Each of the above mentioned security policy features (e.g., security rules, configuration, interface, object and object content.) includes local parameters, and each of the local parameters may form the basis of a point of comparison between security policies. The local parameters of the different network security features that may serve as points of comparison for similarity tests during classification are now described.
• A security rule typically permits or denies network access based on, e.g., source and destination addresses, network protocols, device ports, time of day, and the like. Thus, the network security rule may include the following parameters/points of comparison: name of rule group (e.g. “inside-in” vs. “inside-out”); permit/deny; protocol (e.g., IP, TCP, UDP, ICMP); source (address) vs. source (address); destination (address) vs. destination (address); source vs. destination; device/service ports; interfaces; context (e.g. a deny rule surrounded by other deny rules); and config context (e.g. the rule appears on a branch config).
• A configuration may include: a type of site at which the network policy is employed, e.g., branch or hub; a location of the site (e.g., Europe vs. Asia); a proximity to “outside,” e.g., presence of an outside interface; and a number of interfaces, e.g., ports, in use.
• An interface may include a security level (e.g., as an ordinal variable); a name (e.g., what beyond specific word presence e.g. inside, outside, DMZ, VLAN, and the like); an interface classification/context; and an IP address.
• An object may include a name; object contents; and an object count (e.g. a group that lists 1000 specific objects).
• Object content may include a percent of intersection or ratio; an amount of intersection; a size (e.g. large vs. small IP range); a ratio of covered IP ranges; a type (e.g., internal vs external IP addresses, generic range vs. specific addresses, email-related ports vs. VPN); a contained hierarchy (e.g., number of nodes); and contained nodes (how many levels down the nodes are layered).
• Management entity110 employs network security policy matching algorithms inmethod1500 to determine a level of likeness or commonality between network security policies based on comparisons of the above listed feature parameters/points of comparison.
• Atoperation1518,management entity110 determines different network security policies to be identical if, for example, the corresponding point(s) of comparison are identical or resolve to the same value. For example, if a point of comparison includes network security rules of the different network security policies,management entity110 determines the network security rules to be identical if the rules represent identical strings or resolve to the same value. For example, a comparison between two firewall security rules that each follow the “source-destination-protocol-port-result” format may resolve corresponding rule parameters across the two rules to ascertain identity. Consider the following two example network security rules for an ASA formatted in an access list-extended form:
• • a. Access-list inside_in extended permit TCP object-group InsideNet object-group GenRec eq 8080; and
  • b. Access-list inside_in extended permit TCP object-group InsideNet object-group GenRec object tomcat.
• The two rules above are not identical by literal string comparison, but it may be assumed that the name object “tomcat” in the second rule will resolve to “8080” and thus the first rule/string, when resolved, will be identical to the second rule/string. In an example that compares network security rules for an NGFW, the comparison may include “user” and “action” as additional rule parameters.
• Thus, it is desirable to determine whether a point of comparison in the form of security rules compares rules in their resolved form, as illustrated inFIG. 19, for example. With reference toFIG. 19, there is an illustration of an example access list-extendedrule1900 in which the “name” is resolved to a bound interface.
• Atoperations1520 and1524-1532,management entity110 determines whether different network security policies are similar, unique, or need further investigation based on commonality between the policies, if any.
• With reference toFIG. 20A, there is a flowchart of anexample method2000 of determining commonality/similarity between different network security policies based on matching points of comparison (i.e., corresponding feature parameters of the network security policies).Method2000 may be used inoperations1520 and1524-1532.
• At2005, different points of comparison (i.e., policy feature parameters) are defined. These points of comparison will form a basis for a determination as to whether the different security policies are sufficiently similar as to be placed into the similar policy classification or sufficiently dissimilar as to be placed into the unique policy classification.
• At2010, a weight or coefficient w_iis assigned to each point of comparison.
• At2015, corresponding ones of the points of comparison from different network security policies being compared are compared to each other to arrive at a Boolean result, e.g., match=1, no match=0.
• At2020, each Boolean result is multiplied by the corresponding assigned weight to produce weighted Boolean results.
• At2025, the Boolean results are combined into a match score according to a predetermined expression/equation.
• At2030, the match score is compared to a score threshold. If the compare indicates the match score is equal to or greater than the score threshold, the different network security policies are deemed similar to each other and thus classified into the similar classification. If the compare indicates the match score is below the score threshold, the different network security policies are deemed dissimilar to each other and thus classified into either the unique classification or the investigate classification.
• In an example in whichoperation2005 ofmethod2000 defines as the points of comparison various rule parameters used in the access list-extended model,operation2025 may evaluate the following expression, in which “|match on <point of comparison>?|” defines a match/comparison test that evaluates to a Boolean result:
• • match score=w_i|match on name?|*w₂|match on permit/deny?|*
    • w₃|match on protocol?|*w₄|match on source address?|*
    • w₅|match on destination address?|+
    • [w₆|match on service ports?|+w₇|match on rule context?|]
• In the above equation for match score, both a multiplicative combination and an additive combination of tests results are used. The multiplicative combination is used for points of comparison deemed of higher importance, while the additive combination is used for points of comparison deemed of lower importance. Also, weights w_imay be initially set to 1, but other values may be used. In addition, the score threshold may be set to 2, for example, so that if the match score evaluates to 2 or greater, the network security policies being compared are deemed similar, otherwise the policies are deemed unique.
• With reference toFIG. 20B, there is a flowchart of anexample summary method2050 of classifying network security policies.
• At2055 and2060,management entity110 connects with and imports network security policies fromnetwork security devices130. Each network security policy includes network security rules. Each rule includes rule parameters to cause the corresponding network security device to permit or deny network access based on a network protocol, source and destination addresses, and a device port, for example.
• At2065,management entity110 automatically classifies the network security policies into network security policy (NSP) classifications based on commonality between the network security rules acrossdevices130 associated with the network security policies (see, e.g.,FIG. 14). To do this,management entity110 compares the rule parameters of each rule of each network security policy and, based on the comparison results:
• • a. Classifies network security policies into one or more identical NSP classifications if all of their associated rule parameters are identical to each other;
  • b. Classifies network security policies into one or more similar NSP classifications if only some of their associated rule parameters are equivalent to each other;
  • c. Classifies network security policies into a unique NSP classification if none of the associated rule parameters are equivalent to each other; and
