Systems and methods for dynamic capacity planning of network
An apparatus for capacity planning in a network includes at least one memory storing instructions and at least one processor configured to execute the instructions to obtain network data corresponding to at least one network node and historic dimensional data corresponding to the network, generate transformed data from the network data and the historic dimensional data based on at least one configuration parameter, input the transformed data into a machine learning (ML) model, generate, by the ML module, at least one capacity prediction of the network based on the transformed data, and adjust at least one dimensioning parameter of the network based on the at least one capacity prediction.
1 . An apparatus for capacity planning in a network, the apparatus comprising:
at least one memory storing instructions; and
at least one processor configured to execute the instructions to:
obtain, during a predetermined time frame, network data corresponding to at least one network node and historic dimensional data corresponding to the network;
obtain, from a preconfigured configuration file, at least one threshold condition corresponding to a network parameter threshold, at least one configuration parameter and at least one filtering rule associated with data being obtained during the predetermined time frame;
based on the at least one threshold condition being met, generate transformed data from the network data and the historic dimensional data based on the at least one configuration parameter and the at least one filtering rule;
input the transformed data into a machine learning (ML) model;
generate, by the ML model, at least one temporary capacity prediction of the network based on the transformed data; and
adjust at least one dimensioning parameter of the network based on the at least one temporary capacity prediction,
wherein the at least one threshold condition comprises a central processing unit (CPU) utilization of the network being above a predetermined utilization threshold or a CPU utilization of the at least one network node being above a predetermined utilization threshold.
2 . The apparatus of claim 1 , wherein the at least one processor is further configured to execute the instructions to:
detect an anomaly in the network based on at least one of the network data, the historic dimensional data and the transformed data; and
generate an alarm in response to detecting the anomaly in the network.
3 . The apparatus of claim 2 , wherein the at least one processor is configured to execute the instructions to adjust the at least one dimensioning parameter based on the generated alarm.
4 . The apparatus of claim 1 , wherein the at least one processor is configured to execute the instructions to generate the transformed data based on the at least one filtering rule.
5 . The apparatus of claim 4 , wherein the at least one filtering rule comprises at least one of:
excluding at least a portion of network data or at least a portion of historic dimensional data corresponding to the predetermined time frame; and
prioritizing at least a portion of network data or at least a portion of historic dimensional data based on at least one data feature.
6 . The apparatus of claim 1 , wherein the historic dimensional data comprises at least one of a data of dimensioning of the network, a time of dimensioning of the network, a CPU utilization of the network, a memory utilization of the network, and an added processing capacity to the network.
7 . The apparatus of claim 6 , wherein the added processing capacity to the network comprises at least one of an added CPU to the network, an added memory to the network, and an added server to the network.
8 . The apparatus of claim 1 , wherein the network data corresponding to the at least one network node comprises at least one of an aggregation of a number of successful connections, an aggregation of a number of failed connections, an aggregation of a number of timeouts, a total number of subscribers on the network, a memory utilization percentage, and a number of transactions per a predetermined time period.
9 . A method for capacity planning in a network, the method comprising:
obtaining, during a predetermined time frame, network data corresponding to at least one network node and historic dimensional data corresponding to the network;
obtaining, from a preconfigured configuration file, at least one threshold condition corresponding to a network parameter threshold, at least one configuration parameter and at least one filtering rule associated with data being obtained during the predetermined time frame;
based on the at least one threshold condition being met, generating transformed data from the network data and the historic dimensional data based on the at least one configuration parameter and the at least one filtering rule;
inputting the transformed data into a machine learning (ML) model;
generating, by the ML model, at least one temporary capacity prediction of the network based on the transformed data; and
adjusting at least one dimensioning parameter of the network based on the at least one temporary capacity prediction,
wherein the at least one threshold condition comprises a central processing unit (CPU) utilization of the network being above a predetermined utilization threshold or a CPU utilization of the at least one network node being above a predetermined utilization threshold.
10 . The method of claim 9 , further comprising:
detecting an anomaly in the network based on at least one of the network data, the historic dimensional data and the transformed data; and
generating an alarm in response to detecting the anomaly in the network.
11 . The method of claim 10 , wherein the at least one dimensioning parameter is adjusted based on the generated alarm.
12 . The method of claim 9 , wherein the transformed data is generated based on the at least one filtering rule.
13 . The method of claim 12 , wherein the at least one filtering rule comprises at least one of:
excluding at least a portion of network data or at least a portion of historic dimensional data corresponding to the predetermined time frame; and
prioritizing at least a portion of network data or at least a portion of historic dimensional data based on at least one data feature.
14 . The method of claim 9 , wherein the historic dimensional data comprises at least one of a data of dimensioning of the network, a time of dimensioning of the network, a CPU utilization of the network, a memory utilization of the network, and an added processing capacity to the network.
15 . The method of claim 14 , wherein the added processing capacity to the network comprises at least one of an added CPU to the network, an added memory to the network, and an added server to the network.
16 . The method of claim 9 , wherein the network data corresponding to the at least one network node comprises at least one of an aggregation of a number of successful connections, an aggregation of a number of failed connections, an aggregation of a number of timeouts, a total number of subscribers on the network, a memory utilization percentage, and a number of transactions per a predetermined time period.
17 . A non-transitory computer-readable storage medium storing instructions that, when executed by at least one processor, cause the at least one processor to:
obtain, during a predetermined time frame, first data corresponding to at least one network node in a network and second data corresponding to the network;
obtain, from a preconfigured configuration file, at least one threshold condition corresponding to a network parameter threshold, at least one configuration parameter and at least one filtering rule associated with data being obtained during the predetermined time frame;
based on the at least one threshold condition being met, generate transformed data from the first data and the second data based on the at least one configuration parameter and the at least one filtering rule;
input the transformed data into a machine learning (ML) model;
generate, by the ML model, at least one temporary capacity prediction of the network based on the transformed data; and
adjust at least one dimensioning parameter of the network based on the at least one temporary capacity prediction,
wherein the at least one threshold condition comprises a central processing unit (CPU) utilization of the network being above a predetermined utilization threshold or a CPU utilization of the at least one network node being above a predetermined utilization threshold.
18 . The storage medium of claim 17 , wherein the instructions, when executed by the at least one processor, further cause the at least one processor to:
detect an anomaly in the network based on at least one of the first data corresponding to the at least one network node, the second data corresponding to the network and the transformed data; and
generate an alarm in response to detecting the anomaly in the network.