International Business Machines Corporation

Kyndryl, Inc.

Qi Ming Teng

Guang Cheng Li

Ying Li

Yong Sen Zheng

Tian Cheng Liu

Jie Qiu

Yubo Li

Lin Feng Shen

Rong Yan

Cheng Fang Wang

Jinho Hwang

Sai Zeng

Evelyn Duesterwald

Li Cao

Ling Shao

Andrew John Schofield

David Alvra Wood, Iii

William J Tracey, Ii

Bentley John Hargrave

Susan Patricia Paice

Xin Hui Li

Jun Duan

Christopher Peter Baker

Yan Yan Hu

Disclosed are a computer-implemented method, a system, and a computer program product for system-level tunable parameter identification. Performance characteristic data for an application to be tuned can be obtained by one or more processing units. At least one system-level tunable parameter for the application to be tuned can be identified by one or more processing units based on the obtained performance characteristic data for the application to be tuned and a pattern between training performance characteristic data and a set of training system-level parameter-related correlation coefficients. The set of training system-level parameter-related correlation coefficients can be respective correlation coefficients of system-level tunable parameters with respect to at least one performance metric.

System-level tunable parameter identification

Aspects of the present disclosure relate to converting between structured and tabular data formats. Data can be received in a tabular format. An array can be built for each of a plurality of objects within the data in the tabular format, each object corresponding to at least one identified header of the identified headers. A data row can be parsed using at least one of the built arrays and data within the data row can be added to the structured format in a specific location based characteristics indicated in the at least one array. Data can also be converted from the structured format into the tabular format using the built arrays.

Converting between tabular and structured data formats

A computer-implemented method, system, and computer program product are provided for optimization with domain knowledge requirements. The method includes receiving, by a processor device, domain knowledge requirements for a target system. The method also includes defining, by a domain knowledge manager, a status of the domain knowledge employing a factor responsive to the domain knowledge. The method additionally includes computing, by an effective metric generator, effective performance metrics responsive to the status of the domain knowledge requirements and real performance metrics. The method further includes generating, by an optimization engine, a target system configuration responsive to the effective performance metrics and the real performance metrics. The method also includes improving the target system by changing a state of a function in the target system responsive to the target system configuration.

Target system optimization with domain knowledge

Method and apparatus for reducing resource consumption in container image management. Embodiments include receiving directory information and checksum information corresponding to a plurality of files of a container image. Embodiments include determining, based on the directory information and the checksum information, whether each file of the plurality of files is present in a repository. For each respective file of the plurality of files that is determined to be present in the repository, embodiments include updating a file list for the container image to indicate a respective location in the repository of the respective file. For each given file of the plurality of files that is determined not to be present in the repository, embodiments include storing the given file at a given location in the repository and updating the file list for the container image to indicate the given location in the repository of the given file.

Reducing resource consumption in container image management

The present disclosure provides a computer-implemented method, computer system and computer program product for application containerization. According to the computer-implemented method, an application to be containerized can be traced. Information obtained in the tracing can be analyzed to determine one or more features of the application. An image template for the application can be created based on the one or more features. Then, a container image for the application can be built based on the image template.

Application containerization based on trace information

Distributed traffic scheduling for autonomous self-driving vehicles includes forming a vicinity group of autonomous vehicles by pinging each other using vehicle-to-vehicle communication. A leader vehicle of the vicinity group is elected by transmitting an election messages between each of the autonomous vehicles, each autonomous vehicle responding to a received election message with a vote, the leader vehicle being elected by receiving the most votes. Preferences of a driver of each of the autonomous vehicles are predicted with a classifier by extracting preference labels from online account data. Trip deadline is determined for each autonomous vehicle according to the preferences. Each of the autonomous vehicles in the vicinity group is scheduled to prevent exceeding the trip deadline by performing earliest deadline first scheduling by the leader vehicle. Maneuvers of each of the autonomous vehicles in the vicinity group are controlled according to the scheduling.

Distributed traffic scheduling for autonomous self-driving vehicles

Embodiments of the present invention relate to methods, systems, and computer program products for container image management. In a method, an image layer in a container image may be received by one or more processors, and the container image is to be stored in an image server comprising a group of image layers. A base portion may be selected by one or more processors from the group of image layers based on a similarity analysis between the image layer and the group of image layers. A patch portion may be generated by one or more processors based on a difference between the image layer and the selected base portion. With these embodiments, the container image may be stored based on multiple image layers, and thus the container image may be maintained in a much finer granularity so as to reduce requirements on the bandwidth and time cost for transmitting the container image.

Container image management

A method and system for deploying a cluster, wherein the cluster contains a plurality of cluster members, the method comprising: obtaining a unique identifier for each cluster member; according to the unique identifier of each cluster member, obtaining cluster configuration information for the cluster member; configuring the cluster member using the obtained cluster configuration information of the cluster member; and deploying the cluster. The method and system can enhance the cluster deployment efficiency.

Deploying a cluster

A method and a system of associating a runtime event with a component. The method includes catching the runtime event; catching the context of the current running environment and determining the current component based on said context; and getting the associating policy of current component and determining the responsible component associated with said runtime event based on said associating policy.

Method and system of associating a runtime event with a component

Adaptive profiling for performance analysis of a computer system controls one or more agents to monitor a plurality of events occurring in a target computer system based on an adaptive logic. Collected data may be filtered and analyzed to determine one or more contributor events that attribute to performance of the target computer system. One or more patterns are observed or detected in said collected data, behavior of said one or more agents are adjusted based on said detected one or more patterns. The adaptive logic may be further reconfigured based on said detected one or more patterns.

Mechanism for adaptive profiling for performance analysis

A method and apparatus of partitioned memory dump in a software system is provided, the apparatus comprising: a monitor for, during the execution of the software system, monitoring the memory allocations and deallocations of objects in the software system and according to the results of the monitoring, assigning the allocated objects in the software system into corresponding memory partitions, wherein the memory space of the software system is partitioned according to a given partitioning scheme; an event detector for detecting a triggering event of memory dump; and a dumper for dumping related memory partitions according to the detected triggering event of memory dump. Preferably, the apparatus further comprises an analyzer for analyzing the dumped information, and generating a new dump triggering event or a new partitioning scheme based on the analysis results.

Method and apparatus of partitioned memory dump in a software system

A method and system for increasing a system's performance and achieving improved memory utilization by approximating the memory sizes that will be required for data objects that can be deserialized and constructed in a memory cache. The method and system may use accurate calculations or measurements of similar objects to calibrate the approximate memory sizes.

Growing community of inventors

Beijing, China

Qi Ming Teng