  • d. Classifies network security policies into an investigate classification if the policies require further investigation.
• At2070,management entity110 displays selectable ones of the NSP classifications, including network security rules therein, along with selectable expand view and view filter options (see, e.g.,FIG. 17).
• At2075,management entity110 displays NSP classifications in an expanded view if a selection of the expand view is received. Also,management entity110 displays filtered NSP classifications (and rules therein) if “filter on” parameters are received through the view filter option (see, e.g.,FIG. 18).
Security Policy Unification
• Embodiments directed to normalizing imported rules are now described in the context of security policies for illustrative purposes. As used herein, the terms “normalize” and “unify” and their corresponding derivatives (e.g., normalization and unification) are synonymous and may be used interchangeably.
• Referring again toFIG. 1 (andFIGS. 3-8),management entity110 imports different security policies from different types of security devices over a network, as described above. Each imported security policy is considered to be “native” to the security device from which the policy is imported in that the policy is based on a native policy model associated with that security device. Each of the security devices controls network access by devices associated therewith according to the corresponding native network security policy.
• Management entity110 normalizes the imported native security policies across security devices based on a generic policy model, to produce a normalized network security policy generally representative of all of the native network security polices.Management entity110 may also modify the normalized security policy to suit further generalized security goals, translate the modified normalized security policy to corresponding native security policies representative of the modified normalized policy, and then push the resulting native policies to corresponding ones of security devices. In addition, a normalized policy may be created by a user, automatically translated to suitable native policies, and pushed to multiple security devices (e.g., devices130). A normalized network security policy is also referred to herein as a ‘generic” or “unified” policy network security policy.
• As mentioned above,management entity110 normalizes native security policies that are based on corresponding native policy models to a generic security policy that is based on a generic policy model. In one embodiment, the generic policy model, referred to as the Principal-Action-Resource-Context-Result (PARCR) model, is defined as follows:
• • If {principal} tries to perform an {action} on {resource} within {context} then {result}.
• With reference toFIG. 21, there is an illustration of a PARCR model2100 (also referred to as a “PARCR policy model”). ThePARCR model2100 includes basic PARCR model components, i.e., principal (P), action (A), resource (R), context (C), and result (R), expressed in the “if-then-result” syntax. A normalized or generic network security policy based on thePARCR model2100 includes generic security rules based on the PARCR model. The generic security rules include the PARCR model components (referred to as PARCR “rule components”) expressed in the if-then-result syntax. Thus, normalizing a native security policy includes mapping native features such as native security rules (and their respective rule parameters) expressed according to the native policy model to corresponding PARCR rule components expressed in the PARCR model syntax (i.e., the if-then-result form).
• The examples below show mappings between native security policies and normalized or generic network security policies based on thePARCR model2100, at the rule level. In other words, in the examples, the native rule parameters of native security rules are mapped to corresponding PARCR rule components.
Example 1
• • a. Simple ASA rule (in access list-extended form):
    • Access-list left-to-right extended permit ip host 172.16.1.10 host 192.168.1.100; and
  • b. PARCR rule components and rule:
    • principal=172.16.1.10 (source)
    • resource=192.168.1.100 (destination)
    • context=ip (protocol)
    • result=permit
    • action=implied anything
    • if 172.16.1.10 tries to perform anything on 192.168.1.100 within ip then permit.
Example 2
• • a. More complex ASA rule:
    • Access-list someName extended permit tcp 172.19.103.0 255.255.255.0 object-group ApplicationServers object-group DM_INLINE_TCP_443; and
  • b. PARCR rule components and rule:
    • principal=address range 172.19.103.x
    • resource=ApplicationServers (group of resources)
    • context=tcp+port group DM_INLINE_TCP_443
    • result=permit
    • action=implied anything
    • if 172.19.103.x tries to perform anything on ApplicationServers within tcp and DM_INLINE_TCP_443 then permit.
Example 3
• • a. Simple WSA rule:
    • Block all users from using facebook messaging; and
  • b. PARCR rule component and rule:
    • principal=all users (anyone)
    • resource=facebook messaging
    • context=any
    • result=block
    • action=any
    • if anyone tries to perform anything on facebook messaging within any then block.
Example 4
• • a. More complex WSA rule:
    • Allow all users to use Linked in but only allow HR to post jobs on Linkedin, allow all users to use Linkedin; and
  • b. PARCR rule components and rule:
    • principal=address range 172.19.103.x
    • resource=ApplicationServers (group of resources)
    • context=tcp+port group DM_INLINE_TCP_443
    • result=permit
    • action=implied anything
    • if 172.19.103.x tries to perform anything on ApplicationServers within tcp and DM_INLINE_TCP443 then permit.
Example 5
• • a. Simple WSA rule:
    • Block all users from using facebook messaging; and
  • b. PARCR rule components and rule:
    • principal=all users (anyone)
    • resource=facebook messaging
    • context=any
    • result=block
    • action=any
    • if anyone tries to perform anything on facebook messaging within any then block.
• With reference toFIG. 22, there is a block diagram of an examplepolicy unification module2200 that may be implemented inmanagement entity110 to convert or map between the native network security policies expressed according to native policy models and normalized or generic policies expressed according to the PARCR model (e.g., PARCR model2100).Module2200 includes a Northbound Application Programming Interface (API)2204 (which may be a Representational State Transfer (REST) API) that interfaces with a policy user interface (UI)2205, aunified policy engine2206, aSouthbound model API2208, and multiple, network security device specific, network security device plug-ins2210(1)-2210(N) that interface to corresponding ones ofnetwork security devices130.
• Northbound API2204 allows services or applications such a generalized network security UI or a policy orchestrator (e.g., policy UI2205) to communicate withunified policy engine2206.API2204 allows such services/applications to perform operations such as move policy from one network security device (e.g., one of devices130) to another or perform operations that traverse the devices and that require policies in more than one network security device type. One example is to block all IP addresses from a geo-location. Another example is to set up access blocking based on the IP being used in the access and a service such as (Secure Shell) SSH over HTTP.
• Southbound model API2208 perform actions on a network security device specific (i.e., native) network security policy.API2208 receives native network security policies and rules imported via corresponding plug-ins2210 (from the corresponding network security devices).API2208 also pushes native network security policies down tonetwork security devices130 via the corresponding plug-ins2208.API2210 may also perform some of the mapping or translating between imported native network security policies and normalized/generic network security policies as mentioned above.
• Device plug-ins2210 read and write network security policy information directly from and to network security devices130 (e.g. WSAs). Plug-ins2210 may also include portions of mapping logic to assist with translating native network security policies into the generic/normalized network security policy format. In an example, WSA network access policy rules would define “Block, Monitor, Allow, Warn” as the rule actions. Other WSA network access policy rules, such as WSA time definitions (e.g. “Core Business hours”) may be mapped to the “{context}” generic rule component.
