International Business Machines Corporation

Beijing Xiaomi Mobile Software Co., Ltd.

Zhang Li

Peng Hui Jiang

Chunli Wu

Hong Qing Zhou

Jun Su

Xiao Ling Chen

Xiao Lin Sun

Peter Skov

Yan Lin Ren

Su Ying Rui

Shou Hui Wang

Zhi Yu Yue

Jian Xu

Qi Feng Huo

Xian Wei Zhang

Chang Rui Ren

Jinfeng Li

Xiao Yi Tian

Xiang Dong Hu

Guoxiang Zhang

XiangXing Shi

Zong Xiong Z X Wang

Ling Ling Sh Hu

Jiezhen Lin

In an approach, a processor receives, at an edge node, a message from an IoT device associated with the edge node, the message being embedded with at least one non-fungible token (NFT) and each of the at least one NFT representing a corresponding authorization associated with the IoT device. A processor retrieves, at the edge node, the at least one NFT from the received message. A processor validates, at the edge node, the received message based on the retrieved at least one NFT. A processor, responsive to validating the received message, forwards, by the edge node, the received message to a center node associated with the edge node.

Authorization management in an internet of things (IoT) network

In an approach, a processor obtains an encrypted data key and a first encrypted protection key from a storage device. A processor sends the first encrypted protection key to a first device. A processor obtains a protection key from the first device, wherein the protection key is generated by the first device through decrypting the first encrypted protection key. A processor decrypts the encrypted data key using the protection key to obtain a data key.

Data protection optimization

Shared materialized views are maintained during data changes to the primary data and during creation of new materialized views. Shared data stored for use by shared materialized views is distinguished from data stored by an independent materialized view. A view selector manages data updates to shared materialized views and corresponding mapping table. The view selector directs movement of data between a shared materialized view and an independent materialized view through the lifecycle of the materialized views.

Updating shared and independent materialized views in a multi-tenant environment

Aspects include applying a patch to a first instance of software code and verifying that the first instance of the software code with the patch applied is operating as expected. The verifying includes sending a request to the first instance of the software code and sending the request to a second instance of the software code where the patch was previously applied. The verifying also includes comparing a first result received from the first instance of the software code in response to the request and a second result received from the second instance of the software code in response to the request. The verifying further includes determining that the first instance of the software code with the patch applied is operating as expected based on the first result being within a similarity threshold of the second result.

Managing and distributing patches for multi-tenant applications

In an approach to improve asynchronous replication in database clusters embodiments retain committed transactions during database cluster replication. Embodiments build and run asynchronous replication logic using database nodes running as one or more standalone nodes, wherein the asynchronous replication logic is responsible for data replication from a primary database node to one or more secondary database nodes. Embodiments register a replication reader to a sender node to replicate journal logs and replay the journal logs on the one or more secondary database nodes. Further, embodiments monitor replication lag metrics from the sender node and notify a replication source and a replication target of a generated replication plan, wherein the cluster controller instructs the replication source and the replication target to be ready to initiate the replication plan. Additionally, embodiments adjust a replication plan to allow the asynchronous replication logic to catch up and complete uncommitted transactions.

Retaining committed transactions during database cluster replication

A system is provided that includes a server having a first engine, a second engine, a search word warehouse, and a purchasing engine. The first engine (i) maps topics to product features of products using a topic hotness method which computes a hotness degree of a product on a topic from (a) product identifiers, (b) product features, (c) customer comments, and (d) topics with relevant key words, and (ii) builds a model providing a relevance degree of a topic to a product feature. The model is built based on the hotness degree. The second engine extracts remarkable topics based on product features using a method applied to the relevance degree of the topic to the product feature to identify remarkable topics for each product. The search word warehouse stores the remarkable topics for each product. The purchasing engine conducts purchases of products identified in searches using the remarkable topics.

Scenario enhanced search with product features

Disclosed are a method, apparatus and system for instant message routing. The method includes: receiving a communication request from a source client to a destination client; acquiring the context associated with the source client and the destination client; and selecting a sub-user of the destination client for communication using a policy according to the context. When a source client requests communication with a virtual user, the method can route the communication request using a policy according to the context, to a sub-user of the virtual user which once communicated with the source client.

Instant message (IM) routing to a virtual user consisting of a group of possible sub-users associated with a common IM identity

Growing community of inventors

Beijing, China

Zhang Li