公开(公告)号：US20160337465A1

公开(公告)日：2016-11-17

申请号：US15154141

申请日：2016-05-13

Applicant: CISCO TECHNOLOGY, INC.

Inventor： Marc Solanas Tarre ,  Ralf Rantzau ,  Debojyoti Dutta ,  Manoj Sharma

IPC: H04L29/08 ,  H04L12/861

Abstract: Approaches are disclosed for distributing messages across multiple data centers where the data centers do not store messages using a same message queue protocol. In some embodiment, a network element translates messages from a message queue protocol (e.g., Kestrel, RABBITMQ, APACHE Kafka, and ACTIVEMQ) to an application layer messaging protocol (e.g., XMPP, MQTT, WebSocket protocol, or other application layer messaging protocols). In other embodiments, a network element translates messages from an application layer messaging protocol to a message queue protocol. Using the new approaches disclosed herein, data centers communicate using, at least in part, application layer messaging protocols to disconnect the message queue protocols used by the data centers and enable sharing messages between messages queues in the data centers. Consequently, the data centers can share messages regardless of whether the underlying message queue protocols used by the data centers (and the network devices therein) are compatible with one another.

Abstract translation: 公开了用于在多个数据中心分发消息的方法，其中数据中心不使用相同的消息队列协议来存储消息。 在一些实施例中，网络元件将来自消息队列协议（例如，Kestrel，RABBITMQ，APACHE Kafka和ACTIVEMQ）的消息转换为应用层消息传递协议（例如，XMPP，MQTT，WebSocket协议或其他应用层消息传递协议） 。 在其他实施例中，网络元件将消息从应用层消息传递协议转换为消息队列协议。 使用本文公开的新方法，数据中心至少部分地使用应用层消息传递协议进行通信，以断开由数据中心使用的消息队列协议，并使得能够在数据中心中的消息队列之间共享消息。 因此，数据中心可以共享消息，而不管数据中心（及其中的网络设备）使用的底层消息队列协议是否彼此兼容。

公开(公告)号：US09454702B2

公开(公告)日：2016-09-27

申请号：US13968235

申请日：2013-08-15

Applicant: Cisco Technology, Inc.

Inventor： Mainak Sen ,  Debojyoti Dutta ,  Shaun Kirby

IPC: G06K9/00 ,  G02B5/10 ,  G06Q10/08 ,  G06K9/20 ,  G08B13/196

CPC classification number: G06K9/00771 ,  G02B5/10 ,  G06K9/209 ,  G06Q10/087 ,  G08B13/19626

Abstract: In one embodiment, a processor can receive data representing a view reflected by a mirror of a plurality of mirrors. The plurality of mirrors may be configured in a space to reflect a plurality of views of structures in the space. The mirror of the plurality of mirrors may include a uniquely identifiable feature distinguishable from other objects in the space. The processor can identify the mirror of the plurality of mirrors according to the uniquely identifiable feature. The processor can also determine an attribute of the structures according to the identified mirror and the data representing the view reflected by the mirror.

Abstract translation: 在一个实施例中，处理器可以接收表示由多个反射镜的反射镜反射的视图的数据。 多个反射镜可以被配置在空间中以反映空间中的结构的多个视图。 多个反射镜的镜子可以包括与该空间中的其它物体可区分的唯一可识别的特征。 处理器可以根据唯一可识别的特征来识别多个反射镜的镜像。 处理器还可以根据所识别的镜像确定结构的属性，并且还可以确定表示由镜反射的视图的数据。

公开(公告)号：US20160142336A1

公开(公告)日：2016-05-19

申请号：US14541524

申请日：2014-11-14

Applicant: CISCO TECHNOLOGY, INC.