• Policy engine2206 may implement a policy object model layer to tie network security device specific policies (i.e., native policies) into the generic policy model, e.g., the PARCR model. This may be implemented as a JavaScript Object Notation (JSON) file that ties generic network security policy and rule components/objects (e.g. principal, context, action) to Java classes that can enumerate the supported components/objects and validate the generic network security rules. The policy object model layer may also allow network security device plug-ins2210 to attach network security device specific attributes to object definitions. For example, WSA plug-in2210(1) may indicate that there is an immutable Boolean on the policy called “Is Global Policy” and a mutable integer called “Policy Order,” whose legal values are between n and m. In another embodiment, the policy object model layer may be implemented inSouthbound model API2208.
• In an embodiment, aggregated plug-ins may be used in an environment in which policies are to be defined to cross network security devices. For example, a policy definition may require sub-policies in an ASA and a WSA. Such policies may be defined by higher level plug-ins, which, instead of talking to network security devices directly, would talk to network security device plug-ins and build policies on top of the generic policy definitions exposed by the device plug-ins.
• With reference toFIG. 23, there is an illustration ofexample PARCR mapping2300 performed in part by plug-ins2210 ofpolicy unification module2200. According tomapping2300, native ASA policies including corresponding native rule parameters which protect based IP protocols, native WSA policies including corresponding native rule parameters which protect based on URLs, and native firewall policies including corresponding native rule parameters which protect based on content/applications are each mapped to corresponding generic rule components of the same generic policy model, i.e., the PARCR model.
• With reference toFIG. 24, there is an illustration of an example PARCR model anatomy or break down2400 (referred to as “policy anatomy”). The example ofFIG. 24 represents one possible data model corresponding to an ASA plug-in. It is understood that partitioning between objects in the anatomy may be varied and that other policy anatomies for other data models are possible. The PARCR policy anatomy is based on the following assumptions:
• • a. Policies are groups of rules;
  • b. All rules are if/then statements (see (c));
  • c. If {principal} attempts to perform {action} on {resource} when {context} then {result};
  • d. A policy is high level statement;
  • e. A policy contains a set of rules;
  • f. A rule is a mid-level statement;
  • g. An entity is an object to describe a person or thing, and may map to principal or resource rule components of the PARCR model;
  • h. Entities contain multiple attributes;
  • i. Attributes are static or dynamic states of an entity;
  • j. Context contains environmental attributes external to entities;
  • k. A result is a set of obligations to be performed if the conditions are met; and
  • l. Obligations can be just about any process including simple allow or deny, logging, or complex code launching.
• Going from bottom-to-top inFIG. 24,policy anatomy2400 depicts a mapping of predefined native policy/rule objects in the native domain (bottom ofFIG. 24) to PARCR rule components in the PARCR domain (top ofFIG. 24). In the native domain, an ASA construct2406 (for an ASA device) operates on a variety of objects depicted as boxes inside thebox ASA2406. The ASA (construct)2406 may be implemented in ASA plug-in2210(2), for example. Each of the objects depicted inASA2406 represents a definition of something that the ASA may control. For example: the ASA_GroupObject may define a range of IP addresses protected byfirewall2406, or a group of other objects; the ASA_Object may define or point to all of the other objects defined in the ASA; the ASA_ServiceObject may be a combination of a IP addresses, a service/device port, and a protocol; the ASA_NetworkObject may be an endpoint; and the ASA_Result includes a permit or deny.
• Common object2408 is also depicted in the native domain.Common object2408 is common across multiple security device plug-ins2210, and may include, for example, objects similar to any of the example objects described above in connection withASA2406.
• As depicted inpolicy anatomy2400, various objects in the native domain may represent anentity2412 accessible to the PARCR model.Entity2412 maps to either or both the principal and the resource rule components of the PARCR model, as depicted inFIG. 24. Each of the PARCR rule components depicted inFIG. 24 as boxes labeled principal, action, resource, context, and result includes a name represented as a string, and a universally unique identifier (UUID).
• Also as depicted inpolicy anatomy2400, the PARCR rule components are combined into a generic/PARCR rule2420 according to the if-then-result syntax.Generic rule2420 forms part of a genericnetwork security policy2430.
• The following example shows a mapping between a native network security policy expressed in terms of predefined objects and a generic network security policy based on the PARCR model:
• • a. ASA acl-list using above object model:
    • Access-list executive_access_to_finance_extended permit ip object-group Executive_Network_object-group Finance_Network
  • b. PARCR rule components:
    • principal=ASA_GroupObject named Executive_Network (see definition below)
    • action=any
    • resource=ASA_GroupObject named Finance_Network (see below)
    • context=Context object with namevalue type:ip
    • result=ASA_Result with value “permit”
    • ASAGroupObject.name=Executive_Network
    • ASAGroupObject.asaObjects={ASA_NetworkObject.endpoint, . . . }.
• In normalizing native security policies (based on native policy models) to generic security policies (based on a common generic policy model) as described above, it is desirable to “bridge” between the generic (e.g., PARCR) and native (e.g., WSA) models through corresponding data elements associated with the native and generic policies (as opposed to code). Accordingly, a policy data-driven “policy model bridge” that describes or defines native policies in terms of the PARCR model may be used to map the native policies to PARCR rule components. As described above in connection withpolicy anatomy2400, objects that are not strictly security policies but are referenced by the policies are modeled as entities. An example policy model bridge is described below in connection withFIG. 25.
• With reference toFIG. 25, there is shown an example of apolicy model bridge2500 that may be used to map a simplified WSA access policy to the PARCR model. Other policy model bridges may be used corresponding to other security devices.Policy model bridge2500 may also map native entity objects directed time ranges over which a native rule is to be active and URLs that may need to be blocked or allowed to corresponding PARCR rule components. In the example ofFIG. 25,policy model bridge2500 operates to perform the WSA to PARCR rule mapping shown above in Example 1. In that example,policy model bridge2500 may be implemented as a JSON file that represents a contract between the PARCR model and WSA plug-in2210(1), and may be implemented in the WSA plug-in.
• Policy model bridge2500 (referred to as simply “bridge2500”) includes a “header”2505 that provides basic information about the corresponding plug-in.Header2505 indicates a name of bridge2500 (e.g., “WSA”), which is unique across all plug-ins2210.Header2505 also includes a version (e.g., “1”) to indicate a schema of the policy bridge JSON file being used.Header2505 includes a service definition that indicates an actual implementation of the service as well as a mock implementation. Running a mock implementation simplifies, for example, automated testing ofgeneric policy UI2205. The exposed service retrieves concrete instances of native policies and entities.
