Abstract: Novel tools and techniques are provided for implementing Inference as a Service. In various embodiments, a computing system may receive a request to perform an AI/ML task on first data, the request including desired parameters, in some cases, without information regarding any of specific hardware, specific hardware type, specific location, or specific network for providing network services for performing the requested AI/ML task. The computing system may identify edge compute nodes within a network based on the desired parameters and/or unused processing capacity of each node. The computing system may identify AI/ML pipelines capable of performing the AI/ML task, the pipelines including neural networks utilizing pre-trained AI/ML models. The computing system may cause the identified nodes to run the identified pipelines to perform the AI/ML task.

Abstract: Novel tools and techniques are provided for implementing block-level, bit-mapped binary data access for parallel processing. In various embodiments, a computing system may cause a plurality of compute nodes to perform one or more tasks (e.g., artificial intelligence (“AI”) and/or machine learning (“ML”) tasks) on portions of decompressed data, after causing some, but not all, of the compute nodes to decompress the entire compressed data. In some cases, the computing system may cause each compute node to decompress portions of the compressed data. Once the data has been decompressed and/or bit-mapped, each compute node may be caused to perform at least one task (e.g., AI/ML task) on its assigned portion of the decompressed data. Different portions of the same decompressed data may be processed with the same (AI/ML) algorithm, or the same portions of the same decompressed data may be processed with a different (AI/ML) algorithm, or a combination thereof.

Abstract: Novel tools and techniques are provided for implementing object-based changes to filter-intent over multicast or publication/subscription (“Pub/Sub”) distribution. In various embodiments, a computing system (e.g., a managed device among a plurality of managed devices and/or its corresponding agent) may receive, from a network filter orchestration conductor, a global filter-intent list including a first filter intent that references a corresponding filter-intent object. The computing system may determine whether the at least one first filter intent applies to the managed device. If so, the computing system may translate the at least one first filter intent into a first filter that is specific to a first configuration of the managed device, in some cases, by building the first filter based at least in part on the at least one first filter intent. The computing system may subsequently apply the first filter to one or more network communications handled by the managed device.

Abstract: The present application describes a system and method for utilizing a tunnel in a networking routing protocol to provide a network segment access to additional servers when certain load balancing trigger events are detected.

Abstract: The present application describes a system and method for utilizing a tunnel in a networking routing protocol to provide a network segment access to additional servers when certain load balancing trigger events are detected.

Abstract: Examples of the present disclosure describe systems and methods for restricting the routing scope of an anycast service. In aspects, network traffic and/or network performance data may be received by a network device and/or an endpoint device in an anycast environment. The received network traffic and/or network performance may be evaluated using one or more automated logic systems or algorithms to identify indicators of potential failure or performance degradation for one or more devices in the anycast environment. Upon identifying one or more such indicators, a routing table of one or more network devices in the anycast environment may be dynamically modified to restrict routes to particular endpoint destinations. The network devices in the anycast environment may subsequently route network traffic according to the modified routing table(s).

Abstract: The present application describes a system and method for utilizing a tunnel in a networking routing protocol to provide a network segment access to additional servers when certain load balancing trigger events are detected.