Cisco Technology, Inc.

Applied Materials, Inc.

International Business Machines Corporation

Baidu USA LLC

Netapp, Inc.

Jiangsu Tenesun Electrical Appliance Co., Ltd.

Yi Yang

Alvaro E Retana

Russell Ivan White

James L Ng

Karthik Janakiraman

Abhay Kumar Roy

Donnie Van Savage

Madhavi Wunnava Chandra

David Anthony Cook

Rui Cheng

Alfred C Lindem, Iii

Wojciech Dec

Sina Mirtorabi

Abhijit Basu Mallick

Jiang Wang

Syed Khalid Raza

Krishna Nittala

Steven Edward Moore

Wei Xu

Hai Shan Wu

Rithin Kumar Shetty

Diwakar N Kedlaya

Di Xu

Yao Dong Qi

Aykut Aydin

Kyle Andrew Donald Mestery

Lizhong Sun

Timothy J Kuik

Lan Jun Wang

Carlos M Pignataro

Nagendra Kumar Nainar

Tuan Anh Nguyen

Sanjeev Baluja

Keyur Patel

Akhil Kaushik

Stewart Frederick Bryant

Ian Michael Charles Shand

Jian J Chen

Yuedong Mu

Yihong Chen

Joseph F AuBuchon

Zubin Huang

Bo Qi

David A Cook

Timothy M Gage

Thuan Van Tran

Sanjay Agrawal

Ruchir Gupta

Syed Basheeruddin Ahmed

Ying Wang

Zhiheng Huang

Yujun Wang

Tianshu Wang

Andrew Eric Dunn

Junhua Mao

Donald Earl Slice, Jr

Meenakshi R Kaushik

Liem Hong Nguyen

Xiaodong Yang

Yunyun Wu

Lanjun Wang

Jun Liu

Venkatanarayana Shankaramurthy

Anton Baryshnikov

Haonan Yu

Deepak Kumar

Truong Van Nguyen

Flavio Fernandes

Nattaworn Boss Nunta

Donald Slice

Yinan Yu

Chong Ma

Yafeng Deng

Junhong Li

Xiaowang Wu

Hank R Gibson

Zhonghai Ji

Yufei Zhou

Lichao Huang

Fabien Degouet

Xiaojie Ma

Kun Li

Xiaohu Li

Madhavi W Chandra

Alfred Iii Lindem

Roy Abhay

Systems and methods provide for scaling service discovery in a micro-service environment. A controller can inject a service discovery agent onto a host. At least one of the controller or the agent can identify a first set of micro-service containers that are dependencies of the first micro-service container and a second set of micro-service containers that are dependencies of the second micro-service container. At least one of the controller or the agent can update routing data for the first set of micro-service containers and the second set of micro-service containers. At least one of the controller or the agent can determine the second micro-service container has terminated on the host computing device. At least one of the controller or the agent can update the agent to remove the routing data for the second set of micro-service containers.

Scaling service discovery in a micro-service environment

A chamber includes a target () and a magnetron () disposed over the target (). The magnetron () includes a plurality of magnets (). The magnetron () has a longitudinal dimension and a lateral dimension. The longitudinal dimension of the magnetron () is tilted with respect to the target () so the distances between magnets () and the target () vary. As the magnetron () rotates during operation, the strength of the magnetic field produced by the magnetron () is an average of the various strengths of magnetic fields produced by the magnets (). The averaging of the strengths of the magnetic fields leads to uniform film properties and uniform target erosion.

Tilted magnetron in a PVD sputtering deposition chamber

Exemplary deposition methods may include delivering a silicon-containing precursor and a boron-containing precursor to a processing region of a semiconductor processing chamber. The methods may include delivering a dopant-containing precursor with the silicon-containing precursor and the boron-containing precursor. The dopant-containing precursor may include one or more of carbon, nitrogen, oxygen, or sulfur. The methods may include forming a plasma of all precursors within the processing region of the semiconductor processing chamber. The methods may include depositing a silicon-and-boron material on a substrate disposed within the processing region of the semiconductor processing chamber. The silicon-and-boron material may include greater than or about 1 at. % of a dopant from the dopant-containing precursor.

Doping semiconductor films

Exemplary deposition methods may include delivering a silicon-containing precursor and a boron-containing precursor to a processing region of a semiconductor processing chamber. The methods may include providing a hydrogen-containing precursor with the silicon-containing precursor and the boron-containing precursor. A flow rate ratio of the hydrogen-containing precursor to either of the silicon-containing precursor or the boron-containing precursor is greater than or about 2:1. The methods may include forming a plasma of all precursors within the processing region of a semiconductor processing chamber. The methods may include depositing a silicon-and-boron material on a substrate disposed within the processing region of the semiconductor processing chamber.

Methods to reduce material surface roughness

Presented herein are embodiments of a multimodal Recurrent Neural Network (m-RNN) model for generating novel image captions. In embodiments, it directly models the probability distribution of generating a word given a previous word or words and an image, and image captions are generated according to this distribution. In embodiments, the model comprises two sub-networks: a deep recurrent neural network for sentences and a deep convolutional network for images. In embodiments, these two sub-networks interact with each other in a multimodal layer to form the whole m-RNN model. The effectiveness of an embodiment of model was validated on four benchmark datasets, and it outperformed the state-of-the-art methods. In embodiments, the m-RNN model may also be applied to retrieval tasks for retrieving images or captions.

