CROSS REFERENCE TO RELATED APPLICATION

This application claims the benefit of priority to US provisional patent application No. 62/341,719, the entire contents of which are incorporated herein by reference.

FIELD

[0001] The present invention relates to networking, and in particular, to network routing and network communications.

BACKGROUND

[0002] In today's Internet, an application (or service) sends packets to the wire, and lower layer switches and routers, which are responsible for determining the packet transport path to the destination. The application does not control or determine the path. Likewise, the network doesn't control the application, and therefore can't directly "instruct" the application's behavior. The application and the network do not directly "speak" to one another to optimize the overall behavior, and do not use common identities or policies. Rather, the job of the network is to connect network nodes to other network nodes; not to connect applications or services.

SUMMARY

[0003] An aspect of the present invention relates to a platform, network and network control plane, which have multi-tier, multi-tenant architectures that enable applications to communicate with network elements and to communicate with intermediate platforms in order to determine their own network paths over any combination of networks, and to cooperate with the network to determin the networking which meets the application's, application user's and application administrator needs. The applications can be any application, including but not limited to enterprise applications, SaaS, PaaS, IoT, cognitive, AI, virtual reality, augmented reality, gaming and entertainment. The applications are enabled to "build their own networks", feeding application goals, requirements and policies to a platform which then can create new networks, comprised of software-defined network nodes on commodity servers and compute, including end user devices such as mobile phones and IoT devices. These networks can be ephemeral, specific to a single application flow and torn down after the flow.

[0004] In another aspect, a system of software nodes forms an overlay network on a physical, lower level network of physical nodes, and permits an application to control or at least influence network routing over the physical network.

[0005] Another aspect of the present invention relates to a platform with a multi- tier, multi-tenant architecture that enables application owners, where owner is defined as ASPs, users and/or organizations, to dictate end-to-end policies and goals, independent of underlying access network owners and technology. Policies can change parameters to achieve policy goals, including transport path traversed; networks traversed; encryption and security parameters; logical or physical geographies traversed; network protocols used; number of connections per session; number of different routes per session; number of times to transmit each packet; error handling and threat responses; network interfaces used; redundancy and backup options; TCP and UDP acceleration and optimization options; proxy, gateway, firewall, security element and B2BUA insertion; thresholds and responses to packet loss, jitter and latency; and algorithms used to determine network paths including machine learning and predictive algorithms [0006] Another aspect of the present invention relates to a platform with a multi- tier, multi-tenant architecture that enables application owners to dictate policies for end-to-end desired business results, independent of underlying access networks. Business results can includes quality, measured qualitatively or quantitatively; reliability, connection success ratios, availability and uptime; compliance; security; and throughput and speed.

[0007] Another aspect of the present invention relates to a platform with a multi- tier, multi-tenant architecture that has APIs, SDKs and Portals, which enable application owners to set the policies listed above.

[0008] Another aspect of the present invention relates to a platform with a multi- tier, multi-tenant architecture that includes APIs, SDKs and Portals (web and mobile app), that also enable the policies to be set or changed according to real-time or anticipated (via machine learning or algorithms) conditions, triggers or events. The events can be machine-generated, for example from an IoT sensor or camera, or an AI application observing anomalies or patterns in an IoT feed, or based on human interaction with the application; or based on external events (weather, political, etc) or predicted external events.

[0009] Another aspect of the present invention relates to a platform with a multi- tier, multi-tenant architecture that includes APIs and Portals, and instrumentation to provide application owners with end-to-end data, analytics and visibility of their application, regardless of networks traversed, for example APIs, SDKs and real-time data feeds which provide network performance, security and compliance data to the application and/or application user, in real-time, proactively (e.g. predicted by machine learning) and historically [0010] Another aspect of the present invention relates to platform with a multi-tier, multi-tenant architecture APIs and Portals, which enable the network to communicate to the app in real-time, such that the application may change its current or future behavior in order to meet application owner policies or proactively account for conditions anticipated by machine learning or other algorithms and policies. For example, bi-directional APIs and SDKs which enable the network and app to continually communicate/collaborate to optimize security, quality or make changes due to events, at time according to identities (user identities, application identities, thing (IoT) identities, silicon/chip/hardware identities) which may be communicated from the application or from systems/platforms involved in the application's management.

