Abstract: A network device includes a statelet storage storing statelets that retain state information associated with a packet flow through the network device and that the network device can interact with to control processing performed on packets of the data flow. The network device implements a set of instructions that interpret commands in the data packets to manage and interact with statelets. The statelets in the statelet storage are organized by a statelet key that is derived from information identifying the packet flow. Responsive to the commands in the packets, the network device can create, read, write, or delete statelets from the statelet storage. The statelet storage includes multiple statelets each statelet including multiple fields. The network device may access the statelets to control/monitor a packet flow using information in a network data plane without receiving control information from a network control plane.

Abstract: A method implemented by a network element (NE) in a network, comprising receiving, by the NE, an advertisement comprising preferred path route (PPR) information representing a PPR from a source to a destination in the network, the PPR information comprising a PPR identifier (PPR-ID) and a plurality of PPR description elements (PPR-PDEs) each representing an element on the PPR, receiving, by the NE, a data packet comprising the PPR-ID, and forwarding, by the NE, the data packet having the PPR-ID to a next element on the PPR based on the plurality of PPR-PDEs.

Abstract: A mechanism is disclosed for implementing conditional commands carried by network data packets. A data flow including a data packet is received. The data packet includes a conditional command. A condition and a command are obtained from the conditional command. The mechanism determines that the condition is satisfied. Based on the determination that the condition is satisfied, the command is executed to alter handling of the data flow, alter handling of the data packet, or alter a context for the data flow.

Abstract: A Self-Describing Packet block (SDPB) is defined that allows concurrent processing of various fixed headers in a packet block defined to take advantage of multiple cores in a networking node forwarding path architecture. SPDB allows concurrent processing of various pieces of header data, metadata, and conditional commands carried in the same data packet by checking a serialization flag set upon creation of the data packet, without needing to serialize the processing or even parsing of the packet. When one or h more commands in one or more sub-blocks may be processed concurrently, the one or more commands are distributed to multiple processing resources for processing the commands in parallel. This architecture allows multiple unique functionalities each with their own separate outcome (execution of commands, doing service chaining, performing telemetry, allows virtualization and path steering) to be performed concurrently with simplified packet architecture without incurring additional encapsulation overhead.

Abstract: The system and method of the present disclosure relates to technology for a network system supporting multiple topology transparent zones (TTZs) on an edge node. TTZs have a topology including TTZ nodes consisting of edge nodes and internal nodes connected together by internal links. The edge nodes may also be connected to external nodes outside of the TTZ. The edge node is responsible for generating and distributing link state advertisements (LSAs) to generate virtual paths from the edge node to other edge nodes in each of the TTZs. Each TTZ sharing the edge node and each external node neighboring one of the other edge nodes of the TTZs is virtualized by distribution of the LSAs. However, the internal topology of each TTZ is concealed from each of the other TTZs sharing the edge node and from any neighboring external node connected to one of the other edge nodes.

Abstract: A network element (NE) in a zone routing network includes a memory, receiver, processor, and transmitter. The memory stores a routing information base (RIB) identifying a first next-hop node for a first path in the network to reach a network controller of the network via a first communication node in the network. The first communication node is directly associated with the network controller. The receiver receives a first label-switched path (LSP) creation request message requesting creation of a first LSP through the network. The first LSP creation request message indicates an ingress node of the first LSP. The processor determines that the NE corresponds to the ingress node of the first LSP. The transmitter sends a first LSP creation response message towards the network controller via the first communication node according to the first path to indicate a creation status of the first LSP.

Abstract: A method implemented by a sending host implemented as a network element (NE) comprises receiving, by a receiver of the sending host, a set of locators from a distributed mapping system, each of the locators in the set of locators being an address associated with the receiving host, storing, in a memory of the sending host, the set of locators in an entry of a local identifier (ID)-to-locator database in association with an ID of the receiving host, transmitting, by a transmitter of the sending host, a plurality of packets to a plurality of endpoint routers (xTRs), each of the xTRs associated with each of the locators in the set of locators, and receiving, by the receiver of the sending host, a response from one of the xTRs in response to sending the plurality of packets to the plurality of xTRs, a locator associated with the one of the xTRs being an address of a current location of the receiving host.

Abstract: A method implemented by a network element (NE) configured as a temporal tunnel service (TTS) controller, comprising computing, via a processor of the NE, a path in a network for a temporal label switched path (LSP), wherein the path satisfies a constraint in a scheduled time interval comprising a predetermined start time and a predetermined end time, reserving, at a current time prior to the predetermined start time via the processor, a first network resource on a link along the path computed for the temporal LSP, wherein the first network resource is reserved for the temporal LSP to carry traffic in the scheduled time interval, and sending, at the predetermined start time via a transmitter of the NE, a LSP creation request to a node associated with the temporal LSP to request creation of the temporal LSP along the path in the network.

Abstract: A method implemented by a network controller in a network, comprising receiving, by a temporal label switched path (T-LSP) manager of the network controller, a request for creating a temporal LSP in a scheduled time interval; computing, by a temporal path computation element (T-PCE) of the network controller, a shortest path in a network satisfying a constraint for the label switched path (LSP) in the scheduled time interval; reserving, in a temporal traffic engineering database (T-TED) of the network controller, resources on each link the LSP traverses for the scheduled time interval; and initiating, at a beginning of the scheduled time interval by the T-LSP manager, setup of the LSP in the network through sending a LSP creation request to a path computation client (PCC) on an ingress node of the LSP.