Intelligent image captioning

Methods and systems for controlling concentration profiles of deposited films using machine learning are provided. Data associated with a target concentration profile for a film to be deposited on a surface of a substrate during a deposition process for the substrate is provided as input to a trained machine learning model. One or more outputs of the trained machine learning model are obtained. Process recipe data identifying one or more sets of deposition process settings is determined from the one or more outputs. For each set of deposition process setting, an indication of a level of confidence that a respective set of deposition process settings corresponds to the target concentration profile for the film to be deposited on the substrate is also determined. In response to an identification of the respective set of deposition process settings with a level of confidence that satisfies a level of confidence criterion, one or more operations of the deposition process are performed in accordance with the respective set of deposition process settings.

Controlling concentration profiles for deposited films using machine learning

Embodiments described herein generally relate to doping of three dimensional (3D) structures on a substrate. In some embodiments, a conformal dopant containing film may be deposited over the 3D structures. Suitable dopants that may be incorporated in the film include halogen atoms. The film may be subsequently annealed to diffuse the dopants into the 3D structures.

Conformal halogen doping in 3D structures using conformal dopant film deposition

Methods and apparatus are used for adjusting film stress profiles on substrates. An apparatus may include a PVD chamber with a pedestal configured to support a substrate during processing on a cover positioned on an uppermost surface of the pedestal. The cover is constructed with multiple electrodes such as, for example, a first electrode, a second electrode, and a third electrode. The second electrode is positioned between and electrically separated from the first electrode and the second electrode. A substrate stress profile tuner is electrically connected to the first electrode, the second electrode, and the third electrode and configured to independently adjust an RF voltage level of at least the second electrode and the third electrode relative to RF ground to produce a more uniform film stress profile.

Methods and apparatus for zone control of RF bias for stress uniformity

Systems and methods of using pulsed RF plasma to form amorphous and microcrystalline films are discussed herein. Methods of forming films can include (a) forming a plasma in a process chamber from a film precursor and (b) pulsing an RF power source to cause a duty cycle on time (T) of a duty cycle of a pulse generated by the RF power source to be less than about 20% of a total cycle time (T) of the duty cycle to form the film. The methods can further include (c) depositing a first film interlayer on a substrate in the process chamber; (d) subsequent to (c), purging the process chamber; and (e) subsequent to (d), introducing a hydrogen plasma to the process chamber. Further in the method, (b)-(e) are repeated to form a film. The film can have an in-film hydrogen content of less than about 10%.

Film formation via pulsed RF plasma

One or more techniques and/or computing devices are provided for non-disruptively establishing a synchronous replication relationship between a primary volume and a secondary volume and/or for resynchronizing the primary volume and the secondary volume. For example, a baseline snapshot and one or more incremental snapshots of the primary volume are used to construct and incrementally update the secondary volume with data from the primary volume. A dirty region log is used to track modifications to the primary volume. A splitter object is used to split client write requests to the primary volume and to the secondary volume. A synchronous transfer engine session is initiated to processing incoming client write requests using the dirty region log. A cutover scanner is used to transfer dirty data from the primary volume to the secondary volume. In this way, a synchronous replication relationship is established between the primary volume and the secondary volume.

Non-disruptive baseline and resynchronization of a synchronous replication relationship

A control method and control device for a heat pump-type double-circulation hot-air drying system, relating to a device for supplying or controlling air or gas for drying solid materials or products, and in particular to a control method and control device for a heat pump-type hot-air drying system. The method comprises: configuring a temperature control parameter, and saving a preset temperature control curve parameter; detecting and monitoring an outlet air temperature, and the temperature and humidity of a drying room; dynamically adjusting a set temperature according to a preset temperature control curve; and selecting, according to a current set temperature, a double-circulation dynamic operation mode of a system. The control device uses a micro-processor to realize program control. By building an inner circulation loop for large-volume air circulation, the latent heat of condensation in a refrigerant is fully absorbed, to improve the basic air temperature.

Control method and control device for variable-frequency and variable-capacity heat pump hot-air drying system

Methods of conformally doping three dimensional structures are discussed. Some embodiments utilize conformal silicon films deposited on the structures. The silicon films are doped after deposition to comprise halogen atoms. The structures are then annealed to dope the structures with halogen atoms from the doped silicon films.

Methods for conformal doping of three dimensional structures

A control system and method for a frostless, multivariable coupling and heat pump-based hot blast stove are used for grain drying. A first heat exchanger, a second heat exchanger, a main solution pool, corresponding pipelines, and a temperature detector are configured in the control system. A first heat pump unit, a second heat pump unit and a third heat pump unit are formed. A preheating zone, a low temperature zone, a medium temperature zone and a high temperature zone are sequentially formed on an air supply pipeline from a fresh air inlet to a fresh air outlet. A frostless operation procedure is provided. Through the configuration, the control system and method for a frostless, multivariable coupling and heat pump-based hot blast stove can implement heat supply in a gradient heat-circulation preheating mode and a gradient heat-circulation frostless mode.

Control system and control method for frostless, multivariable coupling, and heat pump-based hot blast stove

Growing community of inventors

Morrisville, NC, United States of America

Yi Yang