In order to enable dynamic scaling of network services at the edge, novel systems and methods are provided to enable addition of add new nodes or removal of existing nodes while retaining the affinity of the flows through the stateful services. The methods provide a cluster of network nodes that can be dynamically resized to handle and process network traffic that utilizes stateful network services. The existing traffic flows through the edge continue to function during and after the changes to membership of the cluster. All nodes in the cluster operate in active-active mode, i.e., they are receiving and processing traffic flows, thereby maximizing the utilization of the available processing power.

Disclosed are examples of systems, apparatus, devices, computer program products, and methods implementing aspects of a decentralized content fabric. In some implementations, one or more processors are configured to execute a software stack to define a fabric node of a plurality of fabric nodes of an overlay network situated in an application layer differentiated from an internet protocol layer. The defined fabric node is configured to: obtain a request for digital content from a client device; obtain, from one or more of the plurality of fabric nodes, a plurality of content object parts of a content object representing, in the overlay network, at least a portion of the digital content; generate consumable media using: raw data stored in the content object parts, metadata stored in the content object parts, and build instructions stored in the content object parts; and provide the consumable media to the client device. In some instances, the consumable media is further generated using a digital contract stored in a blockchain.

Disclosed are examples of systems, apparatus, devices, computer program products, and methods implementing aspects of a decentralized content fabric. In some implementations, one or more processors are configured to execute a software stack to define a fabric node of a plurality of fabric nodes of an overlay network situated in an application layer differentiated from an internet protocol layer. The defined fabric node is configured to: obtain a request for digital content from a client device; obtain, from one or more of the plurality of fabric nodes, a plurality of content object parts of a content object representing, in the overlay network, at least a portion of the digital content; generate consumable media using: raw data stored in the content object parts, metadata stored in the content object parts, and build instructions stored in the content object parts; and provide the consumable media to the client device. In some instances, the consumable media is further generated using a digital contract stored in a blockchain.

A method is disclosed for autonomously routing data using in a peer-to-peer computer network includes automatically updating a peer-to-peer computer network. The method includes automatically sending pulse messages from a first node to neighbor nodes and candidate nodes, receiving return pulses by the first node from at least some of the neighbor nodes and the candidate nodes, calculating round-trip times (RTTs) between the first node and the neighbor nodes or the candidate nodes based on the pulse messages and the return pulses, sorting the nodes in the neighbor nodes and the candidate nodes into orbital bins based on RTTs, and automatically selecting and adding a node from one of the orbital bins based on the RTTs to updated neighbor nodes for the first node, and routing data from the first node to a second node via a relay node in the peer-to-peer computer network.

A method is disclosed for autonomously routing data using in a peer-to-peer computer network includes automatically updating a peer-to-peer computer network. The method includes automatically sending pulse messages from a first node to neighbor nodes and candidate nodes, receiving return pulses by the first node from at least some of the neighbor nodes and the candidate nodes, calculating round-trip times (RTTs) between the first node and the neighbor nodes or the candidate nodes based on the pulse messages and the return pulses, sorting the nodes in the neighbor nodes and the candidate nodes into orbital bins based on RTTs, and automatically selecting and adding a node from one of the orbital bins based on the RTTs to updated neighbor nodes for the first node, and routing data from the first node to a second node via a relay node in the peer-to-peer computer network.

A method is disclosed for autonomously routing data using in a peer-to-peer computer network includes identifying a destination node to receive a data transfer, storing IDs of neighbor nodes sorted into orbital bins according to round-trip times (RTTs) between a source node and the neighbor nodes, sending one or more path packages from the source node to the destination node in a first direct data path from the source node to the destination node, sending path packages from the source node to the neighbor nodes, sending one or more path packages comprising updated hop information from a first hop node to the destination node, calculating total one-way latencies and performance metrics respectively for the path packages received by the destination node, and selecting a relayed data path for the data transfer from the source node to the destination node.

A load balancing method and a device. A network device performs hash calculation on a first traffic flow to obtain a first hash value corresponding to the first traffic flow, and determines, based on a first mapping relationship between the first hash value and a first member port in a link aggregation group, that an egress port of the first traffic flow is the first member port. The network device determines a current first bandwidth of the first member port and adjusts the first mapping relationship to a second mapping relationship between the first hash value and a second member port based on the first bandwidth, to forward a subsequently received first traffic flow through the second member port. Therefore, the network device can meet a requirement of a high bandwidth.

A load balancing method and a device. A network device performs hash calculation on a first traffic flow to obtain a first hash value corresponding to the first traffic flow, and determines, based on a first mapping relationship between the first hash value and a first member port in a link aggregation group, that an egress port of the first traffic flow is the first member port. The network device determines a current first bandwidth of the first member port and adjusts the first mapping relationship to a second mapping relationship between the first hash value and a second member port based on the first bandwidth, to forward a subsequently received first traffic flow through the second member port. Therefore, the network device can meet a requirement of a high bandwidth.

Some embodiments provide a method for a forwarding element that receives a packet. The method determines whether the packet matches any flow entries in a first cache that uses a first type of algorithm to identify matching flow entries for packets. When the packet does not match any flow entries in the first cache, the method determines whether the packet matches any flow entries in a second cache that uses a second, different type of algorithm to identify matching flow entries for packets. The method executes a set of actions specified by a flow entry matched by the packet in one of the first and second caches.

Examples described herein relate to a network interface that includes an initiator device to determine a storage node associated with an access command based on an association between an address in the command and a storage node. The network interface can include a redirector to update the association based on messages from one or more remote storage nodes. The association can be based on a look-up table associating a namespace identifier with prefix string and object size. In some examples, the access command is compatible with NVMe over Fabrics. The initiator device can determine a remote direct memory access (RDMA) queue-pair (QP) lookup for use to perform the access command.