At&t Intellectual Property I, L.p.

Fujitsu Corporation

Edward Daniels

Kaustubh Joshi

Matti A Hiltunen

Abhigyan

Rittwik Jana

Han Q Nguyen

Gueyoung Jung

Yasuhide Matsumoto

Hiroshi Otsuka

The described technology is generally directed towards a cellular network area optimizer. The area optimizer observes cellular network conditions at multiple radio access network (RAN) nodes within a target area. Based on observed conditions, the area optimizer applies a set of parameter values at the multiple RAN nodes. The set of parameter values enhances the overall throughput, while maintaining or improving connection retainability and accessibility, of the multiple RAN nodes under the observed conditions. The area optimizer learns different sets of parameter values to apply in response to different observed conditions by making parameter value adjustments and observing the effect of the adjustments on overall throughput of the RAN nodes in the target area.

Cellular network area optimizer

A system can include a gateway, a plurality of network function nodes, and a distributed load balancer including load balancer nodes each having a flow table portion stored thereon. The load balancer nodes can form a node chain having a tail and head nodes. A load balancer node can receive a packet from the gateway. In response, the load balancer node can generate a query, directed to the tail node, that identifies the packet and a network function identifier associated with a network function node that is proposed to handle a connection. The tail node can determine whether an entry for the connection exists in a flow table portion associated with the tail node. If not, the tail node can initiate an insert request for writing the entry for the connection via the head node. The entry can then be written to all load balancer nodes in the node chain.

Seamless reconfiguration of distributed stateful network functions

According to one aspect disclosed herein, a system can include a set of node peers, including a first subset implemented in software and a second subset implemented in hardware. The first subset can include a software node. The second subset can include a hardware node that includes a hardware cache, a processor, and a memory that stores computer-executable instructions. The hardware node can receive, from a network, a packet, and can determine if data that identifies a path associated with the packet is stored in the hardware cache. If not, the hardware node can query the software node to identify the path associated with the packet, and can receive, in response from the software node, the data that identifies the path, which then can be stored in the hardware cache. The hardware node can forward, along the path, the packet to a network element.

Extending distributed hash table-based software network functions to switching hardware

Concepts and technologies are disclosed herein for a distributed stateful load balancer. A first load balancer can store a first portion of a distributed hash table including data over a first range of hash keys. The first load balancer can be in communication with a second load balancer that stores a second portion of the distributed hash table including data over a second range of hash keys. The first load balancer can receive a packet, where a hash value calculated for the packet does not fall in the first range of hash keys. The first load balancer can identify, based on the hash value, that the second load balancer stores data that identifies a path associated with the packet and query the second load balancer to identify the path. The first load balancer can receive the data and forward, along the path, the packet to a network element.

Distributed stateful load balancer

Methods, systems, and products predict locations of failures. Messages historically received are analyzed to predict a future failure. A pattern in the messages historically received is used to retrieve a template. A location of the future failure is predicted from the template.

Growing community of inventors

Wadsworth, OH, United States of America

Edward Daniels