Nebbiolo Technologies, Inc.

Amazon Technologies, Inc.

Cisco Technology, Inc.

Palani Chinnakannan

Pankaj Bhagra

Flavio Giovanni Bonomi

Chandra Shekhar Joshi

Kannan Devarajan

Lomash Kumar

Venkata Harish Kommaraju

Nicolas Pouyez

Christopher Michael Morrissey

David Delano Ward

Sami Boutros

Dhananjaya Rao

Rex Emmanuel Fernando

Mohammed Ismael Tatar

John H W Bettink

Ruchir Tewari

Sureshkumar Kaliannan

Ravi Bhagavatula

Varadarajan Devnath

Thushar Gowda

Arvind Karuppasamy

Thiru Narayanan

A digital twin service of a provider network allows a user to build high fidelity models ('digital twins') for components and for systems of multiple components. For a given component, the digital twin service receives, from a user, identification of interactions for the component (e.g., inputs/outputs at the boundary), identification of data sources at the system (e.g., sensor that stream data), and/or a definition of behavior of the component (e.g., one or more behavior functions to simulate the component). The digital twin service builds a digital twin model of the component based at least on the interactions, the data sources, and the behavior. The digital twin service may also receive, from the user, an indication to bind one or more outputs of the digital twin model to one or more inputs of at least another of the digital twin models to build a digital twin model of the system.

Digital twin models for components and systems of components

A feature deployment service of a provider network may deploy feature processing units (FPUs) to implement data processing features at both a provider network and edge devices. The use of FPUs may allow a client to use new features at the edge, without delays due to compliance/testing or software upgrades. An FPU includes a model and compute logic that are used to implement a data processing feature. A feature processing service deploys the FPU to an FPU engine at the provider network and also deploys the FPU to edge devices of the client's network that each include an edge FPU engine. The FPU engine at the provider network and the edge FPU engine at each edge device conform to a common specification/API, allowing deployment and use of the same FPU/data processing features at both the cloud and the edge.

Continuous deployment and orchestration of feature processing units in a managed cloud and a provider network for consistent data processing

Systems and methods for security of industrial data streams are provided herein. Methods according to various embodiments include provisioning a fogNode that is communicatively coupled with a fog cloud manager through a forwarder of the fogNode and providing a fogLet within the fogNode, the fogLet communicating with a plurality of operational technology devices. Embodiments include providing fogLet identification information using hardware root of trust of the fogNode, the hardware root of trust of the fogNode being a Trusted Platform Module (TPM) of the fogNode. Embodiments further comprise communicating operational device authentication information with fogLet identification information to a third party tenant application, the third party tenant application validating industrial data streams from the operational technology devices by communicating the operational device authentication information with the fogLet identification information to a third party cloud application.

Systems and methods for securing industrial data streams with a fog root of trust

Adaptive scheduling of compute functions in a fog network is described herein. An example method includes synchronizing kernel and hypervisor scheduling of applications used to control one or more edge devices using a global schedule, wherein the global schedule comprises timeslots for applications, and adapting the timeslots in real-time or near real-time based on application-based time related feedback that is indicative of time delays.

Adaptive scheduling for edge devices and networks

Disclosed herein are enhancements for operating an input/output (I/O) management cluster with end I/O devices. In one implementation, a method of operating an I/O cluster includes, in a first I/O management node of the I/O management cluster, executing a first application to manage data for an I/O device communicatively coupled via at least one switch to the first I/O management node. The method further provides identifying a failure in the first I/O management node related to processing the data for the I/O device and, in response to the failure, configuring the at least one switch to communicate the data for the I/O device with a second I/O management node of the I/O management cluster. The method also includes, in the second I/O management node and after configuring the at least one switch, executing a second application to manage the data for the I/O device.

High availability input/output management nodes

Enterprise grade security for integrating multiple computing domains with a public cloud is provided herein. An example system a forwarder that provides one-way data publishing to a public cloud and a data bus that provides domain-to-domain messaging between a plurality of domains. At least one of the plurality of domains includes operational technology infrastructure devices and operational technology virtual machines. The operational technology virtual machines are communicatively coupled to the operational technology infrastructure devices using one or more operational technology switches. The operational technology switches isolates the operational technology infrastructure devices and facilitates one-way communication and prevents bidirectional communication to the operational technology infrastructure devices from the public cloud.

Enterprise grade security for integrating multiple domains with a public cloud

Provided herein are exemplary systems and methods for a fog computing facilitated flexible factory including establishing a physical production process as part of a work cell, establishing a sensing process as part of the work cell for the physical production process, establishing a monitoring process for the sensing process and the physical production process, establishing a managing process for the monitoring process, the sensing process and the physical production process, establishing a planning process for the managing process, the monitoring process, the sensing process and the physical production process, and establishing a fog node as part of the work cell for all of the processes.

Fog computing facilitated flexible factory

An example method for instantiating a network using a network function virtualization infrastructure (NVFI) pod in a network environment is provided and includes receiving a request to instantiate a logical network topology in the NFVI pod that includes a pre-selected set of interconnected pre-configured hardware resources, the abstract network topology including a virtual network function (VNF) forwarding graph (FG), distilling the VNF FG into various interconnected VNFs, deploying various VNFs of the VNF FG to a plurality of virtual machines, and instantiating the network topology on appropriate hardware resources in the NFVI pod. In specific embodiments, the pre-selected set of interconnected pre-configured hardware resources includes a plurality of compute servers, a plurality of controller servers, a set of network resources, and a plurality of storage resources. It may also include a high performance virtual switch operating in a user space of respective memory elements of a plurality of servers.

Network function virtualization infrastructure pod in a network environment

Policers receive packets of flows of packet traffic, which are to be communicated to monitored resource. The utilization levels of the monitored resource are induced by these flows of packet traffic. Based on the observed utilization levels (including possibly measured durations in one or more of these utilization levels), a determination is made if, and how to adjust policers for policing their respective flow, with policers being adjusted accordingly. In this manner, adaptive policers (typically located remotely from the monitored resource) are adjusted in response to one or more utilization levels (including possible durations at these utilization levels&#x2014;i.e., a persistence of the congestion for the resource) of one or more monitored resources, with these identified utilization levels (and possibly durations) used in determining how much to modify a policing rate.

Growing community of inventors

San Jose, CA, United States of America

Palani Chinnakannan