International Business Machines Corporation

Daniel Nikolaus Bauer

John G Rooney

Paolo Scotton

Luis Garcés-Erice

Luis Garcés Erice

Kristijan Dragicevic

Urs Hunkeler

Peter Urbanetz

Hao Yang

Hui Lei

Minkyong Kim

Thomas A Jobson, Jr

Frank Bagehorn

Marcel Waldvogel

Adeel Qureshi

Liangzhao Zeng

Han Chen

Gidon Gershinsky

Nicholas David Butler

Gero Dittmann

Konstantin Shagin

Marc-Thomas H Schmidt

Florian Hermann Froese

David Rushall

Ilias Iliadis

Paul T Hurley

Andreas Kind

Christopher J Giblin

Sean G Rooney

John N Daigle

Christos-Xenofontas Dimitropoulos

Dimitrios Dechouniotis

John Daigle

A technique for retaining a context in which data resides independently of a data store from which the data originates is disclosed. In relation to a method aspect of the technique, a computer-implemented method provides data with related first metadata, both originating from a data store and extracts the data and the related first metadata independently from the data store. A universal unique identifier of a portion of the data to which portion specific first metadata exists is created as part of the related first metadata. The universal unique identifier of the portion of the data is integrated into the related first metadata, thereby creating modified first metadata as an independently manageable and linkable representation of the related first metadata.

Preserving metadata context in a hybrid cloud context

A method includes obtaining a first snapshot of a source system and deriving a set of key-value pairs reflecting the first snapshot. A mirror operation of the source system is performed to obtain CDC change operations. The method obtains a first CDC log as a first sequence of key-value pairs. A second snapshot of the source system is obtained and a set of key-value pairs is derived, which reflects the second snapshot. The first sequence of key-value pairs is compared with the set of key-value pairs to derive corrective CDC operations, captured as a set of key-value pairs. The corrective CDC operations represent corrections to be performed with respect to the first sequence of key-value pairs. A second CDC log is obtained as a second sequence of key-value pairs. The corrective CDC operations ensure that the second sequence of key-value pairs are coherent with the set of key-value pairs.

Tracking change data capture log history

A computer-implemented method for transferring changes made to data from a data source to a data destination may be provided. The method involves using a message broker with streaming channels for transferring data changes from a data source to a data consumer. To determine if data has changed, a portion of data labeled with a key is sent through a hash function. A hash value with the corresponding key that has already been sent to the data consumer is retrieved from a streaming channel and compared to the newly created hash value. If the newly created hash value is different the data is sent to data consumers by the message broker via one streaming channel, while the new hash value is stored in the other streaming channel.

Universal streaming change data capture

The present disclosure relates to a method for processing data of a database system, the database system being configured to connect to a centralized cache shared with other database systems. The method includes determining a unique identifier of the content of a table of the database system using a common identification method of the database systems. A database command may be generated for processing data of the table. The command references the identifier. It may be determined if the cache comprises results of a same command with the same identifier. The command may be processed at the database system or the results of the command may be retrieved from the cache depending on whether the cache comprises the results.

Processing data of a database system

A computer implemented method for enabling data coherency between a first data source and a second data source of different structure may be provided. The first data source has a first format and a related first log file. Similar conditions apply to the second data source. The method comprises also determining a first entry in the first log file at which the data have a semantic consistency and storing this as first read barrier. The same applies for an entry in the second data source. Additionally, the method comprises building a third data set derived from data of the first data source and data of the second data source and being coherent with them, using data from the first data source up to the first read barrier and data from the second data source up to the second read barrier.

Global coherency for distributed data sources

A computerized method for ingesting data from a relational database into a data lake is provided, wherein a user-defined function (UDF) is associated with a standard operation of extract, transform, load, or ETL, of an ETL pipeline. This UDF is triggered upon performing the standard operation and thereby allow a code associated with the UDF to be executed. Upon migrating data from one or more data sources into the relational database, the standard operation is executed, which triggers the UDF and, in turn, an execution of the code. As per the execution of this code, an entity running on the data lake is notified that a set of data migrated to the relational database is to be ingested according to given ingestion modalities specified by the code. Finally, the set of data can be ingested into the data lake according to the modalities. Related computer program products are also provided.

