Abstract: A model is provided that produces predicted sensor data as a function of at least one input feature that includes an adjustable setting of a cooling infrastructure. The model is able to model a non-linear relationship between the predicted sensor data and the adjustable setting.

Abstract: A data center workload distribution management system includes a cooling cost engine to determine a cooling cost or cooling capacity for each of a plurality of zones of a data center and a workload distribution engine. The workload distribution engine is to identify the zone that has a lowest cooling cost and sufficient cooling capacity and also has sufficient processing capacity for a workload, determine a local cooling efficiency index for at least one location within the identified zone, and distribute the workload to the location having a local cooling efficiency index that indicates the highest cooling efficiency.

Abstract: In a method of determining cooling resource capacity allocations for resource units in a shared environment for optimized operations, cost parameters associated with operating the resource units in at least one of delivering and receiving cooling resources are identified, one or more constraints associated with operating the resource units are identified, in which at least one of the one or more constraints comprises meeting at least a total cooling demand of components in the shared environment, a cost function that correlates the plurality of identified cost parameters is developed, a minimized cost function is solved for through application of a Lagrange multiplier subject to the identified one or more constraints, and cooling resource capacity allocations for the resource units that result in optimized operations in the shared environment based upon the minimized cost function solution are determined.

Abstract: Managing computer resources. Workload state information associated with one or more servers in a data center is accessed by a computer, the workload state information comprises workload placement on one or more servers, cooling conditions of the one or more servers, and performance requirements associated with a data center. Cooling state information of cooling resources associated with the one or more servers are accessed by the computer, the cooling state information comprises cooling conditions of the cooling resources. The workload state information and the cooling state information are compared by the computer. Based on the comparing, the computer generates workload repositioning instructions that instruct repositioning of workload on the one or more servers to meet the performance requirements.

Abstract: A device for reducing temperature variation in a plenum includes at least one pipe and a plurality of sumps containing a fluid operable to vary between a liquid state and a gaseous state depending upon a temperature of the fluid. The plurality of sumps are positioned at various locations within the plenum and the at least one pipe is in fluid communication with the plurality of sumps. In addition, the fluid, in the gaseous state, is caused to move through the at least one pipe and condense, thereby reducing the temperature at the location of the sump where the fluid was vaporized and thereby reducing temperature variation in the plenum.

Abstract: A demand response for an energy-consuming facility is disclosed. A demand response is generated by estimating a likelihood of a coincident peak time period, modeling workloads to be scheduled in the energy-consuming facility, determining a workload schedule based on the likelihood of the coincident peak time period and a plurality of utility charging rates, and scheduling the workloads for execution in the energy-consuming facility according to the determined workload schedule.

Abstract: An air moving device includes an inlet for receiving airflow, an apparatus for cooling the received airflow, an apparatus for moving the airflow, an outlet for exhausting the cooled airflow, at least one louver positioned to direct the cooled airflow exhausted from the outlet, and a motor attached to the louver. The air moving device also includes a controller for controlling the motor to thereby vary a position of the movable louver to thereby vary a region of influence of the air moving device.

Abstract: Methods, apparatus, systems and articles of manufacture to control fluid flow in a data center are disclosed herein. An example method disclosed herein comprises sensing a condition of a fluid in a first contained aisle on a first side of data center equipment, the fluid in the first contained aisle substantially isolated from a second contained aisle on a second side of the data center equipment, and controlling a vent to alter the fluid flow between the first contained aisle and the second contained aisle based on the sensed condition.

Abstract: Embodiments are directed to virtual machine migration according to environmental data in a data center. One embodiment is a method that analyzes environmental data for a data center having plural servers and migrates virtual machines among the servers to increase cooling efficiency in the data center.

Abstract: Quantifying power usage for a service. An example method may include identifying a dependency model for the service, the dependency model based in part on infrastructure providing the service. The method may also include determining power usage for the service using the dependency model.

Abstract: According to an example, synthetic time series data generation may include receiving time series data for a plurality of users, and applying dimensionality reduction to transform the time series data from a high dimensional space n of the time series data to a low dimensional space m, where m<n. The transformed time series data may be used to estimate a time series data density in the low dimensional space m by generating a density function. The density function may be sampled for data, and the sampled data may be transformed back to the high dimensional space n to generate the synthetic time series data.

Abstract: According to an example, synthetic time series data generation may include receiving empirical meter data for a plurality of users, and using the empirical meter data to estimate parameters of a Markov chain. The Markov chain may be used to generate the synthetic time series data having statistical properties similar to the statistical properties of the empirical meter data.

Abstract: According to an example, power infrastructure sizing and workload management of an entity may include receiving power supply and information technology (IT) workload demand input parameter specifications for the entity, and using the power supply and IT workload demand input parameter specifications for a power infrastructure sizing and workload management model for the entity. The power infrastructure sizing and workload management model may be used to generate power supply and IT workload demand output parameter specifications for the entity.

Abstract: A ventilation system includes a ventilation tile. The ventilation tile has a substantially circular opening to allow air to flow through the ventilation tile and a collapsible damper operable to collapse and expand to alter the size of the substantially circular opening in the ventilation tile and thereby variably restrict air flow through the substantially circular opening.

Abstract: In a method for determining a state of an apparatus, detected temperatures are received from a plurality of sensors and are compared to at least one preset condition. The state of the apparatus is determined based upon the comparison.

Abstract: In a method for managing a facility that is to receive resources from a first resource supply, a supply of resources available from the first resource supply is predicted for a predetermined period of time. In addition, a demand for resources in the facility during the predetermined period of time is predicted. A capacity schedule for the facility is planned to meet a predefined operational goal, in which the plan of the capacity schedule uses as inputs, the predicted supply of resources available from the first resource supply and the predicted demand for resources in the facility during the predetermined period of time. Moreover, a determination as to whether the planned capacity schedule meets the predefined operational goal is made.

Abstract: In a method for managing an infrastructure, a utilization of a plurality of infrastructure components in performing an objective is simulated. In addition, a resource demand for the infrastructure