[0011] Another aspect of the present invention relates to platform APIs and Portals, that enable application owners to share identities, authentication, authorization and policies with other systems, both accepting policies from other systems and distributing policies to other systems, such that network elements and application layers can leverage common identities, security and policies

[0012] Another aspect of the present invention relates to a method of routing packets through a physical network. The method comprises providing one or more software nodes, the software nodes being connected to each other over the physical network to form an overlay network; and forming a communications channel between at least one of the software nodes and an application, wherein the application instructs the at least one software node on how to route data over the physical network, or how to query another element to discover routing instructions. The application may use different instructions for different types of data to be transmitted over the physical network. [0013] In some embodiments of the method, the application and the at least one software node dynamically adjusts routing during communications

[0014] In some embodiments of the method, at least one software node is associated with a target node of the physical network, and upon receipt of data from the target node, the at least one software node instructs the target node to transmit to a new target node of the physical network. The node may be a pool of software defined nodes, abstracted via techniques such as load balancing, DNS, GeoDNS and multicasting.

[0015] In some embodiments of the method, communications are transmitted from the overlay network to the application in real time during execution of the application to cause the physical network to change its operation including creating new network nodes to be created in quasi-real-time and added to the network in order to meet certain application needs

[0016] In some embodiments of the method, the application is selected from the group comprising an enterprise application, a software as a service (SaaS), a platform as a service (PaaS), an internet of things (IoT), a cognitive computing application, an artificial intelligence (AI) application, a virtual reality application, an augmented reality application, a gaming application and an entertainment application.

[0017] Another aspect of the present invention relates to an overlay network for use with a data network that routes packets from one location on the data network to another location on the data network. The overlay network comprises one or more software nodes, wherein at least one of the software nodes is configured to receive commands from an application. The application includes software for generating the commands and sending the commands to the at least one of the software nodes. The commands used by the at least one of the software nodes to control routing of packets, generated by the application, through the data network.

[0018] In some embodiments of the overly network, at least one of the software nodes is configured for bidirectional communication with the application.

[0019] In some embodiments of the overly network, at least one of the software nodes is configured to dynamically negotiate with the application to implement changes to the routing of the packets.

[0020] In some embodiments of the overly network, at least one of the software nodes is an edge endpoint.

[0021] In some embodiments of the overly network, the software for generating commands is configured to change the commands based upon at least an Internet of Things (IoT) feed.

[0022] In some embodiments of the overly network, the software for generating commands is configured to change the commands based at least in part upon a sensor or a camera input.

[0023] In some embodiments of the overly network, the destination is a target node.

[0024] In some embodiments of the overly network, the at least one of the software nodes is configured to dynamically transmit to the application, based upon conditions at the target node, an address corresponding to a new target node.

[0025] In some embodiments of the overly network, the application is selected from the group comprising an enterprise application, a software as a service (SaaS), a platform as a service (PaaS), an internet of things (IoT), a cognitive computing application, an artificial intelligence (AI) application, a virtual reality application, an augmented reality application, a gaming application and an entertainment application. [0026] Another aspect of the present invention is a software platform for network communications. The software platform comprises computer-executable instructions configured to generate an overlay network, which may be world-wide, and which enables an application to determine a network path through an underlying access network for data generated by the application.

[0027] In some embodiments of the software platform, computer-executable instructions are provided, which are configured to enable an owner or user of the application to set end-to-end policies, independent of the underlying access network.

[0028] In some embodiments of the software platform, computer-executable instructions are provided, which are configured to enable an owner or user of the application to set policies for end-to-end desired business results, independent of the underlying access network.

[0029] In some embodiments of the software platform, computer-executable instructions are provided, which are configured to generate an API that enables the owner or user of the application to set at least one of the end-to-end policies and the policies for end-to-end desired business results.

[0030] In some embodiments of the software platform, the policies can be set or changed according to at least one of real-time conditions, triggers and events.

[0031] In some embodiments of the software platform, computer-executable instructions are provided, which are configured to generate an API and instrumentation which provides an application with end-to-end data, analytics and visibility of the application, regardless of networks traversed in the underlying access network.

[0032] In some embodiments of the software platform, computer-executable instructions are provided, which are configured to generate one or more APIs and one or more portals that enable the global overlay network to communicate with the application in real-time, such that the application may change its behavior in order to meet application owner policies.

[0033] In some embodiments of the software platform, the one or more APIs enables the owner of the application to share policies with one or more other systems.

[0034] In some embodiments of the software platform, the one or more APIs enables the owner of the application to accept policies from the one or more other systems and distribute policies to the one or more other systems.

[0035] In some embodiments of the software platform, the global overlay network is integrated with the platform and comprises a plurality of software nodes.

[0036] In some embodiments of the software platform, the nodes are connected as traditional networked nodes, as peer-to-peer mesh nodes, or any combination thereof.

[0037] In some embodiments of the software platform, the nodes are agnostic to at least one of network technology and network ownership of the underlying access network.

[0038] In some embodiments of the software platform, the nodes are located in a public cloud, a private cloud, an Internet backbone environment, or on end user devices.

