Cisco Technology, Inc.

Big Switch Networks, Inc.

Subrata Banerjee

Debojyoti Dutta

Ashok Ganesan

Sukhdev S Kapur

Thomas James Edsall

Dinesh Ganapathy Dutt

Ethan M Spiegel

Raja Rao Tadimeti

Ankur Jain

Madhava Rao Cheethirala

Arpan K Ghosh

Sumeet Singh

Silvano Gai

Davide Bergamasco

Joshua Merrill

Badrinarayanan Ramaswamy

Bruno Raimondo

Shankar Ramanathan

Ranganathan Rajagopalan

Srinivasan Ramasubramanian

Robert Edward Adams

Robert Wesley Sherwood

Sriharsha Jayanarayana

Kanzhe Jiang

Huida Dai

Badri Ramaswamy

Sumit Rangwala

Vinay Gaonkar

Rajeev Bhardwaj

Nagasree Ravindra

Pralhad Katti

Venkataraman Swaminathan

Yuanbo Zhu

Arpan Rongong

Jiang Zhu

Jan Medved

Shujin Zhang

Murali Basavaiah

Anand Girish Parthasarathy

Kaushik Narayan

Harsha Bharadwaj

Sanjay Agrawal

Senhua Huang

Joy J Chatterjee

Parthiban Munusamy

Danny J Elder

Arpakorn Boonkongchuen

Robert Horst Starmer

Michael V Dvorkin

Rituparna Agrawal

Erik David Ludvigson

Pavan Basetty

Ruben Hakopian

Rajeev Bharadwaj

Sudheer Yelanduru Raghavendra

Rajendra Shinde

Disclosed are systems, methods and computer-readable storage medium for extending a private cloud to a public cloud. The private cloud can be extended to the public cloud by establishing a virtual private network between a private cloud and a public cloud, receiving one or more access control lists provisioned by the private cloud, determining contracts between an end point group of the private cloud and an end point group of the public cloud based on the one or more access control lists, and extending the end point group of the private cloud to the end point group of the public cloud across the virtual private network.

Extension of a private cloud end-point group to a public cloud

A method for summarizing capabilities in a hierarchically arranged data center includes receiving capabilities information, wherein the capabilities information is representative of capabilities of respective nodes at a first hierarchical level in the hierarchically arranged data center, clustering nodes based on groups of capabilities information, generating a histogram that represents individual node clusters, and sending the histogram to a next higher level in the hierarchically arranged data center. Relative rankings of capabilities may be used to order a sequence of clustering operations.

Data center capability summarization

A controller may be used to control client switches in a network that includes non-client, switches. The controller may form client domains from groups of client switches that are separated by intervening non-client domains formed from non-client switches. The controller may determine a network domain topology from the client domains and non-client domains. The controller may determine a spanning tree that interconnects the nodes of the network domain topology. The controller may control client switches of the client domains to allow only network traffic between the client domains and the non-client domains along the spanning tree. The controller may use the network domain topology to generate inter-domain forwarding maps. The inter-domain forwarding maps may be used to determine network forwarding paths between end hosts in the network.

Systems and methods for forwarding network packets in a network using network domain topology information

A first network device determines capabilities of resources in a section of a network that is accessible using the first network device. The first network device groups the resources into a resource cluster. The first network device advertises the resource cluster in the network, wherein each of a plurality of network devices advertise a resource cluster associated with sections of the network. A second network device receives a request for providing a service. The second network device groups the request into a plurality of request clusters. The second network device selects at least one resource cluster for providing the service based on information associated with the request clusters and the advertised resource clusters. The second network device allocates resources included in the at least one resource cluster for providing the service based on selecting the at least one resource cluster.

Resource allocation mechanism

Techniques are provided for mitigating the effects of slow or no drain devices on a fabric. One or more of the described embodiments can be used alone or in combination to address problems associated with inter-switch link blocking and to address the situation where flows which are not associated with slow/no drain devices suffer the negative impacts of slow or no drain devices on a fabric.

Mitigating the effects of congested interfaces on a fabric

A controller may be used to control client switches in a network that includes non-client switches. The controller may form client domains from groups of client switches that are separated by intervening non-client domains formed from non-client switches. The controller may determine a network domain topology from the client domains and non-client domains. The controller may determine a spanning tree that interconnects the nodes of the network domain topology. The controller may control client switches of the client domains to allow only network traffic between the client domains and the non-client domains along the spanning tree. The controller may use the network domain topology to generate inter-domain forwarding maps. The inter-domain forwarding maps may be used to determine network forwarding paths between end hosts in the network.

