System and method for anomaly detection in distributed cloud environment
By deploying controllers and gateways within the cloud computing platform, the distributed cloud computing system is built, troubleshooting and monitoring problems in multi-cloud network environments are solved, visual management and abnormal detection of cross-cloud resources are realized, and network security and capability are improved. Operative.
Patent Information
- Application Number
- CN202380032896.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-03-15
- Filing Date
- 2023-02-08
- Publication Date
- 2024-11-15
AI Technical Summary
In multi-cloud network environments, network administrators have difficulty effectively troubleshooting and monitoring network connectivity, lack of visualization tools to quickly detect network attacks or abnormal behavior, and traditional systems cannot provide a complete view of topological mapping and resource state across multiple clouds.
By deploying controllers and gateways within the cloud computing platform, building a distributed cloud computing system, using topology system logic to collect and store connection and status information of multiple cloud resources, providing a visual platform to display topology mapping and abnormal detection results, achieving cross-border Cloud visibility and monitoring.
It realizes visual management of multi-cloud network resources and connection status, improves troubleshooting and monitoring efficiency, can quickly detect abnormal behaviors and network attacks, and enhances network security and operability.
Smart Images

Figure CN118974703A_ABST
Abstract
Description
[0001] CROSS-REFERENCE TO RELATED APPLICATIONS
[0002] This application claims priority to U.S. Provisional Patent Application No. 63 / 308,038 filed on February 8, 2022 and U.S. Provisional Patent Application No. 63 / 320,019 filed on March 15, 2022, and incorporates the entire disclosures of each by reference. Technical Field
[0003] Embodiments of the present disclosure relate to the field of network behavior analysis and are directed to resources deployed in one or more cloud computing environments. More specifically, embodiments of the present disclosure relate to a method for generating a network behavior baseline for a specific virtual private cloud during a learning period, the baseline covering multiple metrics, and then analyzing real-time network traffic based on the baseline to detect the presence of abnormal behavior. Background Art
[0004] This section provides background information to facilitate a better understanding of the various aspects of the present disclosure. It should be understood that the statements in this section of this document should be read in this light, and not as admissions of prior art.
[0005] Until recently, enterprises have relied on application software installed on one or more electronic devices that are located near users (hereinafter referred to as "on-premises electronic devices"). For example, these on-premises electronic devices may correspond to endpoint devices (e.g., personal computers, cellular smartphones, netbooks, etc.), locally maintained mainframes, or even local servers. Depending on the size of the enterprise, the purchase of on-premises electronic devices and their corresponding software requires a large upfront capital expenditure, as well as a large ongoing operating cost to maintain the operability of these on-premises electronic devices. These operating costs may include the costs of deploying, managing, maintaining, upgrading, repairing, and replacing these electronic devices.
[0006] Recently, more and more businesses and individuals have begun to rely on public cloud networks (hereinafter referred to as "public clouds") to provide users with a variety of services ranging from word processing application functions to network management. A "public cloud" is a fully virtualized environment with a multi-tenant architecture that provides tenants (i.e., users) with the ability to share computing and storage resources while maintaining data isolation within each user's cloud account. The virtualized environment includes an on-demand cloud computing platform provided by a collection of physical data centers, each of which includes a large number of servers hosted by a cloud provider. Examples of different types of public cloud networks may include, but are not limited to, or are limited to, for example, AMAZON WEB GOOGLECLOUD PLATFORM TM OR ORACLE CLOUD TM .
[0007] This growing reliance on public cloud networks is due in large part to the substantial cost-saving advantages that this particular deployment provides. However, for many types of services (e.g., network management), network administrators face many challenges when, for example, business operations depend on the operability of a single public cloud or the operability of multiple public cloud networks. For example, in situations where an enterprise deploys a network that depends on multiple public cloud networks (hereinafter referred to as a "multi-cloud network"), network administrators cannot effectively troubleshoot connectivity issues that occur within the multi-cloud network. One reason for this ineffective troubleshooting is that administrators or users do not have traditional solutions to visualize the connectivity of their multi-cloud network deployments. Another reason is that cloud network providers only allow users to access a limited number of constructs, thereby controlling the type and amount of network information that users can access. Therefore, the type or amount of network information is rarely sufficient for administrators or users to quickly and effectively troubleshoot and correct network connectivity issues.
[0008] Likewise, there are no conventional solutions to visually monitor traffic exchanges between network devices in different public cloud networks (multi-cloud networks) and retain state information associated with network devices in the multi-cloud network to more quickly detect operational anomalies that may suggest that a cyber attack is ongoing or that the health of the multi-cloud network is compromised. Summary of the invention
[0009] In various embodiments, aspects of the present disclosure relate to a distributed cloud computing system, including: a controller configured to deploy a first virtual private cloud (VPC) in a first cloud computing network, deploy a first gateway in the first VPC, deploy a second VPC in a second cloud computing network, and deploy a second gateway in the second VPC, and wherein a first subset of a plurality of constructs is associated with the first gateway and deployed in the first cloud computing network, and a second subset of the plurality of constructs is associated with the second gateway and deployed in the second cloud computing network; and logic on a non-transitory computer medium that, when executed by one or more processors, causes performance of various operations.
[0010] This Summary is provided to introduce some concepts that are further described below in the Detailed Description. This Summary is not intended to identify key or essential features of the claimed subject matter, nor is it intended to be used to help limit the scope of the claimed subject matter. BRIEF DESCRIPTION OF THE DRAWINGS
[0011] Embodiments of the present disclosure are illustrated by way of example and not limitation in the several figures of the accompanying drawings in which like reference numerals indicate similar elements and in which:
[0012] Figure 1is a diagram of an exemplary embodiment of a distributed cloud computing system including a controller that manages fabrics across multiple cloud networks, according to some embodiments;
[0013] Figure 2A is an exemplary diagram of a logical representation of a controller deployed within a cloud computing platform according to some embodiments;
[0014] Figure 2B is an exemplary illustration of a logical representation of topology system logic deployed within a cloud computing platform according to some embodiments;
[0015] Figures 3A-3C is an interface screen showing a portion of a dashboard of a visualization platform according to some embodiments, the interface screen being intended to show information related to network traffic and constructs within a cloud computing environment;
[0016] Figure 4A is an interface screen displaying portions of a dashboard of a visualization platform according to some embodiments, wherein each portion is configured to show information obtained or determined by a topology system;
[0017] Figure 4B According to some embodiments, the display includes information about Figure 4A A section of the interface screen that discusses the threat details;
[0018] Figure 4C is an interface screen displaying a network topology map showing a compromised gateway in a highlighted capacity according to some embodiments;
[0019] Figure 5A is a graphical representation of functionality associated with a broader security platform according to some embodiments;
[0020] Figure 5B is an interface display of analysis results related to network behavior analysis according to some embodiments;
[0021] Fig. 6A is a graphical representation of an interface screen according to some embodiments, the interface screen displaying a portion of a dashboard of a visualization platform intended to illustrate information related to anomaly detection within network traffic of a cloud computing environment;
[0022] Figure 6B is a graphical representation of an interface screen displaying data regarding a monitored VPC according to some embodiments; and
[0023] Figure 6C is a graphical representation of an interface screen configured to receive user input indicating instructions for managing monitoring of a specified VPC in accordance with some embodiments. DETAILED DESCRIPTION
[0024] I. Terminology
[0025] In the following description, certain terms are used to describe features of the present invention. In some cases, the terms "logic" and "component" represent hardware, firmware and / or software configured to perform one or more functions. As hardware, the logic or component may include circuits with data processing or storage functions. Examples of such circuits may include, but are not limited to, or be limited to, a microprocessor, one or more processor cores, a programmable gate array, a microcontroller, an application specific integrated circuit, a wireless receiver, a transmitter and / or transceiver circuit, a semiconductor memory, or a combinational logic.
