International Business Machines Corporation

Lenovo (singapore) Pte. Ltd.

Keyur Patel

Aspen L Payton

James M Caffrey

Karla Kay Arndt

Robert Andrew Vrabel

SathyaKumar Andre Ramalingam

Albert D Bennah

Surendra Kodali

Robert M Morgan

Robert L Snyder

Techniques are described for monitoring a performance metric. A multiple modeling approach is used to improve predictive analysis by avoiding the issuance of warnings during spikes which occur as a part of normal system processing. This approach increases the accuracy of predictive analytics on a monitored computing system, does not require creating rules defining periodic processing cycles, reduces the amount of data required to perform predictive modeling, and reduces the amount of CPU required to perform predictive modeling.

Multiple modeling paradigm for predictive analytics

Methods, apparatuses, and computer program product for of efficient network bandwidth utilization in a distributed processing system are provided. Embodiments include monitoring, by a network monitor, network usage of a distributed processing system containing a plurality of endpoint devices; creating, by the network monitor, a historical network usage pattern based on the monitoring of the network usage; based on the historical network usage pattern, identifying, by the network monitor, a first set of future time periods predicted to correspond with low network usage; and selecting from the first set of future time periods, by the network monitor, a particular time period to perform a network administrative task on an endpoint device in the distributed processing system.

Efficient network bandwidth utilization in a distributed processing system

Dynamically altering error logging activities in a computing system, including: receiving, by an error logging manager, historical error resolution data; identifying, by the error logging manager in dependence upon the historical error resolution data, a plurality of computing components associated with each error contained in the historical error resolution data; and associating, by the error logging manager in a related component repository, an identification of each of the plurality of computing components associated with each error and an identification of the error.

Dynamically altering error logging activities in a computing system

Damage to a software stack accessed by multiple application programs can be detected by monitoring the rate of stack failures at the stack via a stack monitor and comparing the rate of stack failures with a time related threshold. An alarm is generated when the rate of stack failures exceeds the time related threshold.

Damaged software system detection

Growing community of inventors

Poughkeepsie, NY, United States of America

Keyur Patel