• Afterheader2505,bridge2500 includes “policies”2510, which enumerates substantially all of the policies supported by thebridge2500. Different types of network security devices, such as the WSA, will typically expose network security policies of many different types. Some of these network security policies may be complex and may take advantage of many of the features available in the PARCR model, while other policies may be very simplistic.Policies2510 define the following:
• • a. “Type”—The type of a policy may be unique within the JSON file. Each plug-in can expose multiple policies, which are differentiated by type.
  • b. “Preferences”—A set of generic preferences that govern the behavior of the policy. For example, “readonly” would be used to indicate that the policy can only be read and cannot be modified.
  • c. “Rule mapper”—This object expresses native policy rules in the PARCR model by mapping native objects (e.g., native rule parameters) to the {principal}, {action}, {resource}, {context}, and {result} components of a PARCR rule. Not all of the PARCR components will be applicable to all native rules. It is therefore legal to omit PARCR rule components to indicate that they aren't used by the underlying device.
• Bridge2500 also describes what types ofentities2515 can be referenced inside of network security policies. In other words, the network security policy is a set of rules composed of references to external entities. Sometimes the types of these entities will be defined within a namespace (e.g. plug-in), which means the entity is specific to the native device. For example, if WSAs have the notion of URL categories, but other devices do not, it makes sense to the URL category entity to be scoped to the WSA plug-in. In other cases an entity may be something much more generic and may apply across multiple devices. For example, a principal definition that indicates a user was authenticated by Active Directory may be applicable to WSA and a particular type of firewall and therefore might be defined in a non device-specific plug-in. Thus, a policy model bridge, such asbridge2500, may include a reference mechanism or section used to refer to types within the current namespace and within other namespaces. An example reference section is provided below:
• • namespace”: “identity”,
  • “type”: “ACTIVE_DIRECTORY_AUTHENTICATED”
• Inentities2515,bridge2500 includes a “URL Category” entity that is expressed through a list of resources. Two types of resources are accepted. A user may pick a type and enter a free form string. Then, depending on type, the string would be validated using one of two validators—the first ensures a string is a valid URL and the second that the string is a valid regular expression. As the user enters the string,policy UI2205 may call the validators and give a visual indication if the string is incorrect. Types that start with a hash (‘#’) are build-in types.
• Inentities2515,bridge2500 also includes a time range entity, which indicates a time range to be mapped to the resources rule component of PARCR. For example, a WSA policy may express two custom time ranges. The first one is “Extended Business Hours” and is defined as being Monday through Friday 7 am to 6 pm andSaturday 10 am to 4 pm. The second one is “Core Business Hours” and is defined as being Monday throughFriday 10 am to 2 pm.Bridge2500 maps such time ranges into the PARCR rule.
• A policy model bridge, such asbridge2500, may also include an access policy action, which is a simple entity that does not include any components. A “hidden” flag may be used to indicate that this entity is not exposed to an end user as a first class object and that it can only be referenced in the context of policy definitions. Enumerating this entity class would return three instances, for example, “Block”, “Monitor”, and “Warn.”
• With reference toFIG. 26, there is shown a flowchart of anexample method2600 of converting between normalized network security policies and native network security policies as described above. Reference can be made toFIG. 1 in connection with the description ofFIG. 26.
• At2605,management entity110 receives from security devices corresponding native security policies each based on a native policy model associated with the corresponding network security device. Each of the security devices controls access to resources by other devices according to the corresponding native security policy. Each native network security policy includes a set of one or more native network security rules, each native network security rule including native rule parameters expressed according to the corresponding native policy model.
• At2610,management entity110 normalizes the received native network security policies across the network security devices based on a generic policy model (e.g., the PARCR model), to produce normalized network security policies that are based on the generic policy model and representative of the native network security polices. To do this, for each received native network security policy,management entity110 maps the native rule parameters expressed according to the corresponding native policy model to corresponding generic rule components of the generic policy model to form a generic security rule. For example,management entity110 maps the native rule parameters to PARCR rule components according to the PARCR model in the form: if {principal} tries to perform an {action} on {resource} within {context} then {result}, to generate the generic rule.
• At2615,management entity110 receives a generic security policy (e.g., PARCR rules) based on the generic policy model (e.g. the PARCR model).
• At2620,management entity110 translates the generic security policy to multiple native security policies each based on a corresponding one of the native policy models associated with the corresponding one of the network security devices. To do this,management entity110 maps the generic rule components to native rule parameters expressed according to the corresponding native policy model to form native rules representative of the one or more generic network rules.
• At2625,management entity110 provides the multiple native security policies to the corresponding security devices to enable the security devices to implement the native security policies.
Security Policy Template Creation and Deployment
• As discussed above in connection withFIGS. 14-20B,management entity110 classifies security policies into identical, similar, unique, and investigate NSP classifications as appropriate, and displays selectable ones of the NSP classifications (and their corresponding rules) on a GUI. For example, with reference again toFIG. 17,GUI1700 displays twoselectable NSP classifications1705 and1710, each which groups together identical network security policies/rules. According to embodiments presented below, a user may interact withmanagement entity110 to create a security policy template that combines multiple selected ones of the displayed NSP classifications (and their corresponding network security rules). The user may then build a new security policy based on at least that security policy template, and commit that new network policy to a security device, which may reside in a new site at which the user wishes to control access to a resource.
• The process of creating and using a security policy template is described below in connection withFIGS. 27-33. Reference is also made toFIG. 1 for purpose of these descriptions.
• With reference toFIG. 27, there is an illustration ofGUI1700 in which the user has entered a policy sub-class name “Branch Allow Web Traffic” into sub-class name field/option1720, and selected both ofidentical NSP classifications1705 and1710. In response,management entity110 assigns the combination of selectedNSP classifications1705 and1710 and the rules contained in those NSP classifications to the policy sub-class “Branch Allow Web Traffic.” These actions create the policy sub-class “Branch Allow Web Traffic” that combines all of the network policy rules ofNSP classification1705 and1710, and stores the policy sub-class with other previously created and stored policy sub-classes. A security policy template may be conveniently created based on all of the previously created and stored policy sub-classes (which essentially represent policy sub-templates), as will be described below in connection withFIG. 28. In another embodiment,policy sub-class field1720 may be replaced with a policy template option/field through which the user enters or selects a policy template name, in which case the above described creation process directly results in the creation of a policy template, i.e., the process of creating policy sub-classes is skipped.