Abstract: A method implemented by a path computation element centralized controller (PCECC), the method comprises: receiving a service request to provision for a service from a first edge node and a second edge node in a network; computing a path for a label switched path (LSP) from the first edge node to the second edge node in response to the service request; reserving label information for forwarding traffic of the service on the LSP; and sending a label update message to a third node on the path to facilitate forwarding of the traffic of the service on the path, wherein the label update message comprises the label information.

Abstract: Systems and methods for providing a session layer connection between two or more network endpoints. Session layer connections created and maintained using embodiments of the present disclosure use endpoint identifiers (EIDs) and allow for session layer continuity when a lower-layer connection is broken because of network failures or the movement of an endpoint from one network connection to another.

Abstract: A method of avoiding routing interruption during transition to a topology transparent zone (TTZ) is described. The method includes virtualizing a path from an edge router to another edge router within the TTZ in a first direction without removing original links, wherein the original links are links from the edge router to its neighboring routers within the TTZ, receiving, by the edge router within the TTZ, a router link state advertisement (LSA) from the another edge router within the TTZ while the original links remain in place, wherein the LSA comprises a path from the another edge router to the edge router in a second direction different from the first direction, and removing, by the edge router within the TTZ, the original links after the LSA has been received from the another edge router and the virtualization of the path is complete.

Abstract: The system and method of the present disclosure relates to technology for a network system supporting multiple topology transparent zones (TTZs) on an edge node. TTZs have a topology including TTZ nodes consisting of edge nodes and internal nodes connected together by internal links. The edge nodes may also be connected to external nodes outside of the TTZ. The edge node is responsible for generating and distributing link state advertisements (LSAs) to generate virtual paths from the edge node to other edge nodes in each of the TTZs. Each TTZ sharing the edge node and each external node neighboring one of the other edge nodes of the TTZs is virtualized by distribution of the LSAs. However, the internal topology of each TTZ is concealed from each of the other TTZs sharing the edge node and from any neighboring external node connected to one of the other edge nodes.

Abstract: A mechanism is disclosed for implementing conditional commands carried by network data packets. A data flow including a data packet is received. The data packet includes a conditional command. A condition and a command are obtained from the conditional command. The mechanism determines that the condition is satisfied. Based on the determination that the condition is satisfied, the command is executed to alter handling of the data flow, alter handling of the data packet, or alter a context for the data flow.

Abstract: A method implemented in an inter-provider cloud rendezvous point (CRP-IP). The method includes receiving, via a receiver of the CRP-IP, a Register request from a first service provider cloud rendezvous point (CRP-SP) in a first service provider (SP) network domain, the Register request indicating a first portion of a virtual extensible network (VXN) with a global scope is hosted by the first SP network domain; and transmitting, via a transmitter of the CRP-IP, a Report message to a second CRP-SP in a second SP network domain hosting a second portion of the VXN, the Report message indicating the first portion of the VXN is hosted by the first SP network domain.

Abstract: A method implemented in an inter-provider cloud rendezvous point (CRP-IP). The method includes receiving, via a receiver of the CRP-IP, a Register request from a first service provider cloud rendezvous point (CRP-SP) in a first service provider (SP) network domain, the Register request indicating a first portion of a virtual extensible network (VXN) with a global scope is hosted by the first SP network domain; and transmitting, via a transmitter of the CRP-IP, a Report message to a second CRP-SP in a second SP network domain hosting a second portion of the VXN, the Report message indicating the first portion of the VXN is hosted by the first SP network domain.

Abstract: The system and method of the present disclosure relates to technology for a network system supporting multiple topology transparent zones (TTZs) on an edge node. TTZs have a topology including TTZ nodes consisting of edge nodes and internal nodes connected together by internal links. The edge nodes may also be connected to external nodes outside of the TTZ. The edge node is responsible for generating and distributing link state advertisements (LSAs) to generate virtual paths from the edge node to other edge nodes in each of the TTZs. Each TTZ sharing the edge node and each external node neighboring one of the other edge nodes of the TTZs is virtualized by distribution of the LSAs. However, the internal topology of each TTZ is concealed from each of the other TTZs sharing the edge node and from any neighboring external node connected to one of the other edge nodes.

Abstract: Systems and methods for providing a session layer connection between two or more network endpoints. Session layer connections created and maintained using embodiments of the present disclosure use endpoint identifiers (EIDs) and allow for session layer continuity when a lower-layer connection is broken because of network failures or the movement of an endpoint from one network connection to another.

Abstract: A network element (NE) configured as a local cloud switch point (CSP) comprises a memory configured to store remote virtual routing information in association with an identifier (ID) of a remote CSP in a mapping table, wherein the remote virtual routing information is associated with a remote virtual network associated with the remote CSP, and wherein the remote virtual routing information comprises a virtual extensible network (VXN) type identifier (ID), an address family ID (AFI), and a subsequent AFI (SAFI). The NE further comprises a processor coupled to the memory and configured to perform encapsulation on a data packet to generate a encapsulated data packet by adding an outer header to the data packet based on the AFI, adding a virtualization header to the data packet based on the VXN type ID, and adding an inner header to the data packet based on the SAFI. The NE further comprises a transmitter coupled to the processor and configured to transmit the encapsulated data packet to the remote CSP.

Abstract: A network element (NE) in a zone routing network includes a memory, receiver, processor, and transmitter. The memory stores a routing information base (RIB) identifying a first next-hop node for a first path in the network to reach a network controller of the network via a first communication node in the network. The first communication node is directly associated with the network controller. The receiver receives a first label-switched path (LSP) creation request message requesting creation of a first LSP through the network. The first LSP creation request message indicates an ingress node of the first LSP. The processor determines that the NE corresponds to the ingress node of the first LSP. The transmitter sends a first LSP creation response message towards the network controller via the first communication node according to the first path to indicate a creation status of the first LSP.