Back to results
US 12,762,307 B2Granted
Method and system to determine optimal rack space utilization
AssigneeDell Products L.P.
InventorsVinay Sawal, Udhaya Chandran Shanmugam, Ramya Ramachandran, Mahendaran Pagalpatti Vijayakumar, Ashok Kumar Janardhana
Life of this patent
Estimating the term…
Abstract
Described herein are methods and a system for to optimize utilization of rack space supporting components of an existing client site rack. Real time data as to the rack space is collected, along with an objective function as to expansion. Information as to an initial design configuration is retrieved. A reinforcement learning algorithm processes real time data, the objective function, and initial design configuration to determine a deployment recommendation.
Claims17
A computer-implementable method for optimization of utilization of rack space supporting components comprising:
- collecting, via one or more monitoring interfaces coupled to equipment racks, real time operational data associated with rack space of an existing rack and components installed in the rack, the real-time operational data including rack unit occupancy and infrastructure constraints including power, cooling, and cabling;
- receiving, from a customer information handling system through a network interface, an objective function specifying a desired expansion of computing expansion of computing functionality components for the existing rack;
- retrieving, from a configuration database, initial design configuration information of the existing rack, including reserved rack units and expansion constraints;
- processing, by a rack space optimization engine executing on one or more processors, the real time data, objection function, and initial design configuration information using a reinforcement learning algorithm implemented as a Markov decision process including an agent, policy, state data, and reward function, to determine an optimized rack deployment configuration for installation of one or more additional components within the rack;
- and
- automatically generating and transmitting a deployment recommendation identifying one or more rack unit locations within the existing rack for installation of the additional components to the customer information handling system.
3 more elements
A system comprising:
- a plurality of processing systems communicably coupled through a network, wherein the processing systems include non-transitory, computer-readable storage medium embodying computer program code interacting with a plurality of computer operations optimization of utilization of rack space supporting components comprising:
- collecting, via one or more monitoring interfaces coupled to equipment racks, real time operational data associated with rack space of an existing rack and components installed in the rack, the real-time operational data including rack unit occupancy and infrastructure constraints including power, cooling, and cabling;
- receiving, from a customer information handling system through a network interface, an objective function specifying a desired expansion of computing expansion of computing functionality components for the existing rack;
- retrieving, from a configuration database, initial design configuration information of the existing rack, including reserved rack units and expansion constraints;
- processing, by a rack space optimization engine executing on one or more processors, the real time data, objection function, and initial design configuration information using a reinforcement learning algorithm implemented as a Markov decision process including an agent, policy, state data, and reward function, to determine an optimized rack deployment configuration for installation of one or more additional components within the rack;
- and
- automatically generating and transmitting a deployment recommendation identifying one or more rack unit locations within the existing rack for installation of the additional components to the customer information handling system.
4 more elements
A non-transitory, computer-readable storage medium embodying computer program code for optimization of utilization of rack space supporting components, the computer program code comprising computer executable instructions configured for:
- collecting, via one or more monitoring interfaces coupled to equipment racks, real time operational data associated with rack space of an existing rack and components installed in the rack, the real-time operational data including rack unit occupancy and infrastructure constraints including power, cooling, and cabling;
- receiving, from a customer information handling system through a network interface, an objective function specifying a desired expansion of computing expansion of computing functionality components for the existing rack;
- retrieving, from a configuration database, initial design configuration information of the existing rack, including reserved rack units and expansion constraints;
- processing, by a rack space optimization engine executing on one or more processors, the real time data, objection function, and initial design configuration information using a reinforcement learning algorithm implemented as a Markov decision process including an agent, policy, state data, and reward function, to determine an optimized rack deployment configuration for installation of one or more additional components within the rack;
- and
- automatically generating and transmitting a deployment recommendation identifying one or more rack unit locations within the existing rack for installation of the additional components to the customer information handling system.
3 more elements
Dependencies are inferred from the claim text.