• With reference toFIG. 28, there is an illustration of aGUI2800 displayed bymanagement entity110 and that may be used to create a security policy template (referred to simply as a policy template) based on previously created policy sub-classes or sub-templates.GUI2800 displays selectable previously createdpolicy sub-classes2805 in a sub-class display box. In the example ofFIG. 28,GUI2800 displays policy sub-class “Branch Allow Web traffic” created previously in the manner described above in connection withFIG. 27, along with other previously created policy sub-classes “Branch Gmail Block,” “Allow DNS,” and “Block Social Media.” Each ofpolicy sub-classes2805 contains combined NSP classifications (and the rules therein). In the example ofFIG. 28, the user has selected three of the previously created policy sub-classes, e.g., “Branch Allow Web traffic,” “Branch Gmail Block,” and “Block Social Media.”
• GUI2800 also displays a policy template field/option2810 through which a user may either select an existing policy template name through a drop down menu or add/enter a new policy template name. In the example ofFIG. 28, the user has selected/entered “Branch Policy” as a policy template name to which the three selected policy sub-classes are to be assigned. In response,management entity110 assigns the three selected policy sub-classes to the policy template name “Branch Policy.” These action creates a new policy template named “Branch Policy” that combines all of the network security policies (and rules) of the three selected policy sub-classes. The new policy template “Branch Policy” may be conveniently used to commit new security policies similar to previous security policies (i.e., the security policies in the policy template “Branch Policy”) to security devices in a new deployment site to be managed bymanagement entity110, as will be described below in connection withFIG. 29-32.
• With reference toFIG. 29, there is an illustration of anexample network environment2900 including existing geographically distributed data centers or branch sites2905(1)-2905(5) (referred to simply as “branches”) operating under control ofmanagement entity110.Management entity110 has previously deployed similar security policies across all of existingbranches2905.
• Each of branches2905(1)-2905(5) includes corresponding switches2906(1)-2906(5) connecting corresponding assets2908(1)-2908(5) (e.g., Boston Assets, Chicago Assets, Dallas Assets, Atlanta Assets, and Headquarter (HQ) assets) to corresponding ones of network security devices130(1)-130(5). Each security device130(i) controls access to and between corresponding assets2908(i) and other resources in the corresponding branch2905(i), and access to and from anetwork2909, such as the Internet.
• A geographically distributed new branch2910 (“San Jose”) needs to be added to existingbranches2905 and needs to operate under network security policies similar to those deployed to the existingbranches2905. Similar to existingbranches2905,new branch2910 includes a corresponding switch2906(6) and a corresponding security device130(6) to control access to and between corresponding assets2908(6) and other resources based on security policies available to the security device130(6). As described below,management entity110 may use policy templates to deploy the existing similar network security policies already in place and used acrossbranches2905 to security device130(6) innew branch2910.
• With reference toFIG. 30, there is an illustration of aGUI3000 displayed bymanagement entity110 and that may be used to create a new network security policy based on one or more previously created policy templates.GUI3000 includes a “select policy” field/option3005 through which the user may either select or enter a name of a previously created policy template, and an “add or select labels” field/option3010 through which the user may either select or enter a name of a new branch to which the selected/entered policy template is to be assigned. In the example ofFIG. 30, the user has selected the “Branch Policy” policy template for assignment to the new site “San Jose” (branch2910 inFIG. 29). In response,management entity110 assigns the “Branch Policy” policy template to the label “San Jose” associated with the new branch. Once the “Branch Policy” policy template has been assigned, it becomes a new security policy that may be committed (i.e., downloaded or deployed) to the new branch “San Jose.” An underlying assumption here is thatmanagement system110 has knowledge of a URL or other connection identity associated with security device130(6) at the new branch “San Jose” (seeFIG. 29) so that the management system is able to communicate with the new branch.
• GUI3000 also includes a selectable “Validate” icon that, when selected by the user, causesmanagement system110 to display another GUI that presents all of the network security rules of the selected policy template (e.g., the “Branch Policy” policy template) in editable form, as seen inFIG. 31, described below.
• With reference toFIG. 31, there is an illustration of aGUI3100 displayed bymanagement system110 in response to a user selection of the “Validate” icon inGUI3000.GUI3100 presents in an editable form all of the security rules included in the selected policy template “Branch Policy,” so that the user may review and edit the “Branch Policy.” To this end,GUI3100 also includes edit icons3110 (e.g., add rule and delete rule icons) through which the user may select to edit the presented network security rules.
• With reference toFIG. 32, there is an illustration ofGUI3100 displayed bymanagement system110 after the user has selected the add rule icon and also inserted anew rule3205 that allows “Marketing Users” to access “Facebook” application only for the “San Jose” branch. In response,management system110 updates the new “Branch Policy” to include the added rule.
• Once the new “Branch Policy” has been edited to include the added rule, the user may return toGUI3000 ofFIG. 30, and select the “Commit” icon. In response,management system110 commits (or downloads) the new “Branch Policy” to the security device130(6) at the “San Jose” branch (FIG. 29), which will implement the downloaded policy at that branch. Because the new “Branch Policy” encompasses security rules classified as identical across existingbranches2905, the new branch “San Jose” implements security policies similar, and in some cases identical, to those security policies that the existingbranches2905 implement. The new rule added in connection withGUI3100 represents an exception to the similar security rules that pertains substantially only to the “San Jose” branch.
• With reference toFIG. 33, there is a flowchart of anexample method3300 of creating and deploying security policies based on policy templates, performed by amanagement entity110 through interactions with a user. Initially, and as described above in detail,management entity110 generates NSP classifications. To do this,management entity110 connects with security devices and imports corresponding security policies from the security devices. Each imported policy includes corresponding security rules.Management entity110 classifies the imported security policies across the security devices into identical, similar, unique, and investigate classifications. As used herein, the terms “GUI” and “menu” are synonymous and may be used interchangeably.
• At3305,management entity110 displays selectable ones of the NSP classifications that identify security policies having shared or common (e.g., identical) security rules (see, e.g.,GUI1700 ofFIG. 17).
• Operations3310-3325 described below represent an example by which a new policy template may be created based on selected NSP classifications.
• At3310,management entity110 displays an option to select/enter a name of a policy template (see, e.g.,GUI2800 ofFIG. 28).
• At3315,management entity110 receives an entered/selected policy template name (see, e.g., GUI2800) and selections of the NSP classifications (see, e.g.,GUI1700 ofFIG. 27). The selections of the NSP classifications may be through policy sub-class or sub-template creation menus (see, e.g.,GUI1700 ofFIG. 28). Sub-classes/sub-templates combine NSP classifications prior to template creation inoperation3320.