[0039] In some embodiments of the software platform, the nodes are applications, VNFs, standalone devices running on commodity computers, purpose built appliances, or services embedded in network infrastructure.

[0040] In some embodiments of the software platform, the services embedded in the network infrastructure comprise gateways, firewalls, proxies, deep packet inspection and session border controllers. [0041] Another aspect of the present invention is a method for routing data generated by an application to a destination under the control of the application. The method comprises: a) instructing, with the application, a software node of a global overlay network to initially route the data to a policy-defined target node of an underlying access network; b) at the target node, analyzing needs of the application and a real time status of candidate next hop nodes and associated paths of the underlying network with another software node of the global overlay network or a routing engine; c) selecting, with the another software node or routing engine, one of the candidate next hop nodes of the underlying network, which best meets the needs of the application; d) causing the data to be routed to the selected one of the candidate next hop nodes of the underlying network; and repeating steps b)-d) until the data reaches the destination.

[0042] In some embodiments of the method for routing data, the needs of the application include end-to-end policies of the application.

[0043] In some embodiments of the method for routing data, the needs of the application include policies for end-to-end desired business results of the application.

[0044] In some embodiments of the method for routing data, the policies can be set or changed according to at least one of real-time conditions, triggers and events.

[0045] In some embodiments of the method for routing data, the software nodes are connected as traditional networked nodes, as peer-to-peer mesh nodes, or any combination thereof.

[0046] In some embodiments of the method for routing data, the software nodes are agnostic to at least one of network technology and network ownership of the underlying access network including Internet, mesh networks and peer-to-peer networks [0047] In some embodiments of the method for routing data, the software nodes are located in a public cloud, a private cloud, an Internet backbone environment, on IoT devices, or on end user devices.

[0048] In some embodiments of the method for routing data, the software nodes are applications, VNFs, standalone devices running on commodity computers, purpose built appliances, or services embedded in network infrastructure.

[0049] In some embodiments of the method for routing data, the services embedded in the network infrastructure comprise gateways and session border controllers.

BRIEF DESCRIPTION OF THE DRAWINGS

[0050] FIG. 1 is block diagram illustrating an exemplary embodiment of a mobile overlay virtual network (MOVE) and an exemplary embodiment of a virtual internet for enterprises (VINE).

[0051] FIG. 2 is block diagram illustrating another exemplary embodiment of the VINE and another exemplary embodiment of the MOVE.

[0052] FIG. 3 is a flowchart illustrating the steps of a platform routing and session control method, according to an embodiment of the present invention.

DETAILED DESCRIPTION

[0053] Disclosed herein is a platform, network and network control plane that enable applications to determine, or at least influence, their own network paths. The application tells the switches and routers how and where to route its packets. The "where" may comprise a global overlay network, which includes endpoints, IP Back to Back User Agents (BBUAs), and management and control instance, that is managed according to the present invention. The global overlay network may be built on an existing endpoint and BBUA software. In one non-limiting embodiment, the global overlay network may be built on the endpoint and IP BBUA software disclosed in U. S. Patent 9,071,607, the entire disclosure of which is incorporated herein by reference. Further, the network of the present invention can continually "negotiate" with the application to dynamically make changes. The present invention can allow application users (people and/or devices), the user's administrative domain (usually a business or organization) and provider (often an ASP), to enact policies and business logic, which control networks that they don't own or manage. The network and network control is essentially embedded inside the application, regardless of where the application is physically launched from. Therefore, applications and application managers can control any network from anywhere, according to their needs and goals, without owning or managing the network.

[0054] FIGS. 1 and 2 illustrate exemplary embodiments of global overlay networks according to the present invention. One of the global overlay networks may comprise a mobile overlay virtual enterprise (MOVE) network and another global overlay network may comprise a virtual internet for enterprise (VINE). The MOVE network and the VINE are platforms built on any set of networks for enterprise security, policy control, compliance, performance and reliability network capabilities into applications and devices ubiquitously across any and all broadband access, campus LAN and WAN, as illustrated in FIGS. 1 and 2. The MOVE network travels with the applications and devices. The MOVE network has MOVE endpoints Ml l- M18, and a MOVE backbone M19 in FIG. 1 and MOVE endpoints M21-M25 and a MOVE backbone M26 in FIG. 2, which are software-embedded. The VINE provides enhanced internet performance and security to any application or browser over any internet access, with no software modifications. The MOVE network and the VINE share the same high capability internet-overlay platform that provides:

1) Better Quality of Experience (QoE) via dynamically optimized utilization of multiple parallel Internet Routes and TCP optimization.