[0026] Alternatively, or in combination with the above-mentioned hardware circuits, the logic or component can be software in the form of one or more software modules. A software module may include an executable application, an application programming interface (API), a subroutine, a function, a process, a small application program, a small service program, a routine, a source code, a shared library / dynamically loaded library, or one or more instructions. A software module may be stored in any type of suitable non-transitory storage medium or temporary storage medium (e.g., electrical, optical, acoustic or other forms of propagation signals, such as carrier waves, infrared signals or digital signals). Examples of non-transitory storage media may include, but are not limited to, programmable circuits; semiconductor memories; non-permanent storage devices, such as volatile memory (e.g., any type of random access memory "RAM"); permanent storage devices, such as non-volatile memory (e.g., read-only memory "ROM", power-supported RAM, flash memory, phase change memory, etc.), solid-state drives, hard drives, optical disk drives, or portable memory devices. As firmware, executable code may be stored in a permanent storage device.
[0027] The term "computerized" generally means that any corresponding operations are performed by hardware in combination with software and / or firmware.
[0028] The term "host" can be interpreted as a virtual or physical logic. For example, as an illustrative example, a host can correspond to a virtual logic in the form of a software component (e.g., a virtual machine) that is assigned a hardware address (e.g., a MAC address) and an IP address within the range of IP addresses supported by a specific IP subnet. Alternatively, in some embodiments, a host can correspond to a physical logic, such as an electronic device that is communicatively coupled to a network and assigned a hardware (MAC) address and an IP address. Examples of electronic devices may include, but are not limited to, personal computers (e.g., desktop computers, laptops, tablet computers, or netbooks), mobile phones, stand-alone devices, sensors, servers, or information routing devices (e.g., routers, bridge routers ("brouter"), etc.). Here, the term "internal host" corresponds to a host that resides as part of an "internal" (or local) network, while a "cloud host" corresponds to a host that resides as part of a public cloud network.
[0029] The term "cloud computing infrastructure" generally refers to a networked combination of hardware and software, including one or more servers, each of which includes circuitry for managing network resources, such as additional servers and computing devices. Cloud computing infrastructure also includes one or more communication interfaces and communication interface logic.
[0030] The term "gateway" may refer to a software instance deployed within a VPC that controls the flow of data traffic from the VPC to one or more remote sites that include computing devices that can process, store, and / or continue data routing. The terms "transit gateway" and "branch gateway" may refer to gateways that have similar architectures but are identified differently based on their location / configuration in a cloud computing platform. For example, a "branch" gateway is configured to interact with a target instance, while a "hub" gateway is configured to further assist in the propagation of data traffic (e.g., one or more messages) directed to a branch gateway within a branch VPC or computing devices within an internal network.
[0031] The term "controller" may refer to a software instance deployed within a cloud computing platform that manages certain aspects of the operability of the cloud computing platform. For example, the controller collects information about each VPC and configures the VPC routing table associated with each VPC to establish a communication link (e.g., a logical connection) between a branch gateway and a cloud instance associated with a particular instance subnet. The VPC routing table is programmed to support communication links between different sources and destinations, such as internal computing devices, cloud instances within a particular instance subnet, and the like. In addition, the controller establishes each gateway instance and manages the operability of the gateway by, for example, configuring the gateway routing table of each gateway within each VPC. In addition, the controller can manage the establishment of a secure communication link (e.g., an IPSec tunnel) between each branch VPC and the hub VPC deployed within the cloud computing platform.
[0032] The term "message" generally refers to information having a specified format and transmitted according to an appropriate delivery protocol. Thus, each message may be in the form of one or more packets, frames, or any other bit sequence having a specified format.
[0033] The term "transmission medium" can be interpreted as a physical or logical communication path between two or more electronic devices. For example, as a physical communication path, wires, optical fibers, cables, bus traces, or wireless channels using infrared, radio frequency (RF) can be used. Wired and / or wireless interconnection.
[0034] Finally, the terms "or" and "and / or" as used herein should be interpreted as inclusive, or meaning any one or any combination. As an example, "A, B, or C" or "A, B, and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B, and C". An exception to this definition will only occur if the combination of elements, functions, steps, or actions are inherently mutually exclusive in some way.
[0035] As the invention is susceptible of embodiment to many different forms, it is intended that this disclosure be considered as an exemplification of the principles of the invention and it is not intended that the invention be limited to the specific embodiments shown and described.
[0036] II. General Architecture
[0037] Embodiments of the present disclosure relate to a system configured to provide operational visibility for networking using one or more cloud computing environments (also referred to herein as "multi-cloud networking"). Some embodiments of the system include, for example, logic processed on a first computing resource such as a cloud computing resource, and one or more controllers. Such a system is referred to herein as a topology system.
[0038] As described above, a controller can be a software instance deployed within a cloud computing platform that manages certain aspects of the operability of the cloud computing platform. For example, the controller collects information about each VPC and configures the VPC routing table associated with each VPC to establish a communication link (e.g., a logical connection) between a branch gateway and a cloud instance associated with a particular instance subnet. The VPC routing table is programmed to support communication links between different sources and destinations, such as internal computing devices, cloud instances within a particular instance subnet, etc. Therefore, the controller obtains and stores information that reveals certain characteristics and communication links of resources managed by the controller (e.g., gateways and any subnets within the controller's authority).
[0039] Specifically, the topology system enables the collection and storage of such information across multiple cloud computing environments ("clouds"). By enabling the collection and storage of such information across multiple clouds, the topology system is configured to provide visualization of connections between resources of the multiple clouds and the status of resources and connections between resources of the multiple clouds. Furthermore, the topology system is configured to provide searchability of information detailing resource parameters, connections between resources, and the status of various resources, as well as visualization of search results.
[0040] Embodiments of the present disclosure provide many advantages over current systems that provide dashboards showing controller parameters because current systems do not provide the ability to visualize the connections between resources of multiple clouds and the status of the connections between resources of multiple clouds.
[0041] As an example, an enterprise network may span several clouds, and an administrator of the enterprise may want to see the status of all resources and connections between them on the enterprise network. However, because the enterprise network spans multiple clouds, current systems do not enable administrators to visualize outside of a single cloud. Therefore, by only obtaining a visual of a single cloud, administrators cannot obtain a complete view of resources, connections between them, and their respective states. As used herein, a visual display of resources, connections between them, and their respective states is referred to as a topology map. Current systems do not provide topology maps across multiple clouds. Not only do current systems fail to provide administrators with a complete topology map of the enterprise network, current systems also fail to allow administrators to search across multiple clouds or visualize how state changes in resources or connections in one cloud affect the state of resources or connections in a second cloud.