• At3320,management entity110 creates a new policy template identified by the entered/selected policy template name and that includes all of the security policies identified by the selected NSP classifications (see, e.g., GUI2800). As mentioned above, the policy template may include one or more previously created policy sub-classes/sub-templates that each combine multiple selected ones of the selectable NSP classifications. The user may create the policy sub-classes/sub-templates in a precursor operation to operation3320 (see, e.g.,GUI1700 ofFIG. 27 and GUI2800).
• At3325,management entity110 displays options to enter/select (i) labels associated with or identifying security devices, and (ii) names of previously created policy templates, such as the new policy template (see, e.g.,GUI3000 ofFIG. 30).
• Operations3330-3345 described below represent an example by which a new user policy may be created based on the new policy template.
• At3330,management entity110 receives entries/selections of a label and one or more of the previously created policy templates and, responsive thereto, creates a new security policy based on the security policies in the selected policy template(s) (see, e.g., GUI3000).
• At3335,management entity110 displays a menu which shows editable network security rules included in the new security policy (see, e.g.,GUI3100 ofFIG. 31).
• At3340,management entity110 receives modifications of the network security rules through the menu (see, e.g.,GUI3100 ofFIG. 32).
• At3345,management entity110 updates the new security policy to include the modified/edited network security rules.
• At3350,management entity110 deploys or applies the new security policy to the security device associated with the selected/entered label.
Generalized Security Policy User Interface
• Generating and deploying security policy can be a laborious process. The Policy UI is a graphical user interface tool that allows a user to set network policy using relatively simple graphical actions.
• Referring toFIG. 34, a representation is shown of a graphical userinterface screen display5000 showing a visualization of a network topology. In this view, each cloud icon/graphical element represents a network domain. For example, there arecloud icons5010,5012,5014,5016 and5018 each representing a location, network or level of security, with names as shown in the figure. The outline color or other characteristic of a cloud icon may be used to denote a security risk.
• Connectivity between the network domains is represented by arrow graphical elements, in some cases through a firewall icon. For example,arrow icon5020 andfirewall icon5030 illustrate connectivity between theinternal network5010 and theinternet5018.Arrow icon5022 andfirewall icon5032 illustrate connectivity betweeninternal network5010 andcloud5016.Arrow icon5024 andfirewall icon5034 illustrate connectivity betweenDMZ5014 andcloud5016. A DMZ or demilitarized zone (sometimes referred to as a perimeter network) is a physical or logical subnetwork that contains and exposes an organization's external-facing services to a larger and untrusted network, usually the Internet.Arrow icon5026 andfirewall icon5036 illustrate connectivity betweeninternal network5010 andDMZ5016.Arrow icon5028 andfirewall icon5038 illustrate connectivity betweenDMZ5014 andinternet5018.Arrow icon5029 andfirewall icon5039 illustrate connectivity betweensecure network5012 andDMZ5014. Finally,arrow icon5040 illustrates connectivity between theinternet5018 and someunknown network5018. The outline color or other characteristic of a connectivity arrow may be used to denote a type or level of a security risk.
• An arrow denotes a connection. Some arrows may state an explicit “allow” or “block” policy, while outlines may indicate default connection type. For example, a circle may be created indicating an explicit policy to block all but the arrow. In another example, a user may define that in a particular zone, traffic between one zone and another zone is always allowed unless indicated otherwise by an arrow. Color schemes will typically indicate security levels.
• When a user selects (clicks or double clicks on) a cloud icon shown inFIG. 34, another display screen is presented that shows more details of the security topology for that network domain. An example of such a display screen is shown inFIG. 37, described below. When a user double-clicks on an icon, e.g., an icon for a network security device, status information about that device is displayed.
• On one side of thedisplay screen5000 is a space allocated to display icons representing actors and resources. This may be called, for example, the “Policy Depot”. For example,icon5050 represents a user,icon5052 represents a group of users of a particular type,icon5054 represents a device,icon5056 represents a network andicon5058 represents an application. These icons are referred to below in connection with a description of subsequent figures.
• In addition, when rules from the “Policy Depot” may be dragged and dropped onto a relevant object in the network view in order to deploy the security policy rules onto the that network security device. The network is abstracted away as much as possible focusing on perimeter, inside and out, on zones.
• Turning now toFIG. 35, a graphicaluser interface screen6000 is shown in which the icons depicted inFIG. 34 are overlaid on ageographical map6010 to indicate the locations of each of the networks or devices in the network. For example, inFIG. 35, there is a shown a cloudicon resource icon6020 in the western U.S, aresource icon6030 in western Europe, aresource icon6040 in South Africa, and aresource icon6050 in China, as well ascloud icon6100 for a network domain. In addition, arrow and firewall icons are shown to indicate the connectivity between resources and network domains, as described above in connection withFIG. 34.
• Referring now toFIG. 36, still another example of a graphicaluser interface screen7000 is shown. In this example different network zones are shown, such asnetwork zones7010,7020 and7030. Icons are shown to indicate resources within a given network zone. For example,network zone7010 includescloud icon7012, application icons7014(1),7014(2) and7014(3) and anapplication group icon7016. Similarly,network zone7020 includescloud icon7022 fornetwork1,cloud icon7024 fornetwork2, application icons7026(1),7026(2) and7026(3) and anapplication group icon7028. Arrow and firewall icons are shown between elements inFIG. 35 in a similar manner to that described above in connection withFIG. 34.
• FIG. 37 illustrates still another example graphicaluser interface screen8000. In this example user interface screen, a plurality of icons are displayed. Each icon represents an actor or a resource in a networking environment. Network security policy is defined by receiving user input in the form of lines drawn between icons representing actors and resources to control abilities between actors and resources.
• For example, inFIG. 37,icons8010 and8020 represent actors, that is, users or a class of users in a corporation or enterprise.Icon8010 represents a finance or accountant executive andicon8020 represents an engineer.Icon8010 is located within a region orspace8015 that is intended to represent a corporate network zone.Icon8020 is located within a region orspace8025 that is intended an network zone external to the corporate network.
• Icons8030,8040 and8050 represent resources, such as applications, databases, etc.Icon8030 represents a database system and is shown withinspace8015 because the database system is internal to the corporate network.Icons8040 and8050 are inspace8025 because they are external to the corporate network. As an example,icon8040 represents the website resources of a business process service provider (software-as-a-service provider) that is frequently used by personnel in the corporate network. Thus, there is an ongoing business relationship between the business process service provider and actors in the corporate network.Icon8050 represents website resources of an investment service provider who does not have an ongoing business relationship with the corporation.
• A network administrator may utilize arrows in theuser interface display8000 between actors and resources in order to define network security policy that permits, denies or controls abilities between actors and resources. Characteristics of an arrow, such as the color of the arrow, whether the arrow is dashed or solid, and the weight/thickness of the arrow, etc., may be used to further define network policy.