2) Extremely robust security and network resilience as a Software Defined Perimeter (SDP) per the Cloud Security Alliance (CSA), creating highly secure and trusted end-to-end networks between any IP addressable entities that are highly resilient to network attacks.

3) On-demand, dynamically provisioned, "airgapped" networks.

Enterprise-level management and control.

[0055] The MOVE network and the VINE can be used for B2E (business-to- employee), B2B (business-to-business), and B2C (business-to-consumer) cloud- distributed communications amongst users, machines, and servers, to provide higher network performance, extreme levels of data-in-motion security, and enterprise-level control.

[0056] In some embodiments, the MOVE network and the VINE can comprise an "embeddable enterprise network service" that enables a site-less enterprise-managed overlay VPN to be integrated within applications that require a "site-less" capability for enterprise network security and performance, unconstrained by the inflexibility of a site-level WAN or SD WAN that is bound to one or more physical networks. Such applications include without limitation:

• Enterprise-developed applications.

• 3 ^rd party ISVs. • SaaS (e.g., CRM, sales force automation, office suites, HR, Finance, ERP, document management, CAD, ITaaS, etc.).

• B2C services (e.g., eCommerce, Finance, Healthcare, and etc self service, service desks, Points of Sale, Kiosks).

• IaaS, PaaS and APaaS (e.g., AWS, Azure, Google, SalesForce, Softlayer, BlueMix, MBaaS platforms, etc.).

• IoT implementations (e.g., GE, Telit, PTC, Gemalto, numerous other platforms).

• Cloud storage (e.g., Box, Dropbox, Egnyte, etc.).

• cloud security (e.g., Zscaler, Bluecoat,Websense, Cisco Scansafe, etc.).

• UCaaS.

• API Service Provider platforms (e.g., Twilio, Kandy, Nexmo, Cisco Tropo, etc.).

• Entertainment sites.

• Collaboration platforms (e.g., Webex, Go to Meeting, SFB, etc.).

• Office application suites (e.g., Microsoft, IBM, Google).

[0057] The MOVE network and the VINE, thus, add major new GTM channels for enterprise VPNs. For one example from just above, an ASP such as Microsoft Office365 for B2E or GE Predix for IoT, can integrate strong QoE, hyper-secure MOVE network into their application platform, where:

• The ASP can manage their multi-tenanted MOVE/VINE network as an integral part of their platform service • The ASP can provide each enterprise tenant with autonomous control of their own ASP-bound VPN

o The tenant-level exposure may be fully integrated inside their overall platform management, it may be native white-labeled MOVE/VINE management, or it may be a hybrid.

[0058] In some embodiments, the MOVE network and the VINE may be operative as a "un-network," because they make the VPN part of an application or service instead of being a separate network that the application or service has to traverse.

[0059] The MOVE network and the VINE can operate transparently across a mix of LAN and WAN, private and public, physical and virtual, fixed and mobile networks and network functions that they traverse, as illustrated in FIGS. 1 and 2. The MOVE network and the VINE supplement and relieve the constraints of traditional fixed and Mobile Internet and enterprise WANs and SD WANs.

[0060] In some embodiments, the MOVE network can comprise an enterprise VPN, which provides the enterprise with autonomous performance and compliance monitoring and control for sessions amongst any of the MOVE endpoints Mi l -Ml 8 (FIG. 1) and M21-25 (FIG. 2), which are any mix of:

• Applications using the MOVE SDK.

• IoT Devices using the MOVE SDK or device driver.

• Stand-alone MOVE gateways (e.g., M14-M17 in FIG. and M21- M24 in FIG. 2).

• Public cloud instances of MOVE (MOVE gateway "cloud connect"). • Private cloud instances of the MOVE network (e.g., Ml 8 in FIG. 1 and M25 in FIG. 2).

• Multi-function gateways where a partner has integrated the MOVE network with their own network functions (e.g., Loadbalancers/ADCs, SIP gateways, MCUs, recorders, etc.).

• Internet "breakout" endpoints via the VINE, such as VI 1, VI 2, V21 and V22 as shown.

[0061] In some embodiments, the MOVE network can be implemented within a multitenant platform such as Office365 or GE Predix, but where the ASP has the ability to provide autonomous enterprise network management to each of their client enterprises, and where each enterprise can be enabled to operate to their own authentication, performance and compliance policy for Office365 or GE Predix communications.

In some embodiments, as illustrated in FIGS. 1 and 2, the VINE may comprise a premium overlay shown, for example, as VI 1-12+ V15-18, V21-22 and V25-28.