[0042] As will be discussed in further detail below, embodiments of the present disclosure relate to systems, methods, and apparatus for enabling administrators or other users to see a topology map of the entire enterprise network, even across multiple clouds. In addition, the visualization of the topology map can automatically change as resources or connection states change (e.g., a "dynamic topology map").
[0043] In one embodiment, a network can be deployed across multiple clouds using multiple controllers to manage resources (e.g., gateways) and network connections. In addition, the logic of the topology system can be stored and processed on a server device or cloud computing resource, and multiple controllers are queried for data about the network topology by transmitting one or more dedicated API calls to each controller to obtain specified data, which can be stored by each controller on one or more internal databases. The logic receives the requested data, generates a topology map, and generates one or more GUI screens to display the topology map to an administrator. The logic can be configured to receive user input, such as selection of one or more filters, and display a filtered data set including data across multiple clouds accordingly. In addition, the logic can be configured to receive user input such as a search term, and display a filtered data set including data across multiple clouds accordingly.
[0044] III. Exemplary Graphical User Interface
[0045] refer to Figure 1 , a diagram of an exemplary embodiment of a distributed cloud management system 100 is shown, wherein the cloud computing system features a controller 102 for managing constructs residing in multiple cloud networks (which may include "networked cloud environments" and, in some embodiments, may include the enterprise networks mentioned above), and a software instance 138 (hereinafter referred to as "topological system logic") that visualizes the managed constructs. More specifically, the controller 102 is configured to manage multiple constructs across multiple cloud networks, such as cloud (network) A 104 and cloud (network) B 106. In the exemplary illustration, cloud A 104 provides computing resources ("resources") for transit gateway 114 that communicates with gateways 1181-1182 associated with virtual networks (VNETs) 1161-1162. Cloud B 106 provides resources for transit gateway 120 that communicates with gateways 1241-1242 associated with virtual private clouds (VPCs) 1221-1222. Cloud B 106 also provides resources for a native transit hub 126 that communicates with VPCs 128 and 130. According to this embodiment of the present disclosure, Figure 1 As shown, the transit gateways 114, 120 and the native transit hub 126 are in communication with each other. Thus, it should be clearly understood that the controller 102 manages several constructs, such as the gateways shown, across multiple cloud networks.
[0046] Specifically, the first group of constructs 108 is deployed within cloud A 104, and the second and third group of constructs 110, 112 are deployed within cloud B 106. The controller 102 utilizes a set of APIs to provide instructions and receive data (state information) associated with each of these constructs and state information (link status) related to each connection between these constructs. The construct metadata returned by the construct may depend on the type of construct (e.g., region, VPC, gateway, subnet, instance within a VPC, etc.), where examples of construct metadata may include, but are not limited to, one or more of the following construct parameters (attributes): construct name, construct identifier, encryption enabled, properties of the VPC associated with the construct (e.g., VPC name, identifier and / or region, etc.), cloud properties in which the construct is deployed (e.g., cloud provider, cloud type, etc. on which the construct resides), or the like.
[0047] In addition, the cloud management system 100 includes a topology system logic 138 that is processed on the cloud computing resources 136. In some embodiments, the topology system logic 138 can be logic hosted on a user's Infrastructure as a Service (IaaS) cloud or multi-cloud environment. As an example, the topology system logic 138 can be launched as an instance within a public cloud network (e.g., as a In As an alternative example, the topology system logic 138 may be configured as For example, when booted, the topology system logic 138 is assigned a routable address, such as a static IP address.
[0048] As shown, the topology system logic 138 communicates with the controller 102 via, for example, an API that enables the topology system logic 138 to transmit queries to the controller 102 via one or more API calls. The topology system logic 138, when executed by the cloud computing resources 136, performs operations including querying the controller 102 to obtain construct metadata via API calls in response to specific events. The specific events may be based on periodic intervals or non-periodic intervals or triggering events (e.g., a user request for visualization via user input).
[0049] In some embodiments, in response to receiving a query from the topology system logic 138 via an API call, the controller 102 accesses data stored on or by the controller 102 and returns the requested data via the API to the topology system logic 138. For example, the topology system logic 138 may initiate one or more queries to the controller 102 to obtain topology information associated with the structures managed by the controller 102 (e.g., a list of all gateways managed by the controller 102, a list of all VPCs or VNETs managed by the controller 102, or other data collected from a database table) and status information associated with each structure as described above.
[0050] Upon receiving the requested construct metadata, the topology system logic 138 performs one or more analyses and determines whether any additional construct metadata needs to be requested. For example, the topology system logic 138 may provide a first query to the controller 102 requesting a list of all gateways managed by the controller 102. In response to receiving the requested construct metadata, the topology system logic 102 determines the interconnections between the listed gateways. Subsequently, the topology system logic 138 may provide a second query to the controller 102 requesting a list of all VPCs managed by the controller. In response to receiving the requested construct metadata, the topology system logic 138 determines an association between each VPC and a corresponding gateway.
[0051] For example, in some embodiments, the received construct metadata provides detailed information about each gateway, enabling the topology system logic 138 to generate a data object representing the gateway, such as a database table of construct metadata. The data objects representing multiple gateways are cross-referenced to establish a topology map based on the parameters of each gateway, which parameters may include, among others: cloud network user account name; cloud provider name; VPC name; gateway name; VPC region; sandbox IP address; gateway subnet identifier; gateway subnet CIDR; gateway zone; name of the associated cloud computing account; VPC identifier; VPC state; parent VPC name; VPC CIDR; etc. Similarly, the construct metadata is also used to generate data objects representing each VPC object and each subnet object.
[0052] Additionally, to determine whether a connection within the network is between two transit gateways, the topology system logic 138 may utilize a separate API call to query the controller 102 for a list of all transit gateways. Thus, the topology system logic 138 is then able to determine whether a connection between the first gateway and the second gateway is between the two transit gateways. In some embodiments, as will be discussed below, connections between transit gateways and connections between branch gateways and transits may be visually represented in two different ways.
[0053] In addition to receiving construct metadata from the controller 102, the topology system logic 138 may also receive network data from one or more gateways managed by the controller 102. For example, for each network packet, the network data may include, but is not limited to, the ingress interface, source IP address, destination IP address, IP protocol, UDP or TCP source port, UDP or TCP destination port, ICMP type and code, IP "service type", etc. In one embodiment, the network data may be transmitted from the gateway to the topology system logic 138 using an IP protocol (e.g., UDP). In some embodiments, the network data is collected and exported via the NetFlow network protocol.
[0054] To configure a gateway to transmit network data to topology system logic 138, topology system logic 138 may provide instructions to controller 102, which in turn provides instructions to each gateway managed by controller 102. These instructions provide the IP address of topology system logic 138, which is used as the IP address for addressing network data transmissions.
