Techniques are described for enabling users of a service provider network to create and configure “application profiles” that include parameters related to execution of user workloads at provider substrate extensions. Once an application profile is created, users can request the deployment of user workloads to provider substrate extensions by requesting instance launches based on a defined application profile. The service provider network can then automate the launch and placement of the user's workload at one or more provider substrate extensions using edge-optimized compute instances (e.g., compute instances tailored for execution within provider substrate extension environments). In some embodiments, once such edge-optimized instances are deployed, the service provider network can manage the auto-resizing of the instances in terms of various types of computing resources devoted to the instances, manage the lifecycle of instances to ensure maximum capacity availability at provider substrate extension locations, and perform other instance management processes.

A durability assessment system may receive a request, from a computing system, for a durability index describing an entity. The durability assessment system may determine the durability index based on information about the resource usage by the entity, such as a resource availability score or a resource allocation score. The durability assessment system may compare the obtained resource availability score and resource allocation score to ranges associated with a set of durability indices. Based on the comparison, the durability assessment system may determine a durability index for the entity. The durability index may indicate an ability of the entity to return accessed resources. In some cases, the durability assessment system may provide the durability index to an allocation computing system that is configured to determine whether to grant access to resources based on the durability index.

Embodiments described herein are directed to utilizing a tokenized system to manage network bandwidth. A total network bandwidth availability is determined for a network, and a total number of tokens is determined for that total network bandwidth. The system also determines a total number of users for the network. When a user sends a network usage request to use or access the network, the system selects and allocates a number of tokens for the user based on the network usage request, the total number of network users, and the total number of tokens. The user's device can then access and use the network if the user has a sufficient number of available tokens for that usage. The number of tokens for the user is reduced based on the amount of data used by the user.

Computing systems, devices, and associated methods of allocating computing resources in a distributed computing system to user requests are disclosed herein. In one example, a method includes when none of unallocated computing resources satisfy a constraint on a physical attribute in a received request for the computing resource, identifying one of the allocated resources in the distributed computing system that satisfies the constraint on the physical attribute and recursively determining whether one of the other allocated or unallocated resources in the distributed computing system that satisfies a constraint of a prior request. In response to determining that none of the allocated or unallocated resources satisfies the constraint of the prior request, assignment of the identified one of the allocated resources to the prior request is maintained.

A particular overall architecture for transmission over a bonded channel system consisting of two interconnected MoCA (Multimedia over Coax Alliance) 2.0 SoCs (Systems on a Chip) and a method and apparatus for the case of a “bonded” channel network. With a bonded channel network, the data is divided into two segments, the first of which is transported over a primary channel and the second of which is transported over a secondary channel.

A method and system for allocating tasks among processing devices in a data center. The method may include receiving a request to allocate a task to one or more processing devices, the request indicating a required bandwidth for performing the task, a list of predefined processing device groups connected to a host server and indicating availability of the processing device groups included therein for allocation of tasks and available bandwidth for each available processing device group, assigning the task to a processing device group having an available bandwidth greater than or equal to the required bandwidth for performing the task, and updating the list to indicate that each of the processing device group to which the task is assigned and other processing device group sharing at least one processing device is unavailable. The task may be assigned to an available processing device group having a lowest amount of power needed.

Some embodiments of the invention provide a novel method of performing network slice-based operations on a data message at a hardware forwarding element (HFE) in a network. For a received data message flow, the method has the HFE identify a network slice associated with the received data message flow. This network slice in some embodiments is associated with a set of operations to be performed on the data message by several network elements, including one or more machines executing on one or more computers in the network. Once the network slice is identified, the method has the HFE process the data message flow based on a rule that applies to data messages associated with the identified slice.

Apparatus and methods for optimizing bandwidth utilization and services in a data network infrastructure. In one embodiment, the data network is a managed Hybrid Fiber Coaxial (HFC) cable network, and the network infrastructure is configured to enable dynamic allocate of frequency bands to individual consumer premises device (e.g., DOCSIS-compliant cable modems). In one variant, the improved network infrastructure enables creation of virtual Service Groups (vSGs), and allocation of individual ones of the CM to such vSGs, to some degree irrespective of topological or “hardwired” location within the network. The allocations can be dynamic, and based on factors such as load balancing, evacuation of portions of the physical network topology (such as to support infrastructure upgrades or replacement), or for yet other reasons such as relating to subscriber tier or service level agreement (SLA).

Aspects of the subject disclosure may include, for example, obtaining quality of experience (QoE) information associated with a service from each communication device of a first group of communication devices over a communication network resulting in a group of QoE information. Each communication device of the first group of communication devices generates biometric information resulting in a group of biometric information. Each communication device of the first group of communication devices determines the QoE information based on the biometric information. Further embodiments include generating a group of QoE metrics associated with the service based on the group of QoE information, and adjusting an allocation of first network resources associated with the service based on the group of QoE metrics. Other embodiments are disclosed.

A system and method for managing access congestion in a computer network, the system and method including: determining a plurality of channels within the computer network via a subscriber channel module; determining a set of subscribers for each of the channels of the plurality of channels via the subscriber channel module; determining a congestion level of each of the channels of the plurality of channels via an analysis module; determining each subscriber's impact on the respective channel based on each subscriber's network usage via the analysis module; and determining a reallocation of the subscribers to balance the channels, based on the congestion level and each subscriber's network usage, via a distribution module.