[0062] The CE-to-PE connection may be over local underlay public internet access, encrypted such as with TLS or IP Sec or unencrypted. The VINE may require no software changes to the client application M13 or device M12 (FIG. 1). Similar to a CDN, the VINE can be integrated, via DNS Cname addressing. V19 (FIG. 1) and V29 (FIG. 2) may be controllers in the network, which provide full duplex internet session acceleration, hyper-security, and instrumentation, providing:

• Premium public internet for performant and highly secure web transactions.

• Optional QoS prioritized physical network routing: o On the "backhaul" from the IP BBUAs to public and private cloud endpoints.

o On the "fronthaul" from the edge endpoints to the IP BBUAs.

• "Internet breakout" to/from MOVE enterprise VPNs.

• Private application servers "cloud connect" via cloud-based MOVE Gateways providing managed, secure performance and compliance into all major public clouds and optionally private clouds.

[0063] Application and Device VPNs using TSL or IPsec tunnels over best effort internet are well-established today. The MOVE and the VINE provide a stronger value proposition for commercially-sensitive sessions where Quality of Experience is proportional to revenue and compliance must be rigorously managed. The MOVE network and the VINE provide:

• Better Quality of Experience:

• Multi-factor hyper-security:

endpoints

ARP poisoning None High immunity High protection Immunity

• Rejects

corrupted

packets

Data None Flow fragments can PE to PE sovereignty be routed via and PE to management selected Cloud is sovereignty identical to domains MOVE

• Enterprise Monitoring and Control:

[0064] Various embodiments of the MOVE network, as illustrated in FIG. 1, can have one or more of the following uses, structures, and capabilities:

• Typical use cases: B2E, B2B, B2C, and IoT enterprise networking.

• Form Factors/Overlay Network Presentation: o MOVE SDKs for integration of Enterprise Network into any application (Android, IoS, others): ^■ Direct application SDK.

^■ Embeddable SDK for integration into network-enabled SDK functions (eg UCaaS SDKs).

o MOVE device drivers for integration of Enterprise Network into any LINUX and Windows machines,

o MOVE Gateway for aggregation of local traffic sources onto the Enterprise Network,

o One common Virtual Backbone for both MOVE and VINE. Enterprise Overlay Network Application Policy Management: o Prioritize and selectively route or block sessions to the overlay network:

^■ 5 Tuple resolution: source ΪΡ address/port number, destination IP address/port number and the protocol in use.

Enterprise Overlay Network Monitoring and Analytics.

Heirarchical multi-tenant management, monitoring and analytics to pass network management down the delivery chain:

o E.g., SaaS provider view => MSP partner view => Enterprise

Customer view => Enterprise Department view => End User view.

Transparent interoperability with all mobile broadband access, campus LAN, Data Center LAN, and WAN virtual and physical network functions:

o Enterprise Switches, routers, firewalls, DPIs, IP VPNs, SD WAN, etc.

o Carrier network MPLS, Ethernet, 3G, 4G and Internet. o Cloud networks.

• Highest levels of data-in-motion security:

o Cloud Security Alliance Software Defined Perimeter:

^■ Rigorous isolation from unsecured network and high resilience to attacks.

o "Dynamic Spread Transport" of each session as many

"fragment-flows" across numerous internet paths prevents man-in-the-middle monitoring,

o Separately secure encryption of each of two "airgapped" legs comprising each dynamically shifting fragment flow in a session.

o Detection and discard of corrupted packets,

o Providing High immunity to DDOS, Man in the Middle, and OWASP Top 10 attacks.

• Maximum available network bandwidth and performance:

o "Dynamic Spread Transport" load-balances and multiplexes multiple best available paths across the public internet backbone:

^■ Available bandwidth= sum of all the paths.

^■ Adaptive congestion and high latency avoidance.

o Dynamic Spread Transport load-balances and multiplexes across multiple WAN access networks as available to any WLAN/LAN campus or Public or Private Data Center:

^■ Available bandwidth= sum of all the access networks. o Transport Protocol Optimization: ^■ TCP/IP proxy to eliminate TCP/IP sensitivity to WAN latency.

• MOVE and VINE Interop:

o MOVE policy management, monitoring and analytics encompass multipoint sessions that include a mix of VINE PoP public internet endpoints and MOVE enterprise network endpoints.

• Optional QoS Backhaul and QoS Fronthaul with preferred routing via COS-prioritized TCL Izo WAN:

o Via Policy selection of preferred routing via QoS IP Back-2- Back Agents:

^■ Front Haul to Front Haul QoS for Peer to Peer with QoS.

^■ Front Haul to Back Haul QoS for Endpoint to MOVE Cloud.

^■ Backhaul QoS alone for QoS "cloud connect" paths:

• Best effort is only to the closest VINE PoP.

