Computing resources are managed in a computing environment comprising a computing service provider and an edge computing network. The edge computing network comprises computing and storage devices configured to extend computing resources of the computing service provider to remote users of the computing service provider. The edge computing network collects capacity and usage data for computing and network resources at the edge computing network. The capacity and usage data is sent to the computing service provider. Based on the capacity and usage data, the computing service provider, using a cost function, determines a distribution of workloads pertaining to a processing pipeline that has been partitioned into the workloads. The workloads can be executed at the computing service provider or the edge computing network.

Aspects of the present disclosure relate to allocating workloads to vRANs via programmable switches at far-edge cloud datacenters. Traditionally, traffic allocation is handled by dedicated servers running load-balancing software. However, rerouting RAN traffic to such servers increases both energy and capital costs, degrades end-to-end performance, and requires additional physical space, all of which are undesirable or even infeasible for a RAN far-edge datacenter. Since switches are located in the path of data traffic, workflow policies can be designed to inspect packet headers of incoming traffic, evaluate real-time network information, determine available vRAN instances, and update the packet headers to steer the incoming traffic for processing. As network conditions change, the workflow policies enable the switch to dynamically redirect workloads to alternative vRANs for processing. As a result, RAN processing efficiency and fault tolerance are improved—even with changing network conditions and spikes in data traffic.

Techniques for load balancing communication sessions in a networked computing environment are described herein. The techniques may include establishing a first communication session between a client device and a first computing resource of a networked computing environment. Additionally, the techniques may include storing, in a data store, data indicating that the first communication session is associated with the first computing resource. The techniques may further include receiving, at a second computing resource of the networked computing environment, traffic associated with a second communication session that was sent by the client device, and based at least in part on accessing the data stored in the data store, establishing a traffic redirect such that the traffic and additional traffic associated with the second communication session is sent from the second computing resource to the first computing resource.

Techniques for load balancing communication sessions in a networked computing environment are described herein. The techniques may include establishing a first communication session between a client device and a first computing resource of a networked computing environment. Additionally, the techniques may include storing, in a data store, data indicating that the first communication session is associated with the first computing resource. The techniques may further include receiving, at a second computing resource of the networked computing environment, traffic associated with a second communication session that was sent by the client device, and based at least in part on accessing the data stored in the data store, establishing a traffic redirect such that the traffic and additional traffic associated with the second communication session is sent from the second computing resource to the first computing resource.

A network controller can register WAN edge routers and WAN optimizers distributed across a WAN. The controller can receive a request to establish a WAN optimized connection between first and second hosts. The controller can identify a first WAN optimizer to perform first services (e.g., de-duplication, compression, application acceleration, caching, etc.) for first traffic from the first host to the second host and first complementary services for second traffic from the second host to the first host, and a second WAN optimizer for the second traffic and second complementary services for the first traffic. The controller can establish the optimized connection comprising a first path including the first host, WAN optimizer, and router; a second path including the first router and a second router, and a third path including the second router, WAN optimizer, and host. The controller can route the first and second traffic through the optimized connection.

Systems and methods are described to enable a so-called ‘unified slice’, wherein the unified slice is technology-independent, i.e., constructed from different networking technologies, and spans multiple operators. The method provides an abstraction of a network slice and its segments, and a way to coordinate the end-to-end slice information collection, slice segment configuration and activation across multiple types of networks and operators. The system of invention has the task of coordinating configuration of an end-to-end slice, with user-specified slice parameters, by communicating with the respective slice managers; It receives information to generate an abstract model of each slice segment, and sends the required slice segment attributes to these slice managers so that they can activate the segment after translating them according to capabilities of their network technology.

Embodiments of the present disclosure provide an Internet-of-Things resource access system and method. The system comprises a protocol management subsystem, a data conversion subsystem, and a load balancing subsystem. The protocol management subsystem is configured to obtain protocol frames from shared storage queues of protocol data packets, use a protocol stack to parse the protocol frames into original data payloads and provide the original data payloads to a data conversion subsystem; the data conversion subsystem is configured to perform protocol management, resource binding and data conversion, load Internet-of-Things resources and convert the original data payloads into observation data through multi-threaded concurrency; and the load balancing subsystem is configured to access the Internet-of-Things resources to the system through virtual IP, connect the Internet-of-Things resource to background service nodes through load balancing servers and send the protocol frames to the shared storage queues of the protocol data packets.

A method for processing data packets in a communication network includes establishing a path for a flow of the data packets through the communication network. At a node along the path having a plurality of aggregated ports, a port is selected from among the plurality to serve as part of the path. A label is chosen responsively to the selected port. The label is attached to the data packets in the flow at a point on the path upstream from the node. Upon receiving the data packets at the node, the data packets are switched through the selected port responsively to the label.

The disclosed computer-implemented method includes accessing information related to a playback session in which at least a portion of requested multimedia content is streamed over a network to a client electronic device. The method further includes accessing network topology information for the network to identify which route through the network was used to provide the requested multimedia content during the playback session, including indicating which end node was used to provide the multimedia content. Still further, the method includes accessing network steering factors that indicate why the requested multimedia content was steered through the identified network route, determining, based on the network steering factors, which end node would have been more suited to providing the requested multimedia content for the playback session, and then transferring the requested multimedia content to the determined end node for provisioning during subsequent playback sessions. Various other methods, systems, and computer-readable media are also disclosed.

The disclosed computer-implemented method includes accessing information related to a playback session in which at least a portion of requested multimedia content is streamed over a network to a client electronic device. The method further includes accessing network topology information for the network to identify which route through the network was used to provide the requested multimedia content during the playback session, including indicating which end node was used to provide the multimedia content. Still further, the method includes accessing network steering factors that indicate why the requested multimedia content was steered through the identified network route, determining, based on the network steering factors, which end node would have been more suited to providing the requested multimedia content for the playback session, and then transferring the requested multimedia content to the determined end node for provisioning during subsequent playback sessions. Various other methods, systems, and computer-readable media are also disclosed.