Automated service-oriented performance management
AssigneeSysdig, Inc.
InventorLoris Degioanni
Life of this patent
Estimating the term…
Abstract
A computer-implemented method of monitoring applications executing on a plurality of computing nodes, includes the steps of: deploying agents on each of the computing nodes to detect operations performed on different objects by the executing applications; at each of the computing nodes, storing by a corresponding one of the agents, records of the operations performed on the different objects by the executing applications and associating each of the operations with a set of metadata properties; organizing the operations into groups based on the metadata properties; for one of the groups, generating a monitoring rule based on properties of the stored records of the operations of the one group; and configuring the agents according to the generated monitoring rule.
Claims20
A computer-implemented method of monitoring applications executing on a plurality of computing nodes, said method comprising:
- deploying agents on the computing nodes, at least one agent on each computing node, to detect operations performed on different objects by the executing applications;
- at each of the computing nodes, storing by a corresponding one of the agents, records of the operations performed on the different objects by the executing applications, that contain properties of the operations performed on the different objects by the executing applications, in a plurality of program tables that include a first program table for storing the records of the operations performed on the different objects by a first executing application and a second program table for storing the records of the operations performed on the different objects by a second executing application, and associating each of the operations with a set of metadata properties;
- organizing the operations having records stored in the plurality of program tables into groups based on the metadata properties;
- for one of the groups, generating a monitoring rule based on the properties of the the operations of the one group that are stored in the records thereof;
- and
- configuring the agents according to the generated monitoring rule.
A non-transitory computer readable medium comprising instructions that are executable in a computer system to carry out a method of monitoring applications executing on a plurality of computing nodes, wherein agents are deployed on the computing nodes, at least one agent on each computing node, to detect operations performed on different objects by the executing applications, said method comprising:
- at each of the computing nodes, storing by a corresponding one of the agents, records of the operations performed on the different objects by the executing applications, that contain properties of the operations performed on the different objects by the executing applications, in a plurality of program tables that include a first program table for storing the records of the operations performed on the different objects by a first executing application and a second program table for storing the records of the operations performed on the different objects by a second executing application, and associating each of the operations with a set of metadata properties;
- organizing the operations having records stored in the plurality of program tables into groups based on the metadata properties;
- for one of the groups, generating a monitoring rule based on the properties of the operations of the one group that are stored in the records thereof;
- and
- configuring the agents according to the generated monitoring rule.
A distributed computing environment comprising:
- a plurality of computing nodes having agents deployed thereon, at least one agent on each computing node, to detect operations performed on different objects by applications executing thereon, store records of the operations performed on the different objects by the executing applications, that contain properties of the operations performed on the different objects by the executing applications, in a plurality of program tables that include a first program table for storing the records of the operations performed on the different objects by a first executing application and a second program table for storing the records of the operations performed on the different objects by a second executing application, and associate each of the operations with a set of metadata properties;
- and
- a monitoring computing system that is programmed to perform the steps of:
- organizing the operations having records stored in the plurality of program tables into groups based on the metadata properties;
- for one of the groups, generating a monitoring rule based on the properties of the operations of the one group that are stored in the records thereof;
- and
- transmitting the monitoring rule to the computing nodes,
- wherein the agents are configured according to the transmitted monitoring rule.
Dependencies are inferred from the claim text.