Computerized methods and programs for ingesting data from a relational database into a data lake

A mechanism for declarative service domain federation uses a declarative approach to both the visibility aspect and the management aspect of service domain federation. Instead of manually exporting services, that is, selecting individual services from source domains to be visible in the target domain, using the present system, a federation architect uses a federation console to specify federation intent. The federation intent may include federation rule, query-based service group, and quality-of-service (QoS) policy. Based on the declared federation intent, a planning engine resolves the individual services based on the group query, generates the service visibility configuration, and creates the necessary service proxies and/or mediation objects to satisfy the goal. At runtime, a service monitor collects the performance metrics of federated services and dynamically adjusts the mediation/proxy configurations to maintain the QoS objectives specified by the federation architect.

Declarative service domain federation

An approach is provided for aggregating business data. Based on received columns, attributes, and keys of a dataset, the dataset is registered in a central hub of datasets which include data that is independently and locally maintained. Other datasets in the central hub that include columns, attributes, or keys that match the received columns, attributes, or keys, respectively, are determined and displayed. Responsive to receiving a user selection of one or more of the displayed datasets, cross-links between the dataset and each of the one or more selected datasets are generated. Based on a modification of a selected one of the cross-links and a second user interacting with the central hub, the modification and other cross-link(s) are displayed.

User driven business data aggregation and cross mapping framework

Embodiments of the invention is directed to a method for connecting a device to a network. An example method comprises providing a device assigned with a device identifier and an asymmetric cryptographic key pair that includes a public key and a private key. The device stores the private key on a memory thereof. The device is provided with information as to the assigned device identifier and/or the public key. This information is detectable by a detector so as to be transmissible to a server for it to identify the device identifier and the public key assigned to the device.

Efficient and secure connection of devices to a network without user interfaces

A plurality of protocol stacks are deployed. Each of the protocol stacks includes a plurality of composable protocol modules, and each of the composable protocol modules implements common interfaces. It is detected that a first given one of a plurality of clients wishes to connect to a publication-subscription engine and it is determined whether the first given one of the plurality of clients is to be connected in a secure manner. Responsive to determining that the first given one of the plurality of clients is to be connected in the secure manner, an encrypted instance of a first appropriate one of the plurality of protocol stacks is instantiated to effectuate the secure connection. The first given one of the plurality of clients is authenticated and authorized.

Introducing encryption, authentication, and authorization into a publication and subscription engine

An approach is provided for aggregating business data. Based on received columns, attributes, and keys of a dataset, the dataset is registered in a central hub of datasets which include data that is independently and locally maintained. Other datasets in the central hub that include columns, attributes, or keys that match the received columns, attributes, or keys, respectively, are determined and displayed. Responsive to receiving a user selection of one or more of the displayed datasets, cross-links between the dataset and each of the one or more selected datasets are generated. Based on a modification of a selected one of the cross-links and a second user interacting with the central hub, the modification and other cross-link(s) are displayed. Based on the second user's selection of the modification or one of the other cross-link(s), an analysis of data in the dataset is performed.

Embodiments of the invention provide a method, system, and computer program product for selecting a server as a selected server for performing a specified function in a distributed computing system. In one embodiment, the method comprises identifying a group of servers in the system; and choosing as the selected server the server that has been in the system the longest. In one embodiment, each of the servers in the group has a respective contention period, during which the server sends a contention message, and the selected server is chosen using these contention messages. The servers in the group may have imperfect clock synchronization. In this case, the contention periods of the servers is kept greater than the maximum difference between the server clocks. The invention may be used to select a name server, or to select a server for another service.

Distributed server election with imperfect clock synchronization

An approach is provided for aggregating business data. Based on received columns, attributes, and keys of a dataset, the dataset is registered in a central hub of datasets which include data that is independently and locally maintained. Initial or updated data is loaded into the dataset in the central hub. Other datasets in the central hub that include columns, attributes, or keys that match the received columns, attributes, or keys, respectively, are determined and displayed. In response to receiving a user selection of one or more of the displayed datasets, cross-links between the dataset and each of the one or more selected datasets are automatically generated. The cross-links are based on the columns, attributes, or keys of the displayed datasets that match the received columns, attributes or keys. The cross-links provide an aggregation of business data in the dataset and the one or more selected datasets.

Growing community of inventors

Birmensdorf, Switzerland

Daniel Nikolaus Bauer