[0055] As will be discussed in detail below, the topology system logic 138 can generate a visualization platform including one or more interactive display screens. These display screens can include dashboards, topology maps, and network flow visualizations. In addition, the visualization platform can be configured to receive user input that causes filtering of displayed data.
[0056] For example and still referring to Figure 1 , the topology system logic 138 can generate a topology map visualization of the connections linking the fabrics detected by the controller 102, which is shown by the fabrics within the logical area 132 represented by cloud A 104 and cloud B 106. In addition, the topology system logic 138 can generate various graphical user interfaces (GUIs) that show network traffic flows between, to, and from fabrics managed by the controller 102, traffic flow heat maps, packet captures, network health, link latency, encryption, firewalls, etc., as shown in the second logical area 134.
[0057] Embodiments of the present disclosure provide many advantages over current systems that provide dashboards showing controller parameters because current systems do not provide the ability to visualize connections between structures deployed across multiple cloud networks, the status of resources and connections between resources of multiple clouds, and network data flows across structures of multiple clouds. As an example, an enterprise network may utilize resources deployed in multiple cloud networks, and an administrator of the enterprise network may wish to obtain a visualization of the status of all structures and connections associated with these resources. However, because the enterprise network spans multiple cloud networks, conventional systems are unable to provide such a solution. By only obtaining a textual representation of the status of each structure in a single cloud (e.g., through a command line interface), an administrator cannot obtain a complete view of the structures of the entire enterprise network, the connections between them, and their respective status. In addition, detection of abnormal or malicious network business patterns may not be detectable in the manner provided by current systems.
[0058] As used herein, visualization (or visual display) of constructs, connections between constructs, and the states of each is referred to as a topology map. Current systems fail to provide topology maps across multiple cloud networks, and fail to allow administrators to search across multiple cloud networks, or to visualize how changes in the state of a construct or connection in a first cloud network affects the state of resources or connections in a second cloud network. In some embodiments, a topology map can change automatically as the state of a construct or connection changes, or automatically change upon receiving construct metadata updates in response to certain events, such as at periodic time intervals (e.g., "dynamic topology maps").
[0059] In some embodiments, a network may be deployed across multiple cloud networks using multiple controllers to manage the operability of the network. In some such embodiments, each controller may collect information from the networks and structures it manages, and a single controller may obtain all such information, thereby enabling the visualization platform to provide visibility across a network (or multiple networks) that span multiple controllers.
[0060] refer to Figure 2A , an exemplary diagram of a logical representation of a controller 102 deployed within a cloud management system 100 is shown according to some embodiments. As described above, the controller 102 can be a software instance deployed within a cloud network to help manage the operability of constructs within multiple public cloud networks. According to this embodiment, the controller 102 can be configured with certain logic modules, including VPC gateway creation logic 200, communication interface logic 202, and data retrieval logic 204. The controller 102 can also include a routing table database 206.
[0061] In some embodiments, the gateway creation logic 200 performs operations to create a gateway within the VPC, including creating a virtual machine within the VPC; providing configuration data to the virtual machine; and prompting initialization of the gateway based on the configuration data. In one embodiment, the VPC gateway creation logic 200 starts a virtual machine in the VPC. EC2 instance. The virtual machine is started using a pre-configured virtual machine image published by the controller 102. In a specific embodiment, the virtual machine image is an Amazon Machine Image (AMI). When started, the virtual machine is able to receive and interpret instructions from the controller 102.
[0062] The communication interface logic 202 may be configured to communicate with the topology system logic 138 via an API. The controller 102 may receive queries from the topology system logic 138 through one or more API calls and respond with the requested data through the API.
[0063] The data retrieval logic 204 may be configured to access each construct managed by the controller 102 and obtain construct metadata therefrom. Alternatively or additionally, the data retrieval logic 204 may receive such construct metadata transmitted (or "pushed") from the construct without the controller 102 initiating one or more queries (e.g., API calls).
[0064] The routing table database 206 can store VPC routing table data. For example, the controller 102 can configure the VPC routing table associated with each VPC to establish a communication link (e.g., a logical connection) between a transit gateway and a cloud instance associated with a particular instance subnet. The VPC routing table is programmed to support communication links between different sources and destinations, such as internal computing devices, cloud instances within a particular instance subnet, etc. Therefore, the controller 102 obtains and stores information that reveals certain attributes of resources within the controller 102's authority (e.g., constructs such as gateways, subnets, VPCs, instances within VPCs, etc.) and state information related to the connections (communication links) between these resources.
[0065] refer to Figure 2B, an exemplary diagram of a logical representation of a topology system logic 138 deployed within a cloud computing platform is shown according to some embodiments. The topology system logic 138 may be a software instance deployed using cloud computing resources 136 and configured to communicate with the controller 102 and each gateway managed by the controller 102. The topology system logic 138 is configured with certain logic modules, including a tag logic 208, a tag database 210, an interface generation logic 212, a communication interface logic 214, and a topology snapshot logic 216. In addition, the topology system logic 138 may include a snapshot database 218, a construct metadata database 220, and a network data database 222.
[0066] In some embodiments, the interface generation logic 212, when executed by one or more processors, performs the following operations and results in the generation of Figures 3A-4C , an exemplary interactive user interface shown in 5B.
[0067] In some embodiments, the communication interface logic 214, when executed by one or more processors, performs operations as discussed herein with respect to querying the controller for construct metadata, receiving the requested construct metadata, and receiving network data from one or more gateways managed by the controller. In some embodiments, the received construct metadata and network data may be stored in a construct metadata database 220 and a network data database 222 (which may be separate or combined databases).
[0068] IV. Exemplary User Interface - Topological System Visualization Platform
[0069] Figures 3A-4C , 5B can be configured by the topology system logic 138 to be presented and displayed on various display screens and via various applications. For example, Figures 3A-4C Each user interface shown in 5B can be configured to be displayed on a computer display screen, a laptop, a mobile device, or any other network device including a web browser through a web browser. In addition, Figures 3A-4C , 5B can be configured to be displayed by a dedicated software application that is installed and configured to execute on any of the network devices described above. For example, the topology system logic 138 can be configured to provide the data and user interfaces described herein to a software application (referred to in the art as an "app") that can be installed and configured to be executed by one or more processors of a network device. Thus, when executed, the app causes the user interface described herein to be presented on a display screen (or associated display screen) of the network device.
[0070] 1. Dashboard
[0071] Reference now Figures 3A-3C , according to some embodiments, a graphical user interface (GUI) screen (or "interface screen") is shown that displays portions of a dashboard of a topology system visualization platform ("visualization platform"), wherein each portion is configured to show information obtained or determined by a topology system. Figures 3A-3C The interface screens of may collectively include a “dashboard” 300 that displays various attributes related to a network deployed across one or more cloud providers, and particularly across multiple cloud providers.