^■ Or any mix for multipoint connections.

o With transparent automated rollover to other available Internet Access if QoS IP Back2Back Agents are impaired.

[0065] Various embodiments of the VINE, as illustrated in FIG. 2, can have one or more of the following uses, structures, and capabilities:

^■ Typical Use Cases: B2C, B2B, B2E public intemet connections.

^■ Form Factors:

o Softwareless HTTP or HTTPS/ TLS session access via URL. o CDN-style DNS address resolution (CName) to a local VINE PoP.

o One common Virtual Backbone for both the MOVE and the VINE.

^■ Transparent internet transport with local PoP accesses across a High security, High Performance inter-PoP Virtual backbone sharing the capabilities of MOVE, above:

o A VINE PoP is a shared industrial scale MOVE gateway. o VINE PoP-to-VINE POP for non-MOVE endpoints (including non-MOVE Clouds),

o VINE PoP-to-MOVE Cloud using a Move Gateway Cloud

Connect.

^■ Self-service monitoring, analytics and management of the end to end internet flows.

^■ MO VE and VINE Interop :

o MOVE policy management, monitoring and analytics encompass multipoint sessions that include a mix VINE PoP public internet endpoints and MOVE enterprise network endpoints.

[0066] In various embodiments, the MOVE/VINE Cloud Connect Ml 8, M25/V14, V24 and Cloud Exchange can have the following use, structure, and capabilities:

^■ Use Case: Integrates MOVE/VINE network security, performance, and resilience directly into private and public clouds o Cloud Connect= singular cloud connection using a private or dedicated cloud gateway instance

o Cloud Exchange= on-demand connection to any of many pre- connected clouds via a multi-tenant cloud gateway instance

^■ Reduces dependency on less flexible, more costly physical MPLS VPN-based cloud connect

^■ Form Factor:

o VNF (virtual network function) MOVE Gateways for Public and Private Cloud connections to MOVE enterprise WANs and VINE public internet.

^■ Optional QoS on the backhaul leg from the IP Back to Back Agents to the cloud and also on the Fronthaul from edge endpoints to the Agents.

o Cloud Connect: Private gateways for private or dedicated instances in public and private cloud

o Cloud Exchange: Pre-configured autoscaling multitenant gateways in AWS, Azure/0365, Force.com/SalesForce and other public clouds

[0067] The MOVE network can be implemented as an endpoint software, which establishes spread-transport flows across a global public internet "spread-transport" backbone, and dynamically spreads each session across multiple public internet pathways similar to how spread-spectrum radios dynamically spread radio links across multiple radio frequencies. The MOVE endpoints M11-M18 (FIG. 1) and

M21-M25 (FIG. 2) collaborate to break up/reassemble each flow into multiple fragment-flows that are dynamically load-balanced across highly dispersed public internet pathways:

• Each of many spread-transport fragment-flow pathways is through a core PoP "IP back-to-back agent" (similar to a SIP Back to Back User Agent):

o The Cloud Security Alliance Software Defined Periphery (SDP) encompasses a similar "air gapped" transport, but has typically be implemented without spread-transport back2back agents.

o Each spread-transport fragment-flow pathway is comprised of two independently encrypted separately initiated back to back IP paths, masking the actual endpoint pair members from each other.

• Spread-transport is extremely secure:

o Each session is broken into multiple fragment-flows, dynamically routed across independent internet routes:

^■ Even data-in-motion metadata is un-monitorable.

o Each fragment-flow half-path is individually securely encrypted.

o Corrupted packets are identified and quarantined

^■ Providing ARP poisoning immunity.

o Interception of an entire multi-fragment flow is not possible at any point in the network other than an endpoint, including by

Tata Communications. This protects from even meta-data or flow-behavior monitoring: ^■ Even interception of one encrypted fragment is highly improbable because the fragment flows dynamically hop across different internet pathways.

o The IP Back-to-Back agent moves IP address surfaces into the cloud onto a non-impacting attack target.

• Performance is provided via dynamic load-balancing to utilize the available capacity across the cumulative pathways that the session fragment-flows are spread across:

o The virtual internet path is not limited by the capacity and other impairments of any one internet path.

• Performance may be further enhanced via QoS -prioritized physical network routing of fragments on either or both of o The backhaul path between cloud endpoints and the IP back to back agents

o The fronthaul path between edge endpoints and the IP back to back agents

• Reliability is provided by the same dynamic load balancing:

o Underperforming internet pathways will be automatically identified and taken out of route:

^■ Providing inherent immunity to volumetric DDOS attacks.

