International Business Machines Corporation

Nicholas R Jones

Donald W Schmidt

David Harold Surman

Daniel V Rosa

Karla Kay Arndt

Nicholas C Matsakis

Joseph W Gentile

Robert Miller, Jr

Thomas F Rankin

Robert M Abrams

Douglas Michael Zobre

Andrew M Sica

Tarun Chopra

A computer-implemented method for diagnosing workload performance problems in computer servers includes measuring resource metrics. The method further includes aggregating resource metrics for each consumed resource, and aggregating resource and non-resource activity metrics for each consumer. The method further includes generating condensed diagnostic data, on a synchronized, regular interval with one or more most significant consumed resource names and corresponding consumed resource metrics. Generating the condensed diagnostic data includes grouping consumers into buckets based on bucket and consumer attributes, aggregating resource and non-resource consumer activity metrics across all consumers in each bucket, and including one or more worst offending consumer names with corresponding resource and non-resource consumer activity metrics in each bucket. The condensed diagnostic data with the most significant consumed resources and bucket contents is recorded. Diagnostic data can be generated in a human consumable form or in a machine consumable form for using machine learning.

Establishing consumed resource to consumer relationships in computer servers using micro-trend technology

A computer-implemented method for diagnosing workload performance problems in computer servers includes measuring activity metrics. The method further includes aggregating activity metrics for each user. The method further includes generating condensed diagnostic data for identifying workload performance problems on a synchronized, regular interval. Generating the diagnostic data includes grouping users into buckets based on bucket and user attributes. Generating the diagnostic data includes aggregating activity metrics across all users in each bucket. Generating the diagnostic data includes including one or more most significant users and corresponding activity metrics for each activity in each bucket, and recording bucket contents. Diagnostic data can be generated in a human consumable form for human analysis or in a machine consumable form for machine analysis through machine learning.

Diagnosing workload performance problems in computer servers

A computer-implemented method for dynamic workload bucket reassignment in a system processor includes grouping a plurality of like workloads together via a common workload attribute. The processor calculates an aggregate use of a resource across all of the workloads, and creates a finite number of buckets. The processor then assigns a target resource use for each of the finite number of buckets, and assigns workloads with a same workload attribute value to a bucket of the finite number of buckets. The workload attribute values are distributed across a plurality of the finite number of buckets such that each bucket has a unique nonintersecting range and an aggregate workload resource use in each bucket near a bucket target resource use.

Dynamic workload bucket reassignment

Aspects relate to a computer system and a computer implemented method for resolving abnormal contention on the computer system. The method includes detecting, using a processor and at a hypervisor level of the computer system, abnormal contention of a serially reusable resource caused by a first virtual machine. The abnormal contention includes the first virtual machine experiencing resource starvation of computer system resources used for processing the first virtual machine, causing the first virtual machine to block the serially reusable resource from a second virtual machine that is waiting to use the serially reusable resource. The method also includes adjusting resource allocation at the hypervisor level of the computer system resources for the first virtual machine, processing the first virtual machine based on the resource allocation, and releasing the serially reusable resource by the first virtual machine in response to the first virtual machine processing.

Method and apparatus for resolving contention at the hypervisor level

Aspects relate to a computer implemented method for resolving abnormal contention on a computer system. The method includes detecting, using a processor, abnormal contention of a serially reusable resource caused by a first process, wherein the abnormal contention includes the first process blocking the serially reusable resource from a second process that is waiting to use the serially reusable resource. The method includes collecting, in a computer storage medium, resource data in a serialized resource history database and analyzing the resource data associated with the serially reusable resource and adjusting, using the processor, resource allocation for the first process of the serially reusable resource based on the resource data. The method also includes processing, using the processor and the serially reusable resource, the first process based on the resource allocation and releasing, using the processor, the serially reusable resource by the first process in response to the first process completing.

Method and apparatus for resolving contention in a computer system

Technical solutions are described for transferring standalone dump data of a system to a diagnosis system. An example method includes detecting a creation of a multi-volume standalone dump dataset. The method also includes, in response to detecting the creation, creating an entry in a list of standalone dump datasets, the entry corresponding to the multi-volume standalone dump dataset created. The method also includes detecting a change in data stored in the multi-volume standalone dump dataset; and in response to detecting the change, updating the corresponding entry in the list of standalone dump datasets, and generating a sequential standalone dump based on the multi-volume standalone dump dataset. The method also includes alerting the diagnosis system of the sequential standalone dump. The method also includes transferring the sequential standalone dump for access by the diagnosis system.

Detection and automatic transfer of standalone system dumps

A method and computer program product for testing a system logger. Randomized input is provided into a logging facility and automated testing of the logging facility is performed in a multi-processing environment. Self-verification of data retrieved from the logging facility is performed to ensure no data loss occurs due to external system events or failures driven by a test tool. Real-time analysis of results received is performed by the test tool based on the randomized input into the logging facility. First failure data is captured and self-verification of results of a data recovery operation of the logging facility is performed in a disaster recovery scenario. Self-verification includes writing log blocks of random sizes to the log stream, iteratively forcing log data to be offloaded from primary storage to secondary storage, iteratively updating log stream attributes, and iteratively swapping duplexing modes.

Growing community of inventors

Poughkeepsie, NY, United States of America

Nicholas R Jones