Adaptive metric pruning

Patent metadata

Field	Value
Publication number	US-10713591-B2
Application number	US-201615224416-A
Country	US
Kind code	B2
Filing date	Jul 29, 2016
Priority date	Jul 29, 2016
Publication date	Jul 14, 2020
Grant date	Jul 14, 2020

Abstract

A system for providing adaptive metric pruning includes a processor; a memory; and one or more modules stored in the memory and executable by a processor to perform operations including: receive, by a collector, metadata associated with the business transaction running in a monitored environment; train a machine learning system by providing training data and an anticipated result for the training data to the machine learning system to generate rules for retaining given metadata; predict a retention requirement for the received metadata by providing the received metadata to the machine learning system to apply the generated rules and generate a result for retaining the received metadata; and provide a user interface to display the generated result for retaining the received metadata including a recommendation on how to retain the received metadata according to the result for retaining the received metadata.

First claim

What is claimed is:

1. A system for providing adaptive metric pruning associated with a monitored business transaction running on multiple tiers of nodes and distributed components in a monitored environment, including:
a processor;
a memory; and
one or more modules stored in the memory and executable by the processor to perform operations including:
receive, by a collector, metadata associated with the monitored business transaction, wherein the monitored business transaction is created by detecting an incoming request at an entry point of the monitored environment and tracking activity associated with the incoming request from an originating tier of the multiple tiers to one or more distributed components of the monitored environment, further wherein the metadata is gathered by a plurality of agents executing on devices of an end-to-end processing path used to fulfill a service request for the monitored business transaction in the monitored environment;
train a machine learning system by providing training data and an anticipated result for the training data to the machine learning system to generate rules for retaining previously received metadata;
predict a retention requirement for the received metadata by providing the received metadata to the machine learning system to apply the generated rules and generate a result for retaining the received metadata, wherein different weights are applied to parameters of the received metadata; and
provide a user interface to display the generated result for retaining the received metadata including a recommendation on how to retain the received metadata according to the result for retaining the received metadata.

2. The system of claim 1 , wherein the one or more modules are executable by the processor to train the machine learning system including applying a machine learning algorithm to the training data and the anticipated result to generate the rules for retaining the previously received metadata, wherein the rules include categorizing the previously received metadata into different groups having different retentions.

3. The system of claim 2 , wherein the different groups having different retentions include a group for removing from cache, a group for removing from disk, or a group for further monitoring.

4. The system of claim 2 , wherein the one or more modules are executable by the processor to predict the retention requirement including applying the machine learning algorithm to the received metadata to apply the generated rules to categorize the received metadata into the different groups.

5. The system of claim 4 , wherein the one or more modules are executable by the processor to provide feedback to the machine learning system to update the generated rules.

6. The system of claim 5 , wherein the one or more modules are executable by the processor to provide the feedback by adding the generated result for retaining the received metadata as new training data for the machine learning system.

7. The system of claim 1 , wherein the metadata includes metrics, events, logs, snapshots, or configurations.

8. The system of claim 7 , wherein the metadata are associated with monitored data including average response time, calls per minute, load, or number of slow calls.

9. A method for providing adaptive metric pruning associated with a monitored business transaction running on multiple tiers of nodes and distributed components in a monitored environment, including:
receiving, by a collector, metadata associated with the monitored business transaction, wherein the monitored business transaction is created by detecting an incoming request at an entry point of the monitored environment and tracking activity associated with the incoming request from an originating tier of the multiple tiers to one or more distributed components of the monitored environment, further wherein the metadata is gathered by a plurality of agents executing on devices of an end-to-end processing path used to fulfill a service request for the monitored business transaction in the monitored environment;
training a machine learning system by providing training data and an anticipated result for the training data to the machine learning system to generate rules for retaining previously received metadata;
predicting a retention requirement for the received metadata by providing the received metadata to the machine learning system to apply the generated rules and generate a result for retaining the received metadata, wherein different weights are applied to parameters of the received metadata; and
providing a user interface to display the generated result for retaining the received metadata including a recommendation on how to retain the received metadata according to the result for retaining the received metadata.

10. The method of claim 9 , wherein training the machine learning system includes applying a machine learning algorithm to the training data and the anticipated result to generate the rules for retaining the previously received metadata, wherein the rules include categorizing the previously received metadata into different groups having different retentions.

11. The method of claim 10 , wherein the different groups having different retentions include a group for removing from cache, a group for removing from disk, or a group for further monitoring.

12. The method of claim 10 , wherein predicting the retention requirement includes applying the machine learning algorithm to the received metadata to apply the generated rules to categorize the received metadata into the different groups.

13. The method of claim 12 , including providing feedback to the machine learning system to update the generated rules.

14. The method of claim 13 , wherein providing the feedback includes adding the generated result for retaining the received metadata as new training data for the machine learning system.

15. A non-transitory computer readable medium embodying instructions for providing adaptive metric pruning associated with a monitored business transaction running on multiple tiers of nodes and distributed components in a monitored environment, that, when executed by a processor, cause operations to be performed including:
receiving, by a collector, metadata associated with a business transaction, wherein the monitored business transaction is created by detecting an incoming request at an entry point of the monitored environment and tracking activity associated with the incoming request from an originating tier of the multiple tiers to one or more distributed components of the monitored environment, further wherein the metadata is gathered by a plurality of agents executing on devices of an end-to-end processing path used to fulfill a service request for the monitored business transaction in the monitored environment;
training a machine learning system by providing training data and an anticipated result for the training data to the machine learning system to generate rules for retaining previously received metadata;
predicting a retention requirement for the received metadata by providing the received metadata to the machine learning system to apply the generated rules and generate a result for retaining the received metadata, wherein different weights are applied to parameters of the received metadata; and
providing a user interface to display the generated result for retaining the received metadata including a recommendation on how to retain the received metadata according to the result for retaining the received metadata.

16. The non-transitory computer readable medium of claim 15 , wherein

Assignees

• Appdynamics Llc
• Cisco Tech Inc

Inventors

• Gangadharappa Kiran Kuluvalli

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• G06Q10/067

  Enterprise or organisation modelling · CPC title

• G06N5/04

  Inference or reasoning models · CPC title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06N5/025

  Extracting rules from data · CPC title