Inventor： Yathiraj B. Udupi ,  Debojyoti Dutta ,  Lewis W. Tucker

IPC: H04L12/923 ,  H04L12/743 ,  H04L29/08

CPC classification number: H04L45/7453 ,  G06F9/5077 ,  H04L41/0813 ,  H04L41/0853 ,  H04L47/762 ,  H04L67/10 ,  H04L67/32 ,  H04L67/327 ,  H04L69/24

Abstract: The present disclosure describes a distributed, advertisement-based, solution for scheduling virtual resources in cloud infrastructures such as the OpenStack. The scheduling algorithm distributes the scheduling requirements and host state feasibility checks to the individual hosts in the datacenter, which can periodically send a summarized advertisement to the scheduler controller listing the number of instances of different type(s) of virtual resources that a particular host can support. The scheduler controller, thus no longer has to compute and maintain individual host states, and the scheduling problem is reduced to selecting the feasible advertisements that satisfy a given request. The solution can be extended to a scenario of multiple scheduler controllers using the same distributed, advertisement-based, approach.

公开(公告)号：US20150278343A1

公开(公告)日：2015-10-01

申请号：US14678244

申请日：2015-04-03

Applicant: Cisco Technology, Inc.

Inventor： Sumit Rangwala ,  Debojyoti Dutta ,  Raja Rao Tadimeti ,  Subrata Banerjee ,  Yuanbo Zhu

IPC: G06F17/30 ,  H04L29/08

CPC classification number: G06F17/30598 ,  G06F17/30489 ,  G06F17/30876 ,  G06F2209/503 ,  H04L67/10

Abstract: 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.

Abstract translation: 用于总结分层布置的数据中心中的能力的方法包括接收能力信息，其中能力信息代表层次排列的数据中心中的第一层级的相应节点的能力，基于能力信息组的聚类节点，生成 直方图表示各个节点簇，并将直方图发送到分层排列的数据中心中的下一个较高级别。 功能的相对排名可用于对一系列聚类操作进行排序。

公开(公告)号：US11847500B2

公开(公告)日：2023-12-19

申请号：US16710499

申请日：2019-12-11

Applicant: Cisco Technology, Inc.

Inventor： Johnu George ,  Sourav Chakraborty ,  Amit Kumar Saha ,  Debojyoti Dutta ,  Xinyuan Huang ,  Adhita Selvaraj

IPC: G06F9/50 ,  G06N20/00 ,  G06F9/48 ,  G06F9/38 ,  G06F18/23213

CPC classification number: G06F9/5044 ,  G06F9/3836 ,  G06F9/48 ,  G06F9/4806 ,  G06F9/4843 ,  G06F9/4881 ,  G06F9/50 ,  G06F9/5005 ,  G06F9/5011 ,  G06F9/5022 ,  G06F9/5027 ,  G06F9/5038 ,  G06F9/5055 ,  G06F9/5061 ,  G06F9/5077 ,  G06F18/23213 ,  G06N20/00

Abstract: A method can include receiving, at a workflow controller, a machine learning workflow, the machine learning workflow associated with a first task and a second task. The first task is training a machine learning model and the second task is deploying the model. The method can include segmenting, by the workflow controller, the machine learning workflow into a first sub-workflow associated with the first task and a second sub-workflow associated with the second task, assigning a first workflow agent to the first sub-workflow and assigning a second workflow agent to the second sub-workflow, selecting, by the first workflow agent and based on first resources needed to perform the first task, a first cluster for performing the first task and selecting, by the second workflow agent and based on second resources needed to perform the second task, a second cluster for performing the second task.

公开(公告)号：US11630990B2

公开(公告)日：2023-04-18

申请号：US16358554

申请日：2019-03-19

Applicant: Cisco Technology, Inc.

Inventor： Abhishek Singh ,  Debojyoti Dutta

IPC: G06N3/04 ,  G06F17/18 ,  G06K9/62 ,  G06N3/082

Abstract: The present disclosure provides systems, methods and computer-readable media for optimizing the neural architecture search for the automated machine learning process. In one aspect, neural architecture search method including selecting a neural architecture for training as part of an automated machine learning process; collecting statistical parameters on individual nodes of the neural architecture during the training; determining, based on the statistical parameters, active nodes of the neural architecture to form a candidate neural architecture; and validating the candidate neural architecture to produce a trained neural architecture to be used in implemented an application or a service.

