# US 11,442,516 B1 · Data center control hierarchy for neural networks integration > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 11,442,516 B1 - Document ID: US11442516B1 - Status: Granted - Kind: B1 - Application number: 17/205,360 - Filed: 2021-03-18 - Published: 2022-09-22 - Granted: 2022-09-13 - Earliest claimed: 2021-03-18 - Assignees: Baidu USA LLC - Applicants: Baidu USA LLC - Inventors: Tianyi Gao - Web page: https://patentgrep.com/patent/US11442516B1 ## CPC - G06F 1/26: Power supply means, e.g. regulation thereof (for memories G11C) - G01K 1/026: {arrangements for monitoring a plurality of temperatures, e.g. by multiplexing} - G05B 13/027: {using neural networks only} - G06F 1/206: {comprising thermal management} - G06F 1/263: {Arrangements for using multiple switchable power supplies, e.g. battery and AC (G06F1/30 takes precedence)} - H05K 7/1498: {Resource management, Optimisation arrangements, e.g. configuration, identification, tracking, physical location (thermal management H05K7/20836)} - H05K 7/20836: {Thermal management, e.g. server temperature control} - G06N 3/02: Neural networks - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - G01K: MEASURING TEMPERATURE; MEASURING QUANTITY OF HEAT; THERMALLY-SENSITIVE ELEMENTS NOT OTHERWISE PROVIDED FOR (radiation pyrometry G01J5/00) - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) - H05K: PRINTED CIRCUITS; CASINGS OR CONSTRUCTIONAL DETAILS OF ELECTRIC APPARATUS; MANUFACTURE OF ASSEMBLAGES OF ELECTRICAL COMPONENTS - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS ## Abstract A data center system includes a load section having an array of electronic racks, a thermal management system, and a power flow optimizer. The power flow optimizer is configured to determine a load power requirement of the load section based on workload data of the electronic racks and thermal data of the thermal management system. The data center system further includes a resource section having a number of power sources to provide power to the load section. The resource section includes a resource controller to configure and select at least some of the power sources to provide power to the load section based on the load power requirement provided by the power flow optimizer. The power flow optimizer includes a power flow neural network (NN) model to predict, based on the thermal data and the load data, an amount of power that IT clusters and the thermal management system need. ## Claims ### Claim 1 A data center system, comprising: a load section having a plurality of electronic racks, a thermal management system, and a power flow optimizer, wherein each of the electronic racks includes a plurality of servers and each server containing one or more electronic devices, wherein the thermal management system is configured to provide cooling and/or heating to the electronic devices, and wherein the power flow optimizer is configured to determine a load power requirement of the load section based on thermal data of the thermal management system and load data of the electronic racks; and a resource section having a plurality of power sources to provide power to the load section, wherein the resource section includes a resource controller to configure and select at least some of the power sources to provide power to the load section based on the load power requirement provided by the power flow optimizer, wherein the power flow optimizer includes a power flow neural network (NN) model to predict, based on the thermal data and the load data, an amount of power that the electronic racks and the thermal management system need in order to satisfy a thermal requirement and data processing load requirement of the load section. ### Claim 2 (depends on 1) The data center system of claim 1, further comprising an intermediate section coupled between the resource section and the load section, wherein the intermediate section includes a power bus to distribute power from the resource section to the load section and other subsystems. ### Claim 3 (depends on 2) The data center system of claim 2, wherein the intermediate section further comprises: a subsystem load detector coupled to the other subsystems to determine a subsystem power requirement; and a central controller coupled to the power flow optimizer of the load section and the subsystem load detector to determine a total power requirement based on the subsystem power requirement and the load power requirement, wherein the total power requirement is utilized by the resource controller to configure and select at least some of the power sources. ### Claim 4 (depends on 3) The data center system of claim 3, wherein the load section further comprises: one or more temperature sensors disposed within each server to measure temperatures of the electronic devices; and a workload detector configured to determine a workload of each of the servers, wherein the power flow NN model is to infer the load power requirement based on the temperatures and the workload of each of the servers. ### Claim 5 (depends on 4) The data center system of claim 4, wherein the load section further comprises a power dispatch controller coupled to a power flow controller to proportionally distribute the power received from the resource section to the thermal management system and the servers based on the load