• For example,arrow8100 may be drawn betweenicon8010 and theicon8030. By drawing this arrow8100 (using any arrow drawing tool in a graphical user interface tool set) in the manner shown inFIG. 37, the network security policy is set to allow the finance/accountant executive to access the database system. Again, this is within the corporate network. If no arrow is drawn between theicon8010 andicon8030, then the indication is that no network policy has been set to permit the finance/accountant executive to access the database system. Drawing of thearrow8100 is translated by processing resources of the management entity (FIG. 1) to appropriate rules that are delivered to the one or more security devices that are in the path(s) of traffic flow between the finance/accountant executive and the database system.
• Similarly, by drawing anarrow8200 between theicon8020 and theicon8030, a network security policy is defined to enable the engineer, who is outside the corporate network, to access the database system inside the corporate network.
• By drawing anarrow8300 between theicon8010 and theicon8040, a security policy is defined to enable the finance/accountant executive who is inside the corporate network to access capabilities of the business process service provider who is outside the corporate network. Moreover, it may be desired, in defining this policy, to further require that the traffic between the actor and resource is monitored for indications of network security breaches. To achieve this, the color of thearrow8300 may be set to a particular color (e.g., blue) to indicate that monitoring of traffic and reporting to a network administrator or network management entity will occur for traffic between the actor and resource. If the arrow is drawn with the characteristic to set the policy to monitor traffic between an actor and a resource, a notification may be sent to the actor so that the actor is aware that the traffic will be monitored for indications of security breaches. Moreover, if a security breach is detected from the monitoring of the traffic, an alert notification is sent to a network administrator or other destination.
• Arrow8400 drawn betweenicon8010 andicon8050 results in a security policy being defined to enable the finance/accountant executive who is inside the corporate network to access capabilities of the investment service provider, which is outside of the corporate network. Since the investment service provider does not have a business relationship with the corporation, a characteristic of thearrow8400 may be chosen (e.g., color green or dashed line) to set a network security policy by which the traffic between the actor and resource is ignored, meaning it is not monitored for security breaches. The actor is “on its own”. For example, the actor may be connecting to the resource outside the corporate network for his/her own personal financial reasons, and not for official business of the corporation.
• It is also possible to expressly deny abilities between an actor and a resource by drawing a line of a particular color (e.g., red) or characteristic (blinking).
• To reiterate, the user interface techniques depicted inFIG. 37 involve drawing arrows between icons that represent actors and resources to set network policy in a networking environment. Characteristics of the arrows, such as color, type, line weight, etc., may be used to further define network policy. Drawing of arrows results in generation of one or more rules that are delivered to security devices to achieve the desired policy associated with the arrow. This in turn involves converting between one or more rules of a generic network policy model to one or more rules of a native rule set associated with one or more security devices, as described above in the section under the heading “Security Policy Unification.” While these examples are described in terms of drawing an arrow between icons, it is not meant to be limiting. it is to be understood that any other suitable directional/positional relationship may be represented between icons to achieve the same function as drawing an arrow.
• Reference is now made toFIG. 38.FIG. 38 shows a flow chart for a graphicaluser interface process9000 to set security policy in accordance with an example embodiment. At9010, a plurality of icons are displayed. Each icon represents an actor or a resource in the networking environment. Delineated regions may be displayed in which icons may reside to indicate whether an actor or a resource within or outside a network zone.
• At9020, security policy is defined by receiving user input in the form of lines drawn between icons representing actors and resources to control abilities (e.g., permit access/connectivity, deny access/connectivity, monitor traffic, etc.) between actors and resources.
• In other words, and as shown inFIG. 37, atstep9020, security policy is defined by receiving user input in the form of a line drawn between a first icon representing an actor and a second icon representing a resource to control abilities between the actor and the resource. The first icon is an arbitrary icon representing an actor and the second icon is an arbitrary icon representing a resource.
• An arrow on a line between the first icon representing the actor and the second icon representing the resource determines whether access between the actor and the resource is permitted. A line drawn or not drawn between the first icon representing the actor and the second icon representing the resource determines whether to allow or block abilities between the actor and the resource. A characteristic of the line between the first icon and the second icon further indicates whether traffic between the actor and the resource is to be monitored. As explained above, the characteristic may be color. Moreover, an icon may represent a group of a plurality of actors of a particular type within the networking environment. Also as shown inFIG. 37, delineated regions may be displayed in which the plurality of icons may reside to indicate whether an actor or a resource is located within or outside a network zone.
• Defining security policy thus involves interpreting between the first icon representing the actor and the second icon representing the resource so as to generate one or more security rules for configuring one or more security devices in a path between the actor and the resource. Interpreting a line drawn may involve generating the one or more security rules in accordance with a generic policy model that is applicable across a plurality of security device types. In this case, an additional step is performed when configuring the security devices, of converting the one or more security rules in accordance with the generic policy model to one or more rules of a native rule set associated with one or more security devices. In summary, a graphical user interface that allows a user to set security policy using simple graphical actions.
• Techniques described herein can be embodied in method, apparatus, non-transitory tangible computer readable, and system forms.
• In summary, in one form, a method is provided comprising: at a management entity: connecting with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies; importing, over the network, data describing the security policies from the multiple security devices; and classifying the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.
• In another form, an apparatus is provided comprising: a network interface unit to connect with a network; and a processor coupled to the network interface unit to: connect with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies; import, over the network, data describing the security policies from the multiple security devices; and classify the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.
• In yet another form, a non-transitory tangible computer readable storage media encoded with instructions is provided. The instructions, when executed by a processor, cause the processor to: connect with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies; import, over the network, data describing the security policies from the multiple security devices; and classify the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.
• The above description is intended by way of example only. Various modifications and structural changes may be made therein without departing from the scope of the concepts described herein and within the scope and range of equivalents of the claims.

