International Business Machines Corporation

Vmware, Inc.

Signify Holding B.v.

Nicira, Inc.

Philips Lighting Holding B.v.

Baiao Technology (beijing) Co., Ltd.

Beijing Boe Optoelectronics Technology Co., Ltd.

Beijing Baidu Netcom Science and Technology Co., Ltd.

Fudan University

Xiaomi Inc.

Qin Li

Peng Hui Jiang

Guang Han Sui

Jian Hua Zhang

Jing Li

Bin Xu

Feng Gao

Xiang Sean Zhou

Ping Xiao

Niao Qing Liu

Uday Suresh Masurekar

Amit Chopra

Liang Zhou

Hao Huang

Ayyappan Veeraiyan

Shu Wu

Shailesh Urhekar

Jing Bao

Chengjian Wen

Jonathan D Dunne

Jun Zhou

Feng He

Si Bin Fan

Yong Wu

Zhi Jun Wang

Jikai Yao

Min Chen

Peter Hubertus Franciscus Deurenberg

Ping Ping Cheng

Chen Jiang

Xiaobing Liu

Zhong Huang

Qun Luo

Kang Li

Huaizhou Liao

Runqi Zhang

Xiao Guang Song

Zhiping Li

Jinmiao Lu

Guangfei Wang

Yidie Huang

Xiaoxia Li

Wu Gen Zhan

PeiPei Hu

Wil Ma

Huajie Luo

Zhengyi Zheng

Digital TV antenna

This disclosure describes processes for performing direct memory access ('DMA') between memory of a host and memory of a smart network interface controller (“SNIC”) connected to a bus of the host. The host runs a host thread in a processor of the host and the SNIC runs a SNIC thread in a processor of the SNIC. The host thread and the SNIC thread facilitate direct access of the SNIC thread to memory locations in the memory of the host. The SNIC thread can fetch data directly from and/or write data directly to the memory locations of the memory of the host over the bus.

Methods and systems for establishing direct communications between a server computer and a smart network interface controller

This disclosure is directed to processes and systems for generating data packets in a smart network interface controller ('SNIC') of a host server computer. A smart packet generator (“spktgen”) controller receives a user command that contains directions for how packets are generated in the SNIC. The command is sent to a spktgen daemon that runs in a control core of the multicore processor. The spktgen daemon extracts the type of packet generator and packet parameters recorded in the command and sends the type of packet generator and packet parameters to a spktgen engine that also runs in the control core. The spktgen engine creates threads in each of one or more data cores of the multicore processor. Each thread comprises instructions for generating data packets from the data generated by data generating sources of the host in accordance with the type of packet generator and the packet parameters.

Methods and systems for generating data packets in smart network interface controllers

A method for updating an application is provided. In the method, a first interface image generated during a period that a first instance of the application is running is obtained. The first instance of the application is running on an application trial platform. The application trial platform is communicatively couplable to a server of the application. A plurality of attributes associated with a control contained in the first interface image are extracted. Based on the plurality of attributes, it is determined whether the first interface image indicates that an update of the application is released by the server. In response to determining that the first interface image indicates that the update of the application is released by the server, the first instance of the application is updated on the application trial platform.

Method and apparatus for updating application

In an approach, a processor detects at least one training message delivered by a user on a social media application, the training message being accessible by an operator. A processor records training feedback performed by the operator on the social media application regarding to the at least one training message. A processor trains an analysis engine module using (i) the at least one training message as input and (ii) the training feedback as output. A processor generates a predicted feedback for a subsequent message based on the trained analysis engine module.

Feedback prediction in a social media application

Example methods are provided to perform traffic forwarding between geographically dispersed first site and second site and to support traffic forwarding via a trunk interface. In one example, the method may include receiving, by a first edge device at the first site, network traffic having a plurality of packets via a trunk interface of the first edge device from a virtual tunnel endpoint, the virtual tunnel endpoint having decapsulated the packets prior to communicating the packets through the trunk interface. The method may further include reading an overlay network identifier from each of the packets to identify a source overlay network of the received network traffic from the multiple overlay networks; modifying each of the packets to include a virtual local area network (VLAN) identifier; and forwarding modified network traffic to a second edge device at the second site to identify the destination network based on the VLAN identifier.

Traffic forwarding between geographically dispersed sites

An example operation may include one or more of receiving, by a data server node, requests for data from a plurality of user nodes connected to a blockchain, retrieving, by the data server node, the data from the blockchain, dividing, by the data server node, the data into a plurality of portions and generating copies of the plurality of the portions, changing, by the data server node, at least one bit in each of the copies, encrypting, by the data server node, the copies and recording the copies into blocks on a second blockchain, recording, by the data server node, a plurality of sequence numbers of the blocks, and storing, by the data server node, the plurality of the sequence numbers on a second blockchain to be accessed by the plurality of the user nodes.

Protection of data trading

A luminaire has a light source carrier around a central light output window, with light sources facing a first, at least partially upward, direction. An upper reflector cover reflects light to a range of light output directions, at least partially downwardly, through the central window opening. A lens structure is formed over the set of light sources having a light diffusing output area and reflecting side walls extending between the light sources and the light diffusing output area. This design enables the luminaire to be very slim. The design may also avoid the need for a light diffusing output window.

Luminaire

Embodiments of the invention provide methods, systems, computer program products for managing a call waiting in a queue during a request for staff services provided by a call center. According to the method, a first voice segment received in a call made by a device is recorded first. Next, whether a part of the first voice segment is related to a first predefined voice segment is determined. The volume of device is adjusted in response to the part of the first voice segment being related to the first predefined voice segment. A user of the device is alerted in response to the part of the first voice segment being not related to the first predefined voice segment.

Optimization for a call that waits in queue

A file is transmitted to a plurality of hosts. The hosts are divided into host groups. A host in each host group is selected as a master host. A data block is received by a master host in a host group. The master host sends the data block to both a master host in a next host group and a next host in the host group in order that the next host in the host group sends the data block to a next host of the next host in the host group, and the master host in the next host group sends the data block to both a master host in a next host group of the next host group and a next host of the next host in the next host group.

File transmission in a cluster

A luminaire () is disclosed comprising at least one optical chamber () delimited by a light exit window () and a curved reflective surface () extending from the light exit window to a mounting surface () extending from the curved reflective surface towards the light exit window. Each optical chamber comprises a light source () mounted on the mounting surface; and a light guide () extending across part of the optical chamber. The light guide comprises a total internal reflection major surface () facing the light exit window; a further major surface () facing the curved reflective surface, said further major surface comprising a light outcoupling portion () comprising a plurality of light outcoupling structures (); a first edge surface () optically coupled to the light source; and a further edge surface () opposite the first edge surface arranged to direct light towards the curved reflective surface.

Indirect luminaire

A method for distributed file transfers with high performance and reliability includes creating, on a first Trivial File Transfer Protocol (TFTP) server, a global cache, where the global cache is used to store up to a first portion of a data file. The method further includes storing, on the first TFTP server, in the global cache, one or more continuous data blocks that have exceeded a defined first request rate threshold, where the one or more continuous data blocks make up a subset of the data blocks of the first portion of the data file. The method further includes predicting, on the first TFTP server, a next data block in the data file to be stored in the global cache and in response to predicting the next data block, storing, on the first TFTP server, the next data block in the global cache.

Distributed file transfer with high performance

Growing community of inventors

Beijing, China

Qin Li