System and method for generating performance models of complex information technology systems
AssigneePerot Systems Corporation
InventorsPieter Willem Adriaans, Arno Jan Knobbe, Marc Gathier
Life of this patent
Estimating the term…
Abstract
A system and method for automatically creating performance models of complex information technology (IT) systems. System components and elements are subject to periodic monitoring associated with performance thresholds. A continuity analysis is performed by synchronizing testing functions associated with the predetermined system performance thresholds. Resulting data is accumulated and data mined for component and functional relations within the IT system. Models of the system may then be adapted with results generated from the analysis.
Claims59
In an information technology system having a multiplicity of interconnected nodes, a method for optimizing performance monitoring of said system, said method comprising the steps of:
- (a) performing a continuity analysis on said system;
- (b) automatically generating a plurality of performance models of said system based on periodic measurements of predefined performance levels;
- (c) continuously monitoring, at a plurality of said nodes, the performance of said system at the respective plurality of said nodes;
- (d) collecting, periodically, performance data on said system at said respective nodes;
- (e) applying a plurality of data mining techniques to said periodically collected system performance data and its associated test program data;
- (f) generating a decision tree using said periodically collected system performance data, said decision tree having a multiplicity of decision nodes, each said decision node corresponding to a component of said system;
- (g) comparing a plurality of relationships within said system between said system performance data and said test program data;
- (h) automatically modifying said steps of continuously monitoring and periodically collecting said system performance data at a plurality of said nodes, whereby said autonomous modification iteratively optimizes said continuous performance monitoring of said system;
- and
- (i) automatically updating an adaptive system model according to newly discovered relationships.
An information technology system having a multiplicity of interconnected nodes, said system comprising:
- performance means for performing a continuity analysis on said system;
- generating means for automatically generating a plurality of performance models of said system based on periodic measurements of predefined performance levels;
- monitor means, for continuously monitoring, at a plurality of said nodes, the performance of said system at the respective nodes;
- collection means for periodically collecting, at said plurality of said nodes, performance data for said system at specific time intervals;
- data mining technique application means for applying a plurality of data mining techniques to said collected system performance data and its associated test program data;
- decision tree generation means for generating a decision tree using said periodically collected system performance data, said decision tree having a multiplicity of decision nodes, each said decision node corresponding to a component of said system;
- comparison means for comparing a plurality of relationships within said system between said system performance data and said test program data;
- modification means for automatically modifying said monitor and collection means for the continuous monitoring and periodic collection, respectively, of said system performance data;
- and
- updating means for automatically updating an adaptive system model according to newly discovered relationships.
An article of manufacture comprising a computer usable medium having computer readable program code means embodied thereon for optimizing performance monitoring of at least one node in an information technology system, the computer readable program code means in said article of manufacture comprising:
- computer readable program code means for:
- (a) performing a continuity analysis on said system;
- (b) automatically generating a plurality of performance models of said system based on periodic measurements of predefined performance levels;
- (c) continuously monitoring, at a plurality of said nodes, the performance of said system at the respective plurality of said nodes;
- (d) collecting, periodically, performance data on said system at said respective nodes;
- (e) applying a plurality of data mining techniques to said periodically collected system performance data and its associated test program data;
- (f) generating a decision tree using said periodically collected system performance data, said decision tree having a multiplicity of decision nodes, each said decision node corresponding to a component of said system;
- (g) comparing a plurality of relationships within said system between said system performance data and said test program data;
- (h) automatically modifying said steps of continuously monitoring and periodically collecting said system performance data at a plurality of said nodes, whereby said autonomous modification iteratively optimizes said continuous performance monitoring of said system;
- and
- (i) automatically updating an adaptive system model according to newly discovered relationships.
Dependencies are inferred from the claim text.