18045884. KINETIC POWER CAPPING USING FUZZY LOGIC-BASED DYNAMIC SYSTEM PRIORITIZATION simplified abstract (Dell Products L.P.)
Contents
KINETIC POWER CAPPING USING FUZZY LOGIC-BASED DYNAMIC SYSTEM PRIORITIZATION
Organization Name
Inventor(s)
Rishi Mukherjee of Bangalore (IN)
Shivendra Katiyar of Bangalore (IN)
Lori Lynn Matthews of Austin TX (US)
Elie Antoun Jreij of Pflugerville TX (US)
KINETIC POWER CAPPING USING FUZZY LOGIC-BASED DYNAMIC SYSTEM PRIORITIZATION - A simplified explanation of the abstract
This abstract first appeared for US patent application 18045884 titled 'KINETIC POWER CAPPING USING FUZZY LOGIC-BASED DYNAMIC SYSTEM PRIORITIZATION
Simplified Explanation
The technology described in this patent application involves dynamically adjusting power capping for servers within a subset of servers in a server system. The method includes generating weighted average values for current workload priority, current performance efficiency, and predicted future power usage for a subset of servers, ranking the subset of servers compared to another subset based on these values, and applying a power cap to the subset of servers based on the ranking.
- Explanation of the patent/innovation:
- Dynamic adjustment of power capping for servers within a subset of servers in a server system. - Generating weighted average values for current workload priority, current performance efficiency, and predicted future power usage for a subset of servers. - Ranking the subset of servers compared to another subset based on these values. - Applying a power cap to the subset of servers based on the ranking.
- Potential applications of this technology:
- Data centers - Cloud computing environments - Server farms
- Problems solved by this technology:
- Efficient power management for servers - Optimization of server performance - Reduction of energy consumption
- Benefits of this technology:
- Improved server efficiency - Cost savings on energy consumption - Enhanced performance of server systems
- Potential commercial applications of this technology:
- Energy management software for data centers - Server optimization tools for cloud computing providers - Power capping solutions for server farms
- Possible prior art:
- Power capping technologies in data centers - Server performance optimization tools
- Unanswered questions:
- How does this technology impact overall energy efficiency in server systems?
This technology can significantly improve energy efficiency by dynamically adjusting power capping based on workload priorities and performance efficiency, leading to reduced energy consumption and cost savings.
- What are the potential challenges in implementing this technology in large-scale server systems?
Some challenges in implementing this technology in large-scale server systems may include compatibility issues with existing infrastructure, scalability concerns, and the need for robust monitoring and management systems to handle dynamic power adjustments effectively.
Original Abstract Submitted
Technology described herein relates to dynamic adjustment of power capping for one or more servers of a subset of servers. A method can comprise generating, by a system operatively coupled to a processor, for a first server subset of a server system, weighted average values comprising a first weighted average value of current workload priority at the first server subset, a second weighted average value of current performance efficiency of the first server subset, and a third weighted average value of predicted future power usage for the first server subset, ranking, by the system, the first server subset as compared to a second server subset of the server system that does not overlap servers with the first server subset, wherein the ranking is based on at least one of the weighted average values, and applying, by the system, a power cap to the first server subset based on the ranking.
