Exponential moving maximum (EMM) filter for predictive analytics in network reporting

Patent metadata

Field	Value
Publication number	US-11153176-B2
Application number	US-201815995061-A
Country	US
Kind code	B2
Filing date	May 31, 2018
Priority date	Sep 11, 2014
Publication date	Oct 19, 2021
Grant date	Oct 19, 2021

Abstract

Techniques for an exponential moving maximum (EMM) filter for predictive analytics in network reporting are disclosed. In some embodiments, a process for predictive analytics in network reporting using an EMM filter includes pre-processing network-related data by performing exponential moving maximum (EMM) filtering on the network-related data; and determining predictive analytics based on the EMM filtered network-related data.

First claim

What is claimed is:

1. A system for predictive analytics in network reporting, comprising:
a processor configured to:
pre-process network-related data by performing exponential moving maximum (EMM) filtering on the network-related data, comprising to:
aggregate a number of transactions per a period of time of network events over a plurality of periods of time to obtain the network-related data; and
perform the EMM filtering on the network-related data, comprising to:
identify a local maximum value from the network-related data based on a filtering window size, wherein the filtering window size relates to a period of time; and
identify one or more bubble points based on one or more identified local maximum values over a range of time, wherein the range of time relates to one or more filtering window sizes, wherein a bubble point of the one or more bubble points over shadow other non-local maximum values, and wherein the one or more bubble points relates to the EMM filtered network-related data; and

determine predictive analytics based on the EMM filtered network-related data, comprising to:
generate a trend prediction based on the EMM filtered network-related data, comprising to:
predict network bottlenecks in a network deployment using the identified local maximum value; and

a memory coupled to the processor and configured to provide the processor with instructions.

2. The system recited in claim 1 , wherein the network-related data includes monitored network events data.

3. The system recited in claim 1 , wherein the network-related data includes Domain Name System (DNS) data.

4. The system recited in claim 1 , wherein the network-related data includes Domain Name System (DNS) query and DNS response data.

5. The system recited in claim 1 , wherein to determine predictive analytics based on the EMM filtered network-related data includes to:
adjust a network setting based on the determined predictive analytics.

6. The system recited in claim 1 , wherein the processor is further configured to:
generate a visualization of the trend prediction for display.

7. The system recited in claim 1 , wherein the processor is further configured to:
receive network-related data.

8. The system recited in claim 1 , wherein the processor is further configured to:
receive Domain Name System (DNS) data that is collected from an agent executed on a DNS appliance.

9. The system recited in claim 1 , wherein the processor is further configured to:
collect network-related data from one or more network devices.

10. The system recited in claim 1 , wherein the processor is further configured to:
collect network-related data from one or more network devices; and
store the network-related data in a data store.

11. A method of predictive analytics in network reporting, comprising:
pre-processing network-related data by performing exponential moving maximum (EMM) filtering on the network-related data, comprising:
aggregating a number of transactions per a period of time of network events over a plurality of periods of time to obtain the network-related data; and
performing the EMM filtering on the network-related data, comprising:
identifying a local maximum value from the network-related data based on a filtering window size, wherein the filtering window size relates to a period of time; and
identifying one or more bubble points based on one or more identified local maximum values over a range of time, wherein the range of time relates to one or more filtering window sizes, wherein a bubble point of the one or more bubble points over shadow other non-local maximum values, and wherein the one or more bubble points relates to the EMM filtered network-related data; and

determining predictive analytics based on the EMM filtered network-related data, comprising:
generating a trend prediction based on the EMM filtered network-related data, comprising:
predicting network bottlenecks in a network deployment using the identified local maximum value.

12. The method of claim 11 , wherein the network-related data includes monitored network events data.

13. The method of claim 11 , wherein the network-related data includes Domain Name System (DNS) data.

14. The method of claim 11 , wherein the network-related data includes Domain Name System (DNS) query and DNS response data.

15. The method of claim 11 , further comprising:
adjusting a network setting based on the determined predictive analytics.

16. A computer program product for predictive analytics in network reporting, the computer program product being embodied in a tangible, non-transitory computer readable storage medium and comprising computer instructions for:
pre-processing network-related data by performing exponential moving maximum (EMM) filtering on the network-related data, comprising:
aggregating a number of transactions per a period of time of network events over a plurality of periods of time to obtain the network-related data; and
performing the EMM filtering on the network-related data, comprising:
identifying a local maximum value from the network-related data based on a filtering window size, wherein the filtering window size relates to a period of time; and
identifying one or more bubble points based on one or more identified local maximum values over a range of time, wherein the range of time relates to one or more filtering window sizes, wherein a bubble point of the one or more bubble points over shadow other non-local maximum values, and wherein the one or more bubble points relates to the EMM filtered network-related data; and

determining predictive analytics based on the EMM filtered network-related data, comprising:
generating a trend prediction based on the EMM filtered network-related data, comprising:

predicting network bottlenecks in a network deployment using the identified local maximum value.

17. The computer program product recited in claim 16 , wherein the network-related data includes monitored network events data.

18. The computer program product recited in claim 16 , wherein the network-related data includes Domain Name System (DNS) data.

19. The computer program product recited in claim 16 , wherein the network-related data includes Domain Name System (DNS) query and DNS response data.

20. The computer program product recited in claim 16 , further comprising computer instructions for:
adjusting a network setting based on the determined predictive analytics.

Assignees

• Infoblox Inc

Inventors

• Yu Bin
• Smith Les
• Threefoot Mark

Classifications

• H04L41/147 Primary

  for predicting network behaviour · CPC title

• H04L41/40

  using virtualisation of network functions or resources, e.g. SDN or NFV entities · CPC title

• H04L43/20

  the monitoring system or the monitored elements being virtualised, abstracted or software-defined entities, e.g. SDN or NFV · CPC title

• H04L61/5014

  using dynamic host configuration protocol [DHCP] or bootstrap protocol [BOOTP] · CPC title

• H04L61/4511

  using domain name system [DNS] · CPC title