公开(公告)号：US11595474B2

公开(公告)日：2023-02-28

申请号：US15856716

申请日：2017-12-28

Applicant: Cisco Technology, Inc.

Inventor： Debojyoti Dutta ,  Amit Kumar Saha ,  Johnu George ,  Ramdoot Kumar Pydipaty ,  Marc Solanas Tarre

IPC: G06F7/00 ,  H04L67/1095 ,  G06F11/14 ,  G06F3/06 ,  G06F16/22

Abstract: A method for accelerating data operations across a plurality of nodes of one or more clusters of a distributed computing environment. Rack awareness information characterizing the plurality of nodes is retrieved and a non-volatile memory (NVM) capability of each node is determined. A write operation is received at a management node of the plurality of nodes and one or more of the rack awareness information and the NVM capability of the plurality of nodes are analyzed to select one or more nodes to receive at least a portion of the write operation, wherein at least one of the selected nodes has an NVM capability. A multicast group for the write operation is then generated wherein the selected nodes are subscribers of the multicast group, and the multicast group is used to perform hardware accelerated read or write operations at one or more of the selected nodes.

公开(公告)号：US11120343B2

公开(公告)日：2021-09-14

申请号：US15152379

申请日：2016-05-11

Applicant: CISCO TECHNOLOGY, INC.

Inventor： Aparupa Das Gupta ,  Rahul Ramakrishna ,  Yathiraj B. Udupi ,  Debojyoti Dutta ,  Manoj Sharma

IPC: G06N5/04 ,  G06N5/02 ,  G06F11/30

Abstract: A method for ranking detected anomalies is disclosed. The method includes generating a graph based on a plurality of rules, wherein the graph comprises nodes representing metrics identified in the rules, edges connecting nodes where metrics associated with connected nodes are identified in a given rule, and edge weights of the edges each representing a severity level assigned to the given rule. The method further includes ranking nodes of the graph based on the edge weights. The method further includes ranking detected anomalies based on the ranking of the nodes corresponding to the metrics associated with the detected anomalies.

公开(公告)号：US10938581B2

公开(公告)日：2021-03-02

申请号：US15907018

申请日：2018-02-27

Applicant: Cisco Technology, Inc.

Inventor： Johnu George ,  Amit Kumar Saha ,  Arun Saha ,  Debojyoti Dutta

IPC: G06F15/167 ,  H04L12/06 ,  G06F3/06 ,  G06F16/24 ,  G06F12/08

Abstract: Aspects of the disclosed technology relate to ways to determine the optimal storage of data structures across different memory device is associated with physically disparate network nodes. In some aspects, a process of the technology can include steps for receiving a first retrieval request for a first object, searching a local PMEM device for the first object based on the first retrieval request, in response to a failure to find the first object on the local PMEM device, transmitting a second retrieval request to a remote node, wherein the second retrieval request is configured to cause the remote node to retrieve the first object from a remote PMEM device. Systems and machine-readable media are also provided.

公开(公告)号：US10922287B2

公开(公告)日：2021-02-16

申请号：US15811318

申请日：2017-11-13

Applicant: Cisco Technology, Inc.

Inventor： Johnu George ,  Amit Kumar Saha ,  Arun Saha ,  Debojyoti Dutta

IPC: G06F16/00 ,  G06F16/21 ,  G06F3/06 ,  G06F16/24 ,  G06F12/08 ,  H04L12/06

Abstract: Aspects of the subject technology relate to ways to determine the optimal storage of data structures in a hierarchy of memory types. In some aspects, a process of the technology can include steps for determining a latency cost for each of a plurality of fields in an object, identifying at least one field having a latency cost that exceeds a predetermined threshold, and determining whether to store the at least one field to a first memory device or a second memory device based on the latency cost. Systems and machine-readable media are also provided.