Open Invention Network LLC

Telefonaktiebolaget Lm Ericsson (publ)

Kaloom Inc

Coho Data Inc.

Geoffrey Lefebvre

Andrew Kent Warfield

Daniel Stodden

Brendan Anthony Cully

Jacob Taylor Wires

Timothy John Deegan

Keir Fraser

Kevin Jamieson

Jean Maurice Guy Guyader

Dutch Meyer

Christopher Clark

Suresh Krishnan

Daniel Richer

Harsimrandeep Singh Pabla

Ravi Manghirmalani

Makan Pourzandi

Catherine Truchan

Eric Dyke

Jon Maloy

Erik Rubow

Systems, methods and devices for distributed memory management comprising a network component configured for network communication with one or more memory resources that store data and one or more consumer devices that use data, the network component comprising a switching device in operative communication with a mapping resource, wherein the mapping resource is configured to associate mappings between data addresses associated with memory requests from a consumer device relating to a data object and information relating to a storage location in the one or more memory resources associated with the data from the data object, wherein each data address has contained therein identification information for identifying the data from the data object associated with that data address; and the switching device is configured to route memory requests based on the mappings.

Systems, methods and devices for integrating end-host and network resources in distributed memory

Server and methods for performing synchronization and resynchronization of networking information with client devices. During standard operations, client networking data discovered by a given client device are received by the server from the given client device, stored by the server, and forwarded by the server to other client devices. Client networking data from the other client devices received and stored by the server are forwarded to the given client device. Upon occurrence of a pre-defined event at the given client device preventing exchange of information between the given client device and the server, a resynchronization of the client networking data is performed between the given client device and the server. The synchronization and resynchronization mechanisms are based on unique version numbers respectively generated by the server and the given client device.

Server and methods for synchronizing networking information with client devices

Systems, methods and devices relating to coordinated network communication (e.g. transport-layer communication) of client requests and client request responses between a client and a distributed network service system, the network service nodes of the distributed network service system comprising a storage resource, a network interface, and a computer processor module for sending a coordinated network communication of data request responses to the client upon receipt of (1) network communication of client requests from clients; or (2) communication data channel information from another network service node. There is also provided a network switching device for managing a coordinated network communication of data transactions between clients and a distributed network service system comprising a plurality of network service nodes, the network switching device configured to manage higher-layer data units to coordinate a network communication of data transactions between clients and a distributed network service system.

Methods, devices and systems for coordinating network-based communication in distributed server systems with SDN switching

Computing device and methods for performing synchronization and resynchronization of networking information with a server. During standard operations, local networking data discovered by the computing device are transmitted to the server, and server networking data (discovered by the server) are received from the server. Upon occurrence of a pre-defined event at the computing device preventing exchange of information between the computing device and the server, a resynchronization of the local networking data and server networking data is performed. The synchronization and resynchronization mechanisms are based on unique version numbers respectively generated by the computing device and the server.

Computing device and methods for synchronizing networking information with a topology server

The present disclosure relates to distributed computing systems, and methods and devices relating thereto, said systems comprising a plurality of distributed computing hosts that process client requests from a computing client; and a programmable network switch, comprising two or more network ports for communicatively interfacing said plurality of distributed computing hosts and said computing client, said programmable network switch operable to forward client requests to a given network port based on forwarding rules related to client source addresses; wherein the programmable network switch replaces a request destination address with an incoming network port indicator that identifies on which network port a given client request is received; and wherein the programmable network switch forwards, without reference to the forwarding rules, request responses to the network port associated with the request source address and replaces the response source address with the request destination address associated with the distributing computing system.

Systems, methods and devices for scalable expansion of rules-based forwarding paths in network communications devices for distributed computing systems

A system and method for steering traffic through a set of services is provided. A service path or chain is assigned to a received packet based on a classification of the packet. A position and/or a direction of the traffic in the service path can be determined based on the previous service performed on the traffic. A next destination for the traffic can be assigned in accordance with the assigned service chain and determined position and direction information.

Chaining of inline services using software defined networking

A method for providing a parser by a network node, which includes a table defining a plurality of rules and a plurality of match fields required for extraction from an incoming data packet header in order to apply each of the respective rules is described. The method comprises determining at least one occurrence in the table of a first match field that requires extraction; and determining a parser for use in accordance with the at least one determined occurrence of the first match field. Furthermore, the step of determining at least one occurrence of the first match field comprises initializing a counter associated with the first match field and incrementing the counter for each of the plurality of rules that requires extraction of the first match field. And the step of determining a parser comprises determining the parser for use in accordance with the counter. A network node for carrying this method is also described.

Method for providing a parser to extract information from fields of a data packet header

In one aspect, the teachings herein provide point-to-point communication between all the virtual machines (VMs) belonging to the same tenant in a cloud-based virtual network environment without using any intermediate routing functions, and while providing full virtualization at the L2 and L3 levels. That is, the approach preserves completely isolated address spaces for each tenant at both L2 and L3 levels and supports multiple subnets for each tenant, and thereby allows tenants to define their own virtual networks without requiring the addition of expensive centralized routing and without requiring inter-subnet communications for the same tenant to traverse sub-optimal paths, as happens with software-based routers implemented in one of the VMs belonging to the tenant. Moreover, the methods and apparatuses taught herein provide a distributed solution for communicating between subnets belonging to the same virtual or logical network, thereby avoiding single-point-of-failure issues and offering improved scalability.

Method and apparatus for exchanging IP packets among network layer 2 peers

A load balancer in a communication network tracks active network flows using a Bloom filter and takes a snapshot of the Bloom filter at the time of a scaling event. The load balancer uses the Bloom filter snapshot to differentiate packets belonging to pre-existing network flows from packets belonging to new network flows. Packets belonging to pre-existing network flows continue to be distributed according to a mapping function in use prior to the scaling event. Packets belonging to new network flows are distributed according to a new mapping function.

Growing community of inventors

Montreal, Canada

Geoffrey Lefebvre