公开(公告)号：US10884774B2

公开(公告)日：2021-01-05

申请号：US15311914

申请日：2014-06-16

Applicant: Hewlett Packard Enterprise Development LP

Inventor： Jun Li ,  Hernan Laffitte ,  Donald E. Bollinger ,  Eric Wu

IPC: G06F9/455 ,  G06F16/188

Abstract: Examples relate to deploying distributed applications using virtual nodes. In some examples, virtual nodes are created and are each assigned a core subset of a number of processing cores, an Internet protocol (IP) address, and an in-memory file system configured to provide access to a portion of physically shared memory. At this stage, a distributed application that is configured to be deployed to a plurality of machine nodes is deployed to the plurality of virtual nodes. On a first virtual node, a reference to a first dataset stored in physically shared memory is sent to a second virtual node, where the physically shared memory is accessible to each of the plurality of virtual nodes. Next, on the second virtual node, the first dataset is accessed through the in-memory file system of the first virtual node.

公开(公告)号：US10817496B2

公开(公告)日：2020-10-27

申请号：US15033577

申请日：2013-11-05

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： Vinay Deolalikar ,  Hernan Laffitte

IPC: G06F7/00 ,  G06F16/23 ,  G06N20/00 ,  G06F16/28 ,  G06Q50/00 ,  G06Q10/10 ,  H04L12/58

Abstract: In one implementation, a poster engine may maintain a thread history associated with a response poster, a status engine may maintain a potential thread status of a thread of a forum based on the thread history, and a pool engine may maintain an inspection pool based on the potential thread status. In another implementation, a thread status of a thread of the forum may be maintained based on a correlation rating of a response poster and a pool having a number of threads based on the thread status may be provided.

公开(公告)号：US10089761B2

公开(公告)日：2018-10-02

申请号：US15143273

申请日：2016-04-29

Applicant: Hewlett Packard Enterprise Development LP

Inventor： Krishnamurthy Viswanathan ,  Maria Gonzalez ,  Fei Chen ,  Hernan Laffitte

IPC: G06T11/20 ,  G06T1/60 ,  G06T1/20 ,  G06F17/30

Abstract: Examples herein involve graph processing using a shared memory. An example method includes distributing vertices of a graph to a plurality of graph partition processors of a system, the graph partition processors to process non-overlapping sets of vertices of the graph; storing a master copy of a vertex state of the graph in a shared memory of the system; instructing the graph partition processors to iteratively process respective vertices of the graph partitions based on local copies of the state of the graph stored in respective local memories of the graph partition processors; and updating the master copy of the state of the graph in the shared memory based on the iterative processing of the partitions of the vertices by the graph partition processors until convergence of the graph.

公开(公告)号：US20170316591A1

公开(公告)日：2017-11-02

申请号：US15143273

申请日：2016-04-29

Applicant: Hewlett Packard Enterprise Development LP

Inventor： Krishnamurthy Viswanathan ,  Maria Gonzalez ,  Fei Chen ,  Fei Chen ,  Hernan Laffitte

IPC: G06T11/20 ,  G06T1/60 ,  G06T1/20

CPC classification number: G06T11/206 ,  G06F17/30958 ,  G06T1/20 ,  G06T1/60

Abstract: Examples herein involve graph processing using a shared memory. An example method includes distributing vertices of a graph to a plurality of graph partition processors of a system, the graph partition processors to process non-overlapping sets of vertices of the graph; storing a master copy of a vertex state of the graph in a shared memory of the system; instructing the graph partition processors to iteratively process respective vertices of the graph partitions based on local copies of the state of the graph stored in respective local memories of the graph partition processors; and updating the master copy of the state of the graph in the shared memory based on the iterative processing of the partitions of the vertices by the graph partition processors until convergence of the graph.

公开(公告)号：US20170090973A1

公开(公告)日：2017-03-30

申请号：US15311914

申请日：2014-06-16

Applicant: HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor： Jun Li ,  Hernan Laffitte ,  Donald E. Bollinger ,  Eric Wu

IPC: G06F9/455 ,  G06F17/30

CPC classification number: G06F9/45558 ,  G06F16/188 ,  G06F2009/45562 ,  G06F2009/45583

Abstract: Examples relate to deploying distributed applications using virtual nodes. In some examples, virtual nodes are created and are each assigned a core subset of a number of processing cores, an Internet protocol (IP) address, and an in-memory file system configured to provide access to a portion of physically shared memory. At this stage, a distributed application that is configured to be deployed to a plurality of machine nodes is deployed to the plurality of virtual nodes. On a first virtual node, a reference to a first dataset stored in physically shared memory is sent to a second virtual node, where the physically shared memory is accessible to each of the plurality of virtual nodes. Next, on the second virtual node, the first dataset is accessed through the in-memory file system of the first virtual node.