Hewlett Packard Enterprise Development LP

Anu Mercian

Puneet Sharma

Diman Zad Tootaghaj

Charles F Clark

Chinlin Chen

Terrel L Morris

Jean Tourrilhes

Faraz Ahmed

Lianjie Cao

David Rodrigues Pinheiro

Derek Spock

Vivek Adarsh

Ying Zhang

Sujata Banerjee

Joon-Myung Kang

Renato Chaves De Aguiar

Mario Antonio Sanchez

Raphael Amorim Dantas Leite

Embodiments described herein are generally directed to an edge-CaaS (eCaaS) framework for providing life-cycle management of containerized applications on the edge. According to an example, declarative intents are received indicative of a use case for which a cluster of a container orchestration platform is to be deployed within an edge site that is to be created based on infrastructure associated with a private network. A deployment template is created by performing intent translation on the declarative intents and based on a set of constraints. The deployment template identifies the container orchestration platform selected by the intent translation. The deployment template is then executed to deploy and configure the edge site, including provisioning and configuring the infrastructure, installing the container orchestration platform on the infrastructure, configuring the cluster within the container orchestration platform, and deploying a containerized application or portion thereof on the cluster.

Deployment and configuration of an edge site based on declarative intents indicative of a use case

Example implementations relate to edge acceleration by offloading network dependent applications to a hardware accelerator. According to one embodiment, queries are received at a cluster of a container orchestration platform. The cluster includes a host system and a hardware accelerator, each serving as individual worker machines of the cluster. The cluster further includes multiple worker nodes and a master node executing on the host system or the hardware accelerator. A first worker node executes on the hardware accelerator and runs a first instance of an application. A distribution of the queries is determined among the worker machines based on a queuing model that takes into consideration the respective compute capacities of the worker machines. Responsive to receipt of the queries by the host system or the hardware accelerator, the queries are directed to the master node or one of the worker nodes in accordance with the distribution.

Directing queries to nodes of a cluster of a container orchestration platform distributed across a host system and a hardware accelerator of the host system

An event analysis system is provided. During operation, the system can determine an event description associated with the switch from an event log of the switch. The event description can correspond to an entry in a table in a switch configuration database of the switch. A respective database in the switch can be a relational database. The system can then obtain an event log segment, which is a portion of the event log, comprising the event description based on a range of entries. Subsequently, the system can apply a pattern recognition technique on the event log segment based on the entry in the switch configuration database to determine one or more patterns corresponding to an event associated with the event description. The switch can then apply a machine learning technique using the one or more patterns to determine a recovery action for mitigating the event.

Method and system for facilitating a self-healing network

An optical network includes top networking ports coupled to a packet switch, first media converters, second media converters, and bottom networking ports. The first media converters are coupled to top networking ports, each of the first media converters including a first ASIC transceiver that has a circuit switch function. The second media converters are coupled to the first media converter via optical cables to receive the optical signals. Each of the second media converters includes a second ASIC transceiver that has a circuit switch function. The bottom networking ports are coupled to the second media converters. The first ASIC transceiver and the second ASIC transceiver are configured to transmit a signal from one of the top networking ports to any one of the bottom networking ports, and transmit a signal from one of the bottom networking ports to any one of the top networking ports.

Optical network having combined circuit-packet switch architecture

Systems and methods are provided for managing network devices using policy graph representations. In some embodiments, the method includes receiving configurations for a plurality of network devices; extracting one or more policies from the configurations; extracting a label hierarchy from the configurations, the label hierarchy describing an organization of nodes in a network comprising the network devices; generating a connectivity of a network comprising the network devices based on the one or more policies and the label hierarchy; generating a policy graph representation of the connectivity of the network; and displaying the policy graph representation of the connectivity to a user.

Graph-based policy representation system for managing network devices

An event analysis system in a network is provided. During operation, the system can determine a set of monitoring elements in the network. A respective monitoring element can facilitate monitoring of events in the network. The system can then represent a respective monitoring element as a node of a graph. If the corresponding monitoring elements are related based on a relationship definition, the system can connect a node pair of the graph with an edge. The system can activate a set of nodes of the graph corresponding to an issue in the network based on notifications from one or more monitoring agents on one or more switches in the network. Subsequently, the system can determine a root cause of the issue by traversing the set of activated nodes of the graph and determine a recovery action for mitigating the root cause.

Method and system for root cause analysis of network issues

Systems and methods are provided for self-tuning networks using distributed analytics for network devices. In some embodiments, the method includes instantiating, in a network device, an agent for a network protocol, wherein the agent: monitors a performance parameter for a resource of the network protocol, and responsive to a value of the performance parameter exceeding a threshold value, reports an anomaly for the network protocol to a remote management server, receives a new threshold value from the remote management server, the new threshold value being based on the anomaly reported by the network device, and anomalies reported by other network devices according to respective thresholds employed by the network devices for the performance parameter for the resource of the network protocol, and replaces the threshold value of the performance parameter for the resource of the network protocol with the new threshold value received from the remote management server.

Self-tuning networks using distributed analytics

Example method includes: receiving, by a network device, a plurality of input policy graphs and a composed policy graph associated with the input policy graphs; dividing the composed policy graph into a plurality of sub-graphs, each sub-graph comprising a plurality of edges and a plurality of source nodes and destination nodes that the edges are connected to; selecting a first subset of sub-graphs that include, as a source node, a disjoint part of an original source EPG for each input policy graph; identifying a second subset within the first subset of sub-graphs that include, as a destination node, a disjoint part of an original destination EPG for the each input policy graph; and verifying whether connectivity in the composed policy graph reflects a corresponding policy in the plurality of input policy graphs for each sub-graph in the second subset.

Verifying whether connectivity in a composed policy graph reflects a corresponding policy in input policy graphs

In some examples, a method is described. The method can, for example be performed by a network switch and can include receiving instructions to monitor a switch resource, monitoring time-series data for traffic received and transmitted by the switch, identifying peaks for the switch resource in a specified time period for the monitored data, comparing the identified peaks with data provided during a training phase to identify whether the monitored data includes an anomaly, and generating an alert when the data is identified as containing an anomaly.

Growing community of inventors

Santa Clara, CA, United States of America

Anu Mercian