View source for 20240020157. IDENTIFYING HOTSPOTS AND COLDSPOTS IN FORECASTED POWER CONSUMPTION DATA IN AN IT DATA CENTER FOR WORKLOAD SCHEDULING simplified abstract (HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP)

=IDENTIFYING HOTSPOTS AND COLDSPOTS IN FORECASTED POWER CONSUMPTION DATA IN AN IT DATA CENTER FOR WORKLOAD SCHEDULING=


==Organization Name==

[[:Category:HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP|HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP]]


[[Category:HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP]]

==Inventor(s)==

[[:Category:MANTEJ SINGH Gill of Bangalore (IN)|MANTEJ SINGH Gill of Bangalore (IN)]][[Category:MANTEJ SINGH Gill of Bangalore (IN)]]

[[:Category:DHAMODHRAN Sathyanarayanamurthy of Bangalore (IN)|DHAMODHRAN Sathyanarayanamurthy of Bangalore (IN)]][[Category:DHAMODHRAN Sathyanarayanamurthy of Bangalore (IN)]]

[[:Category:ARUN Mahendran of Bangalore (IN)|ARUN Mahendran of Bangalore (IN)]][[Category:ARUN Mahendran of Bangalore (IN)]]

==IDENTIFYING HOTSPOTS AND COLDSPOTS IN FORECASTED POWER CONSUMPTION DATA IN AN IT DATA CENTER FOR WORKLOAD SCHEDULING - A simplified explanation of the abstract==


This abstract first appeared for US patent application 20240020157 titled 'IDENTIFYING HOTSPOTS AND COLDSPOTS IN FORECASTED POWER CONSUMPTION DATA IN AN IT DATA CENTER FOR WORKLOAD SCHEDULING

==Simplified Explanation==

The patent application describes a system and method for using historical input power data from a server in an IT data center to train a machine learning model. This model can then be used to forecast the power consumption of the server for future time periods. The forecasted power consumption data is analyzed to identify time windows of hotspots (over-utilization) and coldspots (under-utilization) in the server's power consumption.

* The system uses historical input power periodic data from a server to train a machine learning model.
* The trained model is used to forecast the power consumption of the server for future time periods.
* Time windows of hotspots (over-utilization) and coldspots (under-utilization) are identified in the forecasted power consumption data.
* Hotspots are areas or regions of over-utilization in the time series data, while coldspots are areas or regions of under-utilization.
* Hotspots and coldspots are identified by calculating an exponential mean average (EMA) of the forecasted power consumption data.
* Points above the EMA are considered hotspots, and points below the EMA are considered coldspots.
* The identified hotspots and coldspots can be used to schedule workloads for a server or a data center, plan existing workloads more efficiently, or introduce new workloads at more optimal time periods.

Potential applications of this technology:

* Efficient workload scheduling for servers or data centers.
* Improved planning of existing workloads.
* Optimal introduction of new workloads.

Problems solved by this technology:

* Inefficient workload scheduling.
* Lack of accurate power consumption forecasting.
* Inability to identify over-utilization or under-utilization in power consumption.

Benefits of this technology:

* Increased efficiency in workload scheduling.
* Better utilization of server resources.
* Cost savings through optimized power consumption planning.


==Original Abstract Submitted==

systems and methods are provided for using historic input power periodic data from a server in an it data center to train a machine learning (ml) model to obtain forecasted power consumption data of the server for a future time period. time windows of hotspots or coldspots are then identified in the forecasted power consumption data, hotspots being defined as areas or regions of over-utilization in a time series data, and coldspots being defined as areas or regions of under-utilization in a time series data. the hotspots and coldspots are identified by calculating an exponential mean average (ema) of the forecasted power consumption data, taking points above the ema as hotspots and points below the ema as coldspots. the identified hotspots and coldspots can be used to schedule workloads for a server or a data center, to more efficiently plan existing workloads, or to introduce new workloads at more optimal time periods.



[[Category:G06F9/50]]
[[Category:G06N20/00]]