View source for 17956638. IDENTIFYING IDLE-CORES IN DATA CENTERS USING MACHINE-LEARNING (ML) simplified abstract (NVIDIA Corporation)

=IDENTIFYING IDLE-CORES IN DATA CENTERS USING MACHINE-LEARNING (ML)=


==Organization Name==

[[:Category:NVIDIA Corporation|NVIDIA Corporation]]


[[Category:NVIDIA Corporation]]

==Inventor(s)==

[[:Category:Yogesh Dangi of Pune (IN)|Yogesh Dangi of Pune (IN)]][[Category:Yogesh Dangi of Pune (IN)]]

[[:Category:Manas Ranjan Jagadev of San Jose CA (US)|Manas Ranjan Jagadev of San Jose CA (US)]][[Category:Manas Ranjan Jagadev of San Jose CA (US)]]

[[:Category:Sandip Kumar of Pune (IN)|Sandip Kumar of Pune (IN)]][[Category:Sandip Kumar of Pune (IN)]]

[[:Category:Kiran Sutar of Pune (IN)|Kiran Sutar of Pune (IN)]][[Category:Kiran Sutar of Pune (IN)]]

==IDENTIFYING IDLE-CORES IN DATA CENTERS USING MACHINE-LEARNING (ML) - A simplified explanation of the abstract==


This abstract first appeared for US patent application 17956638 titled 'IDENTIFYING IDLE-CORES IN DATA CENTERS USING MACHINE-LEARNING (ML)

==Simplified Explanation==

The patent application describes a method to determine the number of idle cores of a computing device using a machine learning model based on the processes executed by the device. The method involves identifying the processes, using the ML model to decide on the number of cores to power down, and transitioning to a mode with fewer cores to save power.

* Explanation of the patent/innovation:
* Apparatuses, systems, and techniques for determining idle cores of a computing device using machine learning.
* Method involves analyzing processes executed by the device and deciding on the number of cores to power down.
* Transitioning from a high-power mode to a low-power mode based on the ML model's recommendations.

== Potential applications of this technology ==
This technology could be applied in:
* Data centers to optimize power consumption and reduce costs.
* Mobile devices to improve battery life and performance.

== Problems solved by this technology ==
This technology addresses:
* Power inefficiency in computing devices.
* Overheating issues due to unnecessary core usage.

== Benefits of this technology ==
The benefits of this technology include:
* Improved energy efficiency.
* Extended device lifespan.
* Enhanced performance during peak usage times.

== Potential commercial applications of this technology ==
Optimized Power Management in Computing Devices: Enhancing Efficiency and Performance

== Possible prior art ==
There may be prior art related to power management techniques in computing devices, but specific examples are not provided in the abstract.

== Unanswered questions ==
=== How does the machine learning model adapt to different types of processes? ===
The abstract does not detail how the ML model is trained or updated to handle various types of processes and their impact on core usage.

=== What are the potential limitations of transitioning between power modes? ===
The abstract does not mention any potential drawbacks or challenges associated with switching between different core power modes.


==Original Abstract Submitted==

Apparatuses, systems, and techniques to determine a number of idle cores of a computing device using a machine learning (ML) model based on a set of processes executed by the computing device are described. One method determines a set of processes executed by the computing device and determines, using an ML model, a number of cores of the computing device to be powered down based at least on the set of processes. The method updates a first mode of the number of cores to a second mode in which the number of cores consumes less power than in the first mode.



[[Category:G06N5/04]]
[[Category:G06F1/3296]]
[[Category:G06N5/02]]