In one embodiment a method includes at first a network device, identifying an attribute of the first network device, selecting a function that defines how the attribute is to be summarized together with a same attribute of a second network device, generating a message that comprises a set of information comprising an identification of the attribute and the function, and sending the message to, e.g., a next higher node in a network hierarchy of which the network device is a part. Multiple such messages may be received at an aggregation node that then summarizes the attributes according to respective functions, and sends or publishes a summarized version of the attributes to a still next higher node in the network hierarchy.

Extensible attribute summarization

Techniques are provided herein for a device in a network to receive information configured to indicate a control plane traffic load level for one or more server devices that are configured to manage traffic for messaging and presence clients communicating via a messaging and presence protocol. The control plane traffic is associated with the messaging and presence protocol. A determination is made as to when the control plane traffic load level has become unbalanced among the two or more server devices and in response to determining that the control plane traffic load level has become unbalanced, sending a transfer message to one or more clients comprising information configured to initiate migration of one or more clients from a server device that is relatively overloaded to a server device that is relatively underloaded in order to balance the control plane traffic load level among the two or more server devices.

Active load distribution for control plane traffic using a messaging and presence protocol

Techniques are provided for sending from a client in a first network device a first session-initiate message to a second network device that is configured to provide network layer, data link layer, or associated convergence layer based service connection information in order for the second network device to accept or reject a network layer, data link layer, or associated convergence layer based service connection with the first network device. The first session-initiate message is based on a messaging and presence protocol. A session-accept message is received at the client in the first network device that is configured to accept the service connection and provide a network layer, data link layer, or associated convergence layer based service connection information in order for the first network device to establish the service connection with the second network device. The session-accept message is based on the messaging and presence protocol. In response to receiving the session-accept message, the service connection is established.

Resource negotiation for cloud services using a messaging and presence protocol

In one embodiment, an n-dimensional resource vector for each of a plurality of resources in a computer network is determined, each n-dimensional resource vector having n property values for a corresponding resource of the plurality of resources. Upon receiving a request for one or more resources of the plurality of resources, where the request indicates one or more desired property values, the techniques convert the desired property values of the request into an n-dimensional request vector, determine a distance between each resource vector and the request vector, and provide a response to the request, the response indicating one or more closest match resources for the request based on the distances.

Generalized coordinate system and metric-based resource selection framework

Data representing capabilities of devices in a data is aggregated on a cluster-basis. Information representing capability attributes of devices in the data center is received. The information representing the capability attributes is analyzed to generate data that groups devices based on similarity of at least one capability attribute. Aggregation data is stored that represents the grouping of the devices based on similarity of the at least one capability attribute and identifies the devices in corresponding groups.

Clustering-based resource aggregation within a data center

Techniques are provided herein for defragmenting resources within a cloud computing system. The cloud computing system includes a plurality of servers deployed in a plurality of respective racks, wherein the respective racks are deployed in a pod of a data center. An element of the cloud computing system determines for each server in a given rack of servers a number of free resource slots available thereon and a number of resource slots in an idle state, and then further determines whether the number of free resource slots on a first server in the plurality of servers is greater than a predetermined threshold. When the number of free resource slots in the first server is greater than the predetermined threshold, a second server in the plurality of servers is identified with sufficient resource slots thereon to accommodate the number of resource slots in the idle state on the first server, and the resource slots in the idle state on the first server are caused to be migrated to the second server.

Hierarchical defragmentation of resources in data centers

In one embodiment, a method comprises receiving a request for a distributed service, the distributed service offered by a service provider via a data communications network having service delivery locations reachable via a prescribed physical topology; identifying the service delivery locations within a prescribed logical topology overlying the prescribed physical topology, the prescribed logical topology segregating the distributed service from other network traffic on the prescribed physical topology; and identifying one or more of the service delivery locations optimized for providing the distributed service to at least one service consumption location in the prescribed logical topology according to a prescribed service level agreement with the service provider.

Placement of service delivery locations of a distributed computing service based on logical topology

In one embodiment, a method comprises retrieving a request graph specifying request nodes identifying respective requested cloud computing service operations, and at least one request edge specifying a requested path requirements connecting the request nodes; identifying a placement pivot among feasible cloud elements identified in a physical graph representing a data network having a physical topology, each feasible cloud element an available solution for one of the request nodes, the placement pivot having a maximum depth in the physical topology relative to the feasible cloud elements; ordering the feasible cloud elements, according to increasing distance from the placement pivot to form an ordered list of candidate sets of feasible cloud elements; and determining an optimum candidate set, from at least a portion of the ordered list, based on the optimum candidate set having an optimized fitness function in the physical graph among the other candidate sets in the ordered list.

Cloud resource placement using placement pivot in physical topology

Growing community of inventors

Los Altos, CA, United States of America

Subrata Banerjee