Cisco Technology, Inc.

Praveen Parthasarathy Iyengar

Rajendra Kumar Thirumurthi

Samar Sharma

Srinivas Pitta

Mouli Vytla

Narayanaswami Ganapathy

Shrawan Chittoor Surender

Anand Akella

Krishna Bharadwaj Dharwada

Vivek Purushotham

Anand V Akella

A method of managing security rules may include extracting metadata from a data packet received at a first network device. The metadata including network metadata and network system metadata. The method may further include distributing the metadata to at least one service endpoint registered with the first network device, receiving from the at least one service endpoint, an indication as to how traffic associated with the data packet is to be handled, and enabling the traffic based at least in part on feedback received from the at least one service endpoint and creating a first service flow hash entry of a hash table associated with the data packet at the first network device. The first service flow hash entry identified each of a number of services using a unique number. The method may further include distributing the hash table including the first service flow hash entry across a fabric to at least a second network device.

Dynamic management of security rules and policies

Embodiments herein describe an upgrade system that provides suggestions for upgrades using a confidence metric. In one embodiment, the upgrade system tracks network elements in multiple network fabrics to determine whether an upgrade performed on a first network element was successful. The upgrade system can generate one or more vectors that store various data about the network elements such as enabled features, telemetry data, control plane faults, system settings, locations of the network elements in a fabric, and the like. By evaluating these vectors, the upgrade system can derive a reliability confidence metric regarding the upgrade. If the reliability confidence metric exceeds a threshold, this indicates the upgrade was successful. The upgrade system can then compare vectors for different network elements to determine whether to apply the same upgrade to similar network elements.

Self-healing fabrics

Methods and systems for tracking transactions in a network fabric, include: receiving a message generated by a node in the network fabric; assigning a uniform fabric identifier ('UFID') to the node; assigning a uniform transaction identifier ('UTID') to the message&#x201d;; appending the UFID and the UTID to the message; and storing the message in a database.

Method and apparatus for a software-seamed and augmented view of an asynchronous network fabric

A method is described and in one embodiment includes, for each of a plurality of outgoing ports of a first network element: collecting data comprising a number of packets arriving the outgoing port and an amount of power consumed by the outgoing port for a first time interval; calculating a packet per watt ('P/W') metric for the port for the first time interval, wherein the P/W metric comprises the number of packets coming into the port divided by the amount of power consumed by the port during the first time interval; repeating the collecting and calculating for a number of successive time intervals; calculating a mean P/W metric for a time period comprising the first time interval and the successive time intervals; and calculating a variance for the time period comprising the first time interval and the successive time intervals. The method further includes redirecting traffic received at the network element to the outgoing port having the lowest variance.

Power aware switching using analytics

A method is provided in one example embodiment and includes calculating a packet metric for a first one of a plurality of links connected to a network element, wherein the packet metric is equal to a packet count over a period of time T divided by a maximum throughput of the first one of the links; and determining based on the packet metric whether traffic on the one of the links should be switched to another one of the plurality of links to optimize power usage. If a determination is made that the traffic on the first one of the links should be switched to another one of the plurality of links, the method further includes selecting a second link from a subset of the plurality of links; assigning the traffic to the selected second link; and applying a cost saving measure at an interface of the network element to which the first link is connected.

Growing community of inventors

San Jose, CA, United States of America

Praveen Parthasarathy Iyengar