[0068] The following Table, summarizes the uses of the present invention:

B2E • MOVE SDK MOVE Sessions include enterprise - Gateway

developed apps fronting • Edge to Cloud

• MOVE wrappers for enterprise • Edge to Edge (eg VoIP) off-the-shelf apps, private and • Multiple Edges(eg eg SFB public collaboration)

• MOVE drivers for cloud apps • Multiple Edges and PCs and user Cloud appliances • Cloud

Connect

• VINE via Browser

dedicat

TL For Wrappers- ed

• MOVE Gateway for

instance • OEM from someone like Site aggregation s OpenPeak

• Cloud • Partner with

Exchan MDM/MAM players ge

multi- tenant

instance

s

B2C • VINE via Browser T MOVE Sessions include

• MOVE SDK Gateway

enterprise - fronting • Edge to Cloud developed apps enterprise • Edge to Edge (eg VoIP) private and • Multiple Edges(eg public collaboration) cloud apps • Multiple Edges and

Cloud

• Cloud

Connec

t

• Cloud

Exchan

ge

B2B • MOVE to SD WAN, NA Federated entities

MOVE to WAN or independently manage the Federation MOVE to MOVE federated SD WANs, with back to back agreed policy control of Gateway ingress/egress via the GW

IoT • MOVE devices MOVE

• MOVE IOT Gateway

aggregators with local fronting

processing enterprise

• IoT LAN to MOVE private and

Gateway public cloud apps

Enterprise as a • MOVE-integrated MOVE Enterprise Tenant Instance will SaaS network SaaS client Gateway ideally be managed fully "tenant" • Browser TLS to VINE fronting integrated within the SaaS

SaaS management

provider

• Provisioning,

authentication, policy , etc

• Additional reporting may be required

SaaS/PaaS

provider

User Apps • MOVE-integrated MOVE • SaaS provider multi-tenant

SaaS client Gateway MOVE management

• Browser TLS to VINE fronting • Enterprise Tenant Instance

SaaS will ideally be managed provider fully integrated within the

SaaS management

IoT • MOVE-integrated MOVE • SaaS/PaaS provider multi- Device Gateway tenant MOVE management

• MOVE-integrated fronting • Enterprise Tenant Instance IOT aggregator SaaS/PaaS will ideally be managed

• MOVE Gateway provider fully integrated within the

SaaS/PaaS management

TCL UC • Provider Edge MOVE NA

Service Gateways fronting

Infrastructure Network functions

such as SBC and Load

Balancers

• MOVE-integrated

Network Functions

TCL • VINE as an NA

Network extension of IZO

Service WAN

Infrastructure • MOVE/VINE

integration with

TCL CDN

• MOVE as an out-of- band management

network

Network • MOVE integration NA • MOVE management Function within Physical and will ideally be integrated Providers Virtual Network within the network

Functions such as function management SBC , ADC/Load • Candidates: Sonus, balancers, and SD Citrix Netscaler WAN edge

devices(?)

API Platform • MOVE SDK • MOVE • "Native" direct

Services embedded with Gatewa enterprise MOVE

Providers other SDK y management

functions, eg WEB fronting or

RTC enterpri

se • MOVE management private integrate within the and "parent" SDK function public management cloud • Candidates: Twilio, apps Kandy, Cisco Tropo

• MOVE

Gatewa

y

fronting

SaaS or

PaaS

platfor

m

SD WAN • MOVE Gateway • MOVE • SD WAN Cloud Cloud back to back with Gatewa Connect and Cloud

Connect and SD WAN edge y Exchange

Cloud function fronting o Requires coherent

Exchange enterpri orchestration and se management private across both the and SD WAN and public MOVE cloud o One SD WAN to apps MOVE (CE or

• MOVE PE) interconnect Gatewa provides all y MOVE Cloud fronting Connects SaaS or (including QoS PaaS fronthaul + platfor backhaul option m from PE to

Cloud)

SD WAN • SD NA • Each SD WAN

Exchange WANI MOVE maintains autonomous

SD WAN2 management.including policy at the SD WAN exchange edge.

• MOVE Exchange edge policy is ^" standardized" and published to all.

[0069] In still further embodiments of the present invention, extended managed services with monitoring and management can be coherently integrated in the same multi-tier mult-tenant platforms that monitor and manage the MOVE network and the VINE. These may be peer services to the MOVE network and the VINE, or have the MOVE network and the VINE integrated within them.

UC PaaS Unified Communications Platform-as-a- Kandy, Twilio, others Service over MOVE/VINE

Distributed Secure data store behind MOVE/VINE Dispersive

Cloud Storage Technologies Storage

CASB Cloud Access Security Management Palerra, Skyhigh,

with and behind MOVE/VINE Netskope, vArmour, etc

[0070] The three part combination of Mobile Device Protection, MOVE/VINE data in motion protection, and Cloud Server protection provides a highly secure end-

2-end platform for B2E, B2C IoT, and other business applications.