power requirement received from the power flow controller. ### Claim 6 (depends on 5) The data center system of claim 5, wherein the load power requirement includes information regarding how to dispatch the power to the thermal management system and the servers, and wherein the power dispatch controller is configured to output total power needed by the load section to the central controller. ### Claim 7 (depends on 3) The data center system of claim 3, wherein the resource section comprises a resource optimizer to receive the total power requirement from the central controller to generate power source configuration information, wherein the resource controller is configured to configure the power sources based on the power source configuration information. ### Claim 8 (depends on 7) The data center system of claim 7, wherein the resource optimizer comprises an NN model to determine the power source configuration information based on total power needed. ### Claim 9 (depends on 7) The data center system of claim 7, wherein the power source configuration information comprises information specifying an amount of power to be provided by each of the plurality of power sources. ### Claim 10 (depends on 7) The data center system of claim 7, wherein the plurality of power sources comprises a utility power source, a photovoltaic (PV) power source, and a battery power source. ### Claim 11 (depends on 3) The data center system of claim 3, wherein the central controller comprises an NN model to infer total power needed based on the load power requirement and the subsystem power requirement. ### Claim 12 (depends on 2) The data center system of claim 2, wherein the data center system is a first data center subsystem of a plurality of data center subsystems, and wherein a power bus of the intermediate section is coupled to a power bus of an intermediate section of a second data center subsystem of the plurality of data center subsystems. ### Claim 13 (depends on 12) The data center system of claim 12, wherein a central controller is shared by the first data center subsystem and the second data center subsystem. ### Claim 14 (depends on 12) The data center system of claim 12, wherein the plurality of data center subsystems is a first data center cluster of a plurality of data center clusters. ### Claim 15 (depends on 14) The data center system of claim 14, wherein each of the data center clusters is controlled by a respective cluster controller, and wherein a central controller is shared by the plurality of data center clusters. ### Claim 16 A method for managing a data center system, the method comprising: determining, using a power flow optimizer, a load power requirement of a load section having a plurality of electronic racks, a thermal management system based on thermal data of the thermal management system and load data of the electronic racks, wherein each of the electronic racks includes a plurality of servers and each server containing one or more electronic devices, wherein the thermal management system is configured to provide cooling and/or heating to the electronic devices; and configuring and selecting, by a resource controller of a resource section having a plurality of power sources to provide power to the load section, at least some of the power sources to provide power to the load section based on the load power requirement provided by the power flow optimizer, wherein the power flow optimizer includes a power flow neural network (NN) model to predict, based on the thermal data and the load data, an amount of power that the electronic racks and the thermal management system need in order to satisfy a thermal requirement and data processing load requirement of the load section. ### Claim 17 (depends on 16) The method of claim 16, wherein the data center system further comprises an intermediate section coupled between the resource section and the load section, wherein the intermediate section includes a power bus to distribute power from the resource section to the load section and other subsystems. ### Claim 18 (depends on 17) The method of claim 17, further comprising: determining a subsystem power requirement using a subsystem load detector coupled to the other subsystems; and determining a total power requirement based on the subsystem power requirement and the load power requirement using a central controller coupled to the power flow optimizer of the load section and the subsystem load detector, wherein the total power requirement is utilized by the resource controller to configure and select at least some of the power sources. ### Claim 19 (depends on 18) The method of claim 18, further comprising: measuring temperatures of the electronic devices using one or more temperature sensors disposed within each server; and determining a workload of each of the servers using a workload detector, wherein the power flow NN model is to infer the load power requirement based on the temperatures and the workload of each of the servers. ### Claim 20 (depends on 19) The method of claim 19, further comprising proportionally distributing, using a power dispatch controller coupled to a power flow controller, the power received from the resource section to the thermal management system and the servers based on the load power requirement received from the power flow controller.