[0072] For example, Figure 3A The illustrated dashboard 300 includes several display portions 302, 306, and 308. A navigation panel 304 is also illustrated as part of the visualization platform generated by the topology system logic 138. The display portion 302 displays information related to information generated by a controller (e.g., Figure 1 The information displayed may include, but is not limited to, the number of deployed gateways, the number of current virtual private network (VPN) users, the number of user accounts, the number of temporary gateways (TGWs), the number of network connections (optionally filtered by cloud computing service), the number of border gateway protocol (BGP) connections, etc.
[0073] Figure 3A The display portion 306 of the dashboard 300 includes a list of virtual data centers that contain network resources, which again optionally span multiple cloud networks. Specifically, the display portion 306 includes a user input field (e.g., a check box) that is configured to receive user input that indicates how the content displayed by the dashboard 300 is to be used by one or more specific cloud networks (e.g., CLOUD (GCP), ORACLECLOUD (OCI)) filtering. In some embodiments, a virtual data center is a pool of cloud computing resources that can be hosted on a public cloud.
[0074] In addition, display portion 308 shows a world map, for example, which includes a graphical representation (such as icon 309) of each virtual data center listed in display portion 306, and a location on the world map representing its geographical location. Display portion 308 can be filtered according to the "filter by cloud" selection provided in display portion 306, and can be configured to receive user input to adjust the magnification of the map (e.g., "zoom in" or "zoom out").
[0075] Navigation panel 304 includes links to each general visualization provided by the visualization platform including dashboard 300 , which may encompass or provide access to any interface screen disclosed herein, such as interface screens 400 , 418 , 420 , and 504 .
[0076] Reference now Figure 3B , an illustration of a portion of a dashboard 300 displaying a plurality of graphs and charts is shown via a plurality of display portions 310 and 312. Each of the display portions 310 and 312 displays the distribution of resources throughout a multi-cloud deployment.
[0077] For example, as an illustrative embodiment, display portion 310 features a plurality of bar graphs showing metrics for resources managed by the controller; however, it should be understood by reading the drawings of the present disclosure that a bar graph is merely one type of illustration that can be used to present data, and the present disclosure is not intended to be so limited to the specific graphical representation type shown. Display portion 310 shows that the data displayed on the dashboard corresponds to constructs and network services across multiple cloud networks by specifically displaying "Accounts by Cloud", "Gateways by Cloud", and "Transit Gateways by Cloud". Similarly, display portion 312 provides a graphical representation of gateway metrics, including "Gateways by Type", "Gateways by Region", and "Gateways by Scale". In some embodiments, gateway metrics include one or more of the total number of gateways deployed, the number of virtual private network (VPN) users, the number of user accounts associated with one or more gateways, the number of transit gateways, the number of gateways deployed by a particular cloud computing resource provider, the number of Border Gateway Protocol (BGP) connections, or the number of temporary gateway attachments.
[0078] Figure 3A-3B Various metrics and characteristics of the gateways are shown, where the metrics may include one or more of the following: the total number of gateways deployed, the number of virtual private network (VPN) users, the number of user accounts, the number of transit gateways, the number of gateways deployed by a specific cloud computing resource provider, the number of Border Gateway Protocol (BGP) connections, or the number of temporary gateway attachments.
[0079] Additionally, one or more metrics may be derived from or based on gateway characteristics, which may include one or more of the cloud computing network in which each gateway is deployed, the type of each gateway, the size of each gateway, or the geographic region in which each gateway is deployed.
[0080] Reference now Figure 3C, showing an illustration of another graphical representation of network functionality or operation or operability based on data collected and processed by the topology system logic 138 and displayed as part of the dashboard 300. More specifically, according to this illustrative embodiment, the display portion 314 provides a graphical representation of network traffic between resources across multiple cloud networks over an adjustable time period (e.g., 24 hours). The time period can be adjusted by the topology system logic 138 based on receipt of user input. For example, user input corresponding to a user selection of a portion of the graphic shown can be received. In response to such received user input, the topology system logic 138 can change the graphical representation to target the selected portion that can now be represented by a smaller time interval (e.g., 15 minutes, 30 minutes, 1 hour, etc.).
[0081] In some embodiments, the dashboard 300 (and Figure 4A -5G) is the result of a user input requesting such a visualization. In some embodiments, in response to receiving the request, the topology system logic 138 will request the construct metadata as discussed above and store the construct metadata and the latest network data received from the gateway in a data store (e.g., construct metadata database 220 and / or network data database 222, which, as described above, can be a single database). In addition, the topology system logic 138 then generates the requested visualization based on the stored data.
[0082] In some embodiments, the topology system logic 138 will automatically update the visualization (e.g., generate an updated visualization and cause a re-rendering of the display screen) at periodic time intervals (e.g., every 30 seconds, every 1 minute, etc.). In some embodiments, upon the occurrence of a triggering event, such as upon receipt of a user input requesting a refresh of the display screen, an updated visualization will be generated and displayed. The updated visualization will be updated based on newly received or acquired construct metadata and / or network data since the previous rendering.
[0083] V. Monitoring Capabilities – ThreatIQ
[0084] As discussed above, the distributed cloud management system 100 can provide a display including visual elements to showcase structures residing in multiple cloud networks. In addition, the distributed cloud management system 100 can enable administrators to assign structures to segments within a networked cloud environment, where structures outside the same segment can be prevented from communicating with each other. These segments are enabled by security domains, and the ability of structures within a segment to communicate with each other is dictated by security domain policies. Security domains can be generated and enabled via the controller 102 (or optionally via the topology system logic 130). Any security domain policies for each segment can also be generated and enabled in a similar manner. Therefore, the interface generated by the topology system logic 138 can show logical and physical views of domain segments and their connection relationships.
[0085] An additional feature provided by the distributed cloud management system 100 is a security monitoring feature ("ThreatIQ") that enables monitoring of security threats in the networked cloud environment, generates and transmits alerts when threats are detected in the networked cloud environment (e.g., in network traffic flows), and can be configured to block traffic associated with the threats. All of these capabilities apply to the entire networked cloud environment (multi-cloud or single cloud) managed by the controller 102.
[0086] In some embodiments, when the current behavior exceeds a certain threshold (eg, approaches the outer limits of a baseline range or exceeds the baseline range by an amount), an alert is generated.
[0087] The ThreatIQ feature provides visibility into known malicious threats that attempt to communicate with constructs throughout the networked cloud environment. In some embodiments, the controller 102 or topology system logic 138 may store (or otherwise access) a list of well-known malicious sites or IP addresses that are known to be bad actors ("threat IPs"). The topology system logic 138 obtains network traffic and construct data (in real time) from gateways deployed by the controller 102 and / or the controller 102 itself, and the topology system logic 138 analyzes the network traffic and construct data to detect traffic from threat IPs. In some embodiments, this analysis may include comparison with a database of known malicious hosts.
[0088] Reference now Figures 4A-4C , according to some embodiments, illustrates an example of Figure 1 An exemplary graphical user interface of monitoring features enabled by a distributed cloud computing system. Figure 4A , according to some embodiments, an interface screen showing portions of a dashboard of a visualization platform, wherein each portion is configured to show information obtained or determined by a topology system. Interface screen 400 may be an additional interface screen including dashboard 300 discussed above. Figure 1The data collected by the distributed cloud computing system used in the analysis used to generate the interface screen discussed above may also be the data used in the analysis that leads to the generation of interface screen 400. More specifically, the data may include various attributes related to a network deployed across one or more cloud providers, and especially across multiple cloud providers.