[0071] FIG. 3 is a flowchart illustrating the steps of a platform routing and session control method, according to an embodiment of the present invention. In block 100, the application, device or third party platform (hereinafter "application") instructs one of the software nodes of the global overlay network of the present invention (e.g., a

MOVE or a VINE endpoint) to initially route packets to a policy-defined target node of an underlying access network (by DNS name, IP address, or policy match). The underlying network comprises wire, and lower layer switches and routers, which transport the packets to the destination. The target node of the underlying network is selected according to the needs of the application and one or more various identities including, but not limited to user identities, application identities, thing (IoT) identities, and silicon/chip/hardware identities. The packets can be an entire application stream, parts of the stream, or packet by packet routing. At the target node, in block 110, an analysis is performed, which may consider one or more of the needs of the application (the policies, business needs and the like describe above) and the real time status of: 1) each candidate next hop node of the underlying network; 2) the path to each candidate next hop node of the underlying network; and 3) the paths from each candidate next hop node of the underlying network to other nodes in that network. This analysis may be made locally with another software node of the global overlay network (e.g., a MOVE or a VINE endpoint) associated with the initial target node, or alternatively, the software node may query a routing engine to perform the analysis to determine the next hop node. The software node and/or the routing engine contains one or more optimization algorithms to perform the analysis. In some embodiments, the software node and/or routing engine can also use machine learning to continuously make changes to preferred routes, networks to be traversed, preferred next hop, or any other of the parameters discussed above. In block 120, one of the candidate next hop nodes of the underlying network is selected by the software node, which best meets the needs of the application based on the analysis performed in block 110 or the routing engine selects one of the candidate next hop nodes of the underlying network, which best meets the needs of the application based on the analysis performed in block 110 in response to the query of the software node and communicates the selection to the software node. In block 130, the software node instructs the initial target node of the underlying network to forward the packets to the candidate next hop node selected in block 120. In block 140, the initial target node forwards the packets to the selected candidate next hop node. In block 150, the blocks

110-140 are repeated along the path until the "packets" reach their destination.

[0072] In some embodiments, upon receipt of or the determination of the selected next hop node, the software node may forward the packets to another device and provide that device with the address of the selected next hop node (the new target node). In other embodiments, the software node managing and controlling the target node and any of the software nodes along the path may pass instructions to the application, directly or indirectly, including potentially providing the application with a replacement target node (a new DNS name or IP address to replace this target). The instructions may instruct the application to change its behavior for reasons, such as security, compliance or quality, or prompt the user for a decision. The software nodes of the network may also function as gateways, or insert gateways into the path, for example to do signaling or media interworking, or to encrypt/decrypt streams. The platform of the present invention has the capability to instantly and programmatically (no human involvement) deploy new software nodes (during the current application session) if necessary to meet the current or anticipated (according to machine learning or other algorithms) needs of the application or service. Each of the software network nodes of the global overlay network is acting according to instructions, identities, policies and feedback from the application, and the application managers, such that the application needs are determining the selected nodes of the underlying network, and the transport links between them. Whereas today's networks connect nodes according to the network's policies, the present invention connects applications and services according to their identities, policies and needs.

[0073] The global overlay network is integrated with the platform and application technology described above. The software nodes of the global overlay network may be connected as traditional networked nodes, and/or as peer-to-peer mesh nodes, and/or combinations of both. Unlike public Internet nodes, they do not all need to have public addresses or use standard Internet protocols and methods. The software nodes of the global overlay network are agnostic to the underlying network technology and network ownership. The software nodes can be located in public cloud, private cloud and Internet backbone environments, and on end user devices as describe earlier (e.g., MOVE and VINE). The software nodes may be applications,

VNFs, standalone devices running on commodity compute, purpose built appliances, or services embedded in network infrastructure such as gateways and session border controllers.

[0074] Any and all of the above methodologies may be applied differently for different sets of data to be transmitted by the same application. For example, the same application may have different policies and requirements, and thus issue different instructions to the software nodes, for different types of data or for the same data transmitted at different times or to different recipients.

[0075] While exemplary drawings and specific embodiments of the present disclosure have been described and illustrated, it is to be understood that that the scope of the invention as set forth in the claims is not to be limited to the particular embodiments discussed. Thus, the embodiments shall be regarded as illustrative rather than restrictive, and it should be understood that variations may be made in those embodiments by persons skilled in the art without departing from the scope of the invention as set forth in the claims that follow and their structural and functional equivalents.