Claims (25)

What is claimed is:

1. A method comprising:

at a management entity:

connecting with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies;

importing, over the network, data describing the security policies from the multiple security devices; and

classifying the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.

2. The method ofclaim 1, wherein:

the importing includes importing each security policy as one or more security rules, each security rule including a set of rule parameters configured to permit or deny access to a resource based on a network protocol, source and destination addresses, and a device port; and

the classifying includes classifying the security policies into the security policy classifications based on commonality between the security rules across the multiple security devices.

3. The method ofclaim 2, wherein the classifying includes:

classifying security policies which have identical security rules into one or more identical security policy classifications;

classifying security policies which have similar security rules into one or more similar security policy classifications; and

classifying security policies which have unique security rules into a unique security policy classification.

4. The method ofclaim 3, further comprising:

classifying security policies not already classified into a security policy classification indicating further investigation is needed for certain security policies; and

generating an alert notification indicating that one or more security policies need further investigation.

5. The method ofclaim 3, further comprising comparing the rule parameters of each rule of each security policy across the security policies, wherein the classifying further includes classifying based on results of the comparing by:

classifying the security policies into the identical security policy classification if all of their associated rule parameters are equivalent to each other;

classifying the security policies into the similar security policy classification if only some of their associated rule parameters are equivalent to each other; and

classifying the security policies into a unique security policy classification if none of the associated rule parameters are equivalent to each other.

6. The method ofclaim 3, further comprising displaying a list of the rule parameters for each of the security policy classifications.

7. The method ofclaim 3, further comprising:

displaying the security policy classifications as selectable security policy classifications;

displaying a policy template naming option through which a policy template name may be entered;

receiving an entered policy template name and selections of multiple security policy classifications; and

assigning all of the security policies in the multiple selected security policy classifications to a security policy template having the entered policy template name.

8. The method ofclaim 2, further comprising:

displaying the security policy classifications in unexpanded views;

displaying the an expand option associated with each of the displayed security policy classifications;

receiving a selection of one of the expand options; and

displaying the security policy classification associated with the selected expand option in an expanded view that exposes device names of all of the security devices associated with that security policy classification.

9. The method ofclaim 8, wherein:

the displaying further includes displaying a filter option to specify a rule parameter associated with the security policy classifications;

receiving a specified rule parameter through the filter option; and

displaying all of the rules in each security policy classifications that include a rule parameter that matches the specified rule parameter.

10. An apparatus comprising:

a network interface unit to connect with a network; and

a processor coupled to the network interface unit to:

connect with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies;

import, over the network, data describing the security policies from the multiple security devices; and

classify the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.

11. The apparatus ofclaim 10, wherein:

the processor imports by importing each security policy as one or more security rules, each security rule including a set of rule parameters configured to permit or deny access to a resource based on a network protocol, source and destination addresses, and a device port; and

the processor classifies by classifying the security policies into the security policy classifications based on commonality between the security rules across the security devices.

12. The apparatus ofclaim 11, wherein the processor classifies by:

classifying security policies which have identical security rules into one or more identical security policy classifications;

classifying security policies which have similar security rules into one or more similar security policy classifications; and

classifying security policies which have unique security rules into a unique security policy classification.

13. The apparatus ofclaim 12, wherein the processor further:

classifies security policies not already classified into a security policy classification indicating further investigation is needed for certain security policies; and

generates an alert notification indicating that one or more security policies need further investigation.

14. The apparatus ofclaim 12, wherein the processor further compares the rule parameters of each rule of each security policy across the security policies, and classifies further based on results of the compare by:

classifying the security policies into the identical security policy classification if all of their associated rule parameters are equivalent to each other;

classifying the security policies into the similar security policy classification if only some of their associated rule parameters are equivalent to each other; and

classifying security policies into the unique security policy classification if none of the associated rule parameters are equivalent to each other.

15. The apparatus ofclaim 12, wherein the processor generates for display a list of the rule parameters for each of the security policy classifications.

16. The apparatus ofclaim 11, wherein the processor further:

generates for display the security policy classifications in unexpanded views;

generates for display an expand option associated with each of the displayed security policy classifications;

receives a selection of one of the expand options; and

generates for display the security policy classification associated with the selected expand option in an expanded view that exposes device names of all of the security devices associated with that security policy classification.

17. The apparatus ofclaim 16, wherein the processor further:

generates for display a filter option to specify a rule parameter associated with the security policy classifications;

receives a specified rule parameter through the filter option; and

generates for display all of the rules in each security policy classifications that include a rule parameter that matches the specified rule parameter.

18. A non-transitory tangible computer readable storage media encoded with instructions that, when executed by a processor, cause the processor to:

connect with multiple security devices across a network, each security device configured to operate in accordance with one or more security policies;

import, over the network, data describing the security policies from the multiple security devices; and

classify the imported security policies into security policy classifications based on commonality in information included in the security policies across the multiple security devices.

19. The computer readable storage media ofclaim 18, wherein:

the instructions to cause the processor to import include instructions to cause the processor to import each security policy as one or more security rules, each security rule including a set of rule parameters configured to permit or deny access to a resource based on a network protocol, source and destination addresses, and a device port; and

the instructions to cause the processor to classify include instructions to cause the processor to classify the security policies into the security policy classifications based on commonality between the security rules across the security devices.

20. The computer readable storage media ofclaim 19, wherein the instructions to cause the processor to classify include instructions to cause the processor to:

classify security policies which have identical security rules into one or more identical security policy classifications;

classify security policies which have similar security rules into one or more similar security policy classifications; and

classify security policies which have unique security rules into a unique security policy classification.

21. The computer readable storage media ofclaim 20, further comprising instructions to cause the processor to:

classify security policies not already classified into a security policy classification indicating further investigation is needed for certain security policies; and

generate an alert notification indicating that one or more security policies need further investigation.

22. The computer readable storage media ofclaim 20, further comprising instructions to cause the processor to compare the rule parameters of each rule of each security policy across the security policies, wherein the instructions to cause the processor to classify include further instructions to cause the processor to, based on results of the compare:

classify the security policies into the identical security policy classification if all of their associated rule parameters are equivalent to each other;

classify the security policies into the similar security policy classification if only some of their associated rule parameters are equivalent to each other; and

classify security policies into the unique security policy classification if none of the associated rule parameters are equivalent to each other.

23. The computer readable storage media ofclaim 20, further comprising instructions to cause the processor to generate for display a list of the rule parameters for each of the security policy classifications.

24. The computer readable storage media ofclaim 19, further comprising instructions to cause the processor to:

generate for display the security policy classifications in unexpanded views;

generate for display an expand option associated with each of the displayed security policy classifications;

receive a selection of one of the expand options; and

generate for display the security policy classification associated with the selected expand option in an expanded view that exposes device names of all of the security devices associated with that security policy classification.

25. The computer readable storage media ofclaim 24, further comprising instructions to cause the processor to:

generate for display a filter option to specify a rule parameter associated with the security policy classifications;

receive a specified rule parameter through the filter option; and

generate for display all of the rules in each security policy classifications that include a rule parameter that matches the specified rule parameter.

Images