[0089] Figure 4A The interface screen 400 may be referred to as a "threat view" and includes a geographic map portion 402 showing the approximate geographic locations 403 of known malicious IPs that have communicated with the networked cloud environment. An enumeration of the number of unique threats (enumeration 404), a threat count (enumeration 406), and an all threat count (enumeration 408) may be provided. In addition, the severity level of the detected threat IPs may be provided by a graphical representation 410, and the attack classification of each threat may be provided by a graphical representation 412. In addition, a graphical representation 414 showing threats within a specific time period may be provided, and a graphical representation 416 showing total threats over a period of time may be provided.
[0090] refer to Figure 4B , according to some embodiments, a display including a reference Figure 4A Detailed information about each threat record is shown in the interface screen 418, including the source IP of the threat, the destination IP, the gateways through which the threat-IP traffic passed, associated traffic flow data (date and time, source and destination ports, etc.), and threat information, such as why the particular threat was considered a threat. For each threat record, a network topology map 420 can be accessed, in which the associated compromised gateway is highlighted (see Figure 4C ). Once the topology map 420 is viewed, a drill-down to the instance level can be performed on the map, where the compromised instance (which is communicating with and sending out to the threat IP) is highlighted. This topology view can easily identify the subnet that the compromised server is deployed on, as well as the transit gateway it uses to communicate with the threat IP.
[0091] While ThreatIQ's threat view provides visibility into threats detected in your network, additional functionality can include taking action on these threats, such as enabling notification alerts (e.g., via a preferred communication channel (email)) when threat-IP traffic is first detected, or viewing historical information about when alerts were triggered via an interface screen generated by the topology system logic 138, including the names of gateways within the threat-IP traffic flow. In addition, threat IP traffic can be blocked. When a threat IP is first detected in a traffic flow, the controller 102 or topology system logic 138 instantiates security rules (stateful firewall rules) on all gateways within the flow (all gateways within a VPC, VNET, virtual cloud network (VCN)) to immediately block traffic associated with the threat IP. If the threat IP is removed from the database of threat IP sources, the controller 102 can automatically remove the security rules for that particular threat IP from the affected gateways, and the associated traffic is no longer blocked. Otherwise, the security rules for that particular threat IP remain implemented.
[0092] VI. Network behavior analysis
[0093] In addition to the features discussed above, as provided by Figure 1 As part of a broader security platform enabled by distributed cloud computing systems, topology system logic can include network behavior analysis features. Figure 5A , shows a graphical representation of functionality associated with a broader security platform 500 according to some embodiments. Specifically, Figure 5A Network behavior analysis 502 is shown as a component of such a platform 500, which also includes one or more components such as a next-generation firewall, distributed control (e.g., structure segmentation), known threat signature detection, distributed inspection (e.g., inspecting network traffic in and out of each segment), and / or malicious IP detection.
[0094] As security experts and network administrators understand, cloud networking introduces new threat vectors, which often require new defense approaches. For example, as the exposure of networked cloud environments expands, fundamental security challenges evolve due to the migration of constructs and network services in the cloud, such that there is no single perimeter, for example, the number of ways or points to access a particular construct may be infinite.
[0095] Thus, the migration to cloud computing has increased complexity for security professionals and network administrators. For example, traditional security solutions (e.g., solutions that were typically deployed prior to cloud computing migration) leave gaps in an enterprise's network security. For example, traditional security systems focus on a single checkpoint (e.g., ingress / egress) and are typically signature-based (e.g., primarily leveraging signatures of known bad actors). However, such systems are unable to protect against threats in the cloud because traditional cloud computing does not have a single access point (e.g., when all of an enterprise's network equipment is located on-site, there is no single entry point into the enterprise network). In addition, systems that rely on signatures of known bad actors are often outdated by the time the signatures are generated because bad actors may hide their IP addresses or otherwise distort their signatures in order to not be detected. Furthermore, such systems have difficulty responding to zero-day attacks because such threats have not been previously discovered and no signatures have been developed.
[0096] Furthermore and more specifically, Figure 1 A distributed cloud computing system provides visibility and the ability to perform analysis on a networked cloud environment spanning multiple clouds. Cloud service providers can provide some analysis within a single cloud, but cannot provide network behavior analysis across a construct spanning multiple clouds.
[0097] It should be understood that these deficiencies in traditional security systems pose high business risks to enterprises, including data loss, leakage, and resource hijacking, which may lead to loss of customer trust and have catastrophic financial impacts.
[0098] refer to Figure 5B , according to some embodiments, an interface display of analysis results related to network behavior analysis is shown. Interface 504 can be an additional interface screen including the dashboard 300 discussed above, which includes a graphical representation of a baseline 506 established for a specific set of network services, which is shown as a band (or range) of measured values (e.g., bytes) over a period of time. Baseline 506 is a fingerprint of an aspect of a networked cloud environment generated over time, and a baseline refers to a specific metric. Example metrics for which a baseline can be generated include, but are not limited to, data leakage (data outbound on a single or aggregated VPC / VNET), lateral movement (network services move from one VPC / VNET to another within a cloud networking environment), port and / or protocol usage (e.g., the number of new ports used over time, the amount of data transmitted using a specific protocol over time), distributed denial of service (DDoS) attacks (aggregate inbound bandwidth of VPC / VNET), port scan detection (ports and protocols used), and / or unencrypted business flows (identifying encrypted ports and protocols used).
[0099] Taking the data leak of a single VPC as an example, historical (or current) network traffic data can be used to generate an initial baseline that indicates the amount of outbound data transmitted from a particular VPC. In one embodiment, the average of the outbound data for a particular time period can be determined to determine the expected outbound data for those particular time periods, which represents the baseline. For example, the outbound data from a particular VPC in a day (e.g., from 12:00am to 11:59pm) can be averaged (using outbound data from that particular VPC for weeks, months, years, etc., if any) so that the baseline represents the expected outbound data on any particular day. In some embodiments, the baseline for a particular day can be composed of 1440 ranges, each representing a one-minute interval, where the range is expressed in a particular metric, such as 900-1200 bytes.
[0100] However, the baseline may instead be more granular in its representation, such that expected outbound traffic is represented at minute (or several-minute) intervals throughout the day (e.g., network traffic is expected to fluctuate over the course of a day-to-day basis). Thus, such a baseline may include expected outbound data from a particular VPC at minute intervals on any given calendar day. In some embodiments, such a baseline may be further refined to account for expectations of outbound data transmissions on a particular day (e.g., a particular weekday, a particular weekend day (Saturday or Sunday), a particular holiday, etc.). Such refinement to minute (or several-minute) intervals throughout the day, or further refinement to individual calendar days, may be useful in improving the accuracy of a baseline of outbound data for a particular VPC during a particular time period on a particular day, e.g., such a baseline may more closely represent the actual expected outbound data for a particular VPC at any particular minute of a calendar day.
[0101] In some embodiments, a baseline may be generated using machine learning techniques, such as supervised or unsupervised learning techniques. As an example, a feed-forward neural network model may be trained using historical data and configured to learn a baseline. An illustrative training process includes providing a multi-layer neural network with timestamped historical outbound data (as training data) for a particular VPC that is known to be expected (e.g., non-malicious) outbound data. The neural network generates a model that learns the expected outbound data for a given time period (e.g., minute intervals), thereby generating a baseline 506. Generation of the baseline may be performed by the topology system logic 138.
[0102] Interface 504 also includes a current behavior indicator 508, which represents a curve of actual network traffic over time for a given metric. Figure 5B As shown, current behavior indicator 508 can be displayed as an overlay on baseline 506 so that a security expert or administrator can easily see how the current behavior compares to baseline 506. Figure 5BAs shown, spikes 510, 512 may represent abnormal (irregular, unexpected) behavior due to the fact that the current behavior is outside the baseline 506 (which represents expected behavior). Spikes are not the only type of abnormal behavior that may be involved; rather, abnormal behavior may be any behavior outside the baseline.
[0103] The network behavior analysis feature can detect differences between current behavior (actual behavior) and a baseline, generate an alert, and transmit the alert to a security expert or administrator (or store the alert for later access). Therefore, compared to traditional security systems, especially those that utilize signatures of known bad actors, the network behavior analysis feature is advantageous by determining the expected behavior of one or more specific constructs (e.g., individual VPCs or aggregations thereof) and detecting abnormal behavior of the one or more specific constructs, regardless of whether such behavior is associated with a known bad actor. When abnormal behavior is detected, remedial actions can be taken, such as blocking certain network services, diverting certain network services, etc.
[0104] In addition, baselines can be tuned over time to adapt to the changing behavior of constructs or aggregates thereof. For example, as a networked cloud environment grows over time to add constructs within a particular VPC (e.g., increase the number of virtual machines (VMs) operating within the VPC), the various baselines for that particular VPC may change over time (and similarly when VMs are removed). As an illustrative example, as additional VMs are deployed within a particular VPC, the baseline volume of outbound data will likely change over time. Therefore, baselines can be tuned at regular intervals (e.g., using rolling historical data for adjustment). For example, a baseline for outbound data for a particular VPC can be tuned on a monthly basis by determining a new baseline for outbound data for that particular VPC using a network traffic metric that indicates outbound data for only the past three months.
[0105] In some embodiments, machine learning or other statistical analysis can also be used to predict future behavior based on current behavior. Thus, regression analysis (e.g., using any of linear regression, decision trees, support vector regression, Lasso regression, and / or random forests) can be used to determine a predicted value for the current behavior (e.g., at a specific future time). The predicted value can then be compared to a baseline for the future time so that similar alerts can be generated and / or remedial actions can be taken, as discussed above.
[0106] Another advantage of the above network behavior analysis signature is that such signatures can be specific to a particular enterprise and can be customized accordingly. Therefore, unlike signatures of threat IPs that are generated and distributed to all enterprises, network behavior analysis signatures are specific to networked cloud environments and are developed based on specific historical data (e.g., historical data of a specific VPC / VNET or an aggregation thereof). Therefore, this is not a "one size fits all" approach.
[0107] 1. Further technical details
[0108] In view of the above description, especially corresponding to Figure 5A-5B The description of the present invention now provides further technical details on the implementation and operability of anomaly detection performed by the topology system logic deployed within the cloud computing platform. Figure 1 As shown, the topology system logic 138 is communicatively coupled to the controller 102 and the gateway of each deployed VPC (and as described above, the term "VPC" includes any of a VPC, VNET, VCN, or equivalent virtual private cloud, where such terminology may depend solely on the corresponding cloud service provider). Therefore, the topology system logic 138 obtains data about network traffic flows from either / both of the controller 102 and the various gateways. As a result, the topology system logic 138 can monitor network traffic metrics at the VPC level (e.g., by monitoring network traffic at one or more gateways of each VPC). Also as discussed above, the controller 102 is configured to deploy each VPC and the corresponding gateway, and therefore has access to this data related to network traffic flows.
[0109] The ability to monitor network traffic flows at the VPC level provides several advantages to users of the topology system logic 138, one of which includes anomaly detection at the VPC level. Importantly, anomaly detection may differ from traditional anomaly detection in network traffic, which is performed in the prior art as a centralized location for the entire network. For example, traditional anomaly detection may include routing all network traffic entering or exiting an enterprise network to a central analysis location before entering or exiting the enterprise network, thereby causing congestion points that typically slow down the flow of network traffic.
[0110] Furthermore, analysis at a centralized location cannot provide detailed information about a specific VPC and cannot detect all anomalies. For example, network traffic analyzed at a centralized location may appear normal at the enterprise network level, but at the VPC level, network traffic may surge in an abnormal manner and / or certain IP addresses or ports may be new or abnormal for a specific VPC. However, the volume of network traffic and / or IP addresses / ports may not be abnormal for the entire enterprise and therefore will not be flagged as abnormal.
[0111] Therefore, VPC-level anomaly detection provides a much more granular level of analysis and monitoring than existing technologies can provide. Figures 6A-6C Further discussed, specific VPCs in the distributed cloud computing network can be selected for monitoring, so that a baseline (or fingerprint) for each selected VPC can be generated based on monitoring the selected VPC during a learning period (e.g., a predefined time period). The fingerprint includes an analysis of various metrics of the selected VPC, such as the amount of network traffic entering / exiting the VPC (optionally, at different times of the day, different days of the week, etc.), source and destination IP addresses, ports used, etc. As an example, the fingerprint can include an average of network traffic transmitted to and from the selected VPC for each hour of each day of the week, and a list of known IP addresses and ports used. Therefore, when comparing real-time traffic to the fingerprint, when new or unknown IP addresses or ports are used, the network administrator or security expert can mark anomalies for inspection. In addition, when the network traffic to / from the selected VPC is outside a predetermined number of standard deviations from the average (e.g., for a given hour as just an illustrative time frame), anomalies can be marked. This level of granularity is not achievable in the prior art. Furthermore, the prior art is unable to perform such a level of granularity—particularly across distributed cloud computing networks that span cloud networks provided by multiple cloud service providers—because prior art systems do not deploy and control VPCs and corresponding gateways. As a result, anomaly detection provided by topology detection system 138 as discussed herein provides technical benefits and improvements in the field of network security.
[0112] Reference now Fig. 6A , a graphical representation of an interface screen showing a portion of a dashboard of a visualization platform, the interface screen being intended to show information related to anomaly detection within network traffic of a cloud computing environment, is shown according to some embodiments. Fig. 6A The dashboard display screen (dashboard) 600 may be Figure 3A 600 includes content obtained from the controller 102 and any gateways deployed therefrom and any statistics or data derived or generated therefrom (where the topology system logic 138 obtains and derives or generates such data or statistics). The dashboard 600 includes a display portion 602 that includes several links (tabs), the selection of which results in the display of certain data or statistics related to network traffic generated by the topology system logic 138. Specifically, Fig. 6ASelection of an "anomaly" feature 604 is shown, which results in displaying content corresponding to the anomaly detection feature described above. When anomaly feature 604 is selected, dashboard 600 includes a configuration display portion (or display pane) 606 that includes an indication of the number of VPCs being monitored and the number of VPCs that are in a learning phase (also referred to herein as a learning period) (to the left of configuration display pane 606). In addition, configuration display pane 606 includes a detection sensitivity user input (UI) component that is configured to receive user input corresponding to a selection of a detection sensitivity level (e.g., low, recommended, or high). Adjustment of the detection sensitivity level can refer to an adjustment of the number of standard deviations from one or more metrics of the VPC fingerprint required to flag an anomaly. For example, a low sensitivity detection level may correspond to an anomaly being flagged when the real-time data (one or more metrics) is at least 5 standard deviations (+ / -) from the fingerprint of the VPC, a recommended sensitivity detection level may correspond to an anomaly being flagged when the real-time data (one or more metrics) is at least 3 standard deviations from the fingerprint of the VPC, and a high sensitivity detection level may correspond to an anomaly being flagged when the real-time data (one or more metrics) is at least 1 standard deviation from the fingerprint of the VPC. Note, however, that these are exemplary levels and may be adjusted in real time. Additionally, the configuration display pane 606 includes a UI component (e.g., a toggle switch) that turns the alert on and off.
[0113] Dashboard 600 also includes a set of time filters 608, which can be adjusted by user input, wherein the implementation of the time filter can affect the displayed metrics or statistics displayed in one or more metric display panes 610-624, each of which provides a graphical or textual representation of anomaly detection. Examples of such metrics or statistics include, but are not limited to: total number of anomalies detected (display pane 610); VPCs with anomalies (display pane 612); metrics with deviations (display pane 614); anomalies by deviation label (low, medium, high, etc.) (display pane 616); anomalies by VPC (display pane 618); metrics with maximum deviation (display pane 620); anomalies over time (display pane 622); and total anomalies over time (display pane 624).
[0114] In addition, the dashboard 600 includes an anomaly detection pane 626, which provides a list or other graphical / textual representation of detected anomalies. For example, the anomaly detection pane 626 shows a tabular anomaly list with a set of rows 628, each row being related to an anomaly. Each row may include certain information corresponding to the anomaly, such as a timestamp of the detection time, the name or identifier of the VPC where the anomaly was detected, the cloud service provider of the cloud computing network where the VPC is deployed, multiple metrics monitored for the VPC (or, multiple metrics that meet the selected sensitivity detection level), the deviation level of one or more metrics, and optional feedback. The feedback icon 630 is shown as being selected (e.g., "thumbs down"), which indicates negative feedback. The topology system logic 138 may interpret negative feedback as an indication that the detected anomaly should not be considered as an anomaly of at least a specific VPC. Feedback may be received by the feedback icon 630 through user input. The feedback icon 630 may alternatively or additionally also include a positive feedback option (which will be interpreted as confirmation of the anomaly). The received feedback may be used to tune the fingerprint in the same manner as discussed above.
[0115] Reference now Figure 6B , according to some embodiments, a graphical representation of an interface screen displaying data about monitored VPCs is shown. Display screen 632 is shown as a "pop-up window", which can be an auxiliary display screen presented as an overlay of dashboard 600. However, in other embodiments, display screen 632 can be directly integrated into dashboard 600. Display screen 632 can include a list of VPCs being monitored and VPCs in the learning phase (e.g., table 634). For example, each row of table 6345 can be related to a VPC and include information such as the VPC name / identifier, the cloud service provider of the cloud computing network in which the VPC is deployed, the region of the cloud computing network, the status of the learning phase of the VPC, and the detection status. Display screen 632 can be accessed by selecting a user input of an "edit" or "menu" UI component in the configuration display pane 606 (e.g., on the left side adjacent to the indication of the number of VPCs being monitored and the number of VPCs in the learning phase).
[0116] refer to Figure 6C, a graphical representation of an interface screen configured to receive user input indicating instructions for managing monitoring of a specified VPC is shown according to some embodiments. Display screen 636 is shown as a "pop-up window" in a similar manner to display screen 632, but in some embodiments it can also be directly integrated into dashboard 600. Display screen 636 includes content about VPC monitoring management. For example, display screen 636 includes a first list (table) 638 of all VPCs that can be used for monitoring. For each VPC, the first list 638 may include the VPC name / identifier, the cloud service provider that provides the cloud computing network in which the VPC is deployed, and the corresponding region. A second list (table) 640 may also be provided, which includes a list of all VPCs currently monitored by topology system logic 138, and information similar to the first list 638 may be provided for each VPC. Display screen 636 may be configured to receive user input indicating that one or more VPCs are to be monitored. In one embodiment, display screen 636 may utilize a drag-and-drop UI method so that a VPC from the first list 638 can be dragged and dropped into the second list 640 to indicate that the VPC is to be monitored. However, other user input methods known to those of ordinary skill in the art may be utilized.
[0117] In addition, the display screen 636 includes a UI component 642 that indicates a number of learning periods (e.g., in weeks) for the newly added VPC. In some examples, the UI component 642 can be a text box that is configured to display learning (e.g., 4 or "four") and is further configured to receive user input to adjust the learning period.
[0118] In the foregoing description, the invention has been described with reference to specific exemplary embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention as set forth in the appended claims.
Claims
1. A distributed cloud computing system, comprising: a controller configured to deploy a first virtual private cloud (VPC) in a first cloud computing network, deploy a first gateway in the first VPC, deploy a second VPC in a second cloud computing network, and deploy a second gateway in the second VPC, and wherein a first subset of the plurality of constructs is associated with the first gateway and deployed in the first cloud computing network, and a second subset of the plurality of constructs is associated with the second gateway and deployed in the second cloud computing network; as well as Logic stored on a non-transitory computer medium, which, when executed by one or more processors, results in the performance of various operations, including: generating a first fingerprint of the first VPC, the first fingerprint being a statistical measurement of a plurality of network metrics during a learning phase; generating a second fingerprint of the second VPC, the second fingerprint being a statistical measurement of a plurality of network metrics during a learning phase; receiving metadata about each of the first gateway and the second gateway from the controller; receiving network data from each of the first gateway and the second gateway, wherein the metadata and the network data identify each of the plurality of constructs, a communication path between each of the constructs, and in which cloud computing network each of the constructs is deployed; Detecting anomalies in one or more network traffic metrics of the first VPC or the second VPC based on a comparison of the received network traffic and the corresponding fingerprint; and Generate alerts for detected anomalies.
2. The distributed cloud computing system of claim 1 , wherein the logic, when executed by one or more processors, causes the performance of further operations, the further operations comprising: A visualization is presented on a display screen of a network device, showing multiple metric display portions, each metric display portion providing a textual or graphical representation of a metric related to anomaly detection of at least one of a first VPC, a second VPC, a first cloud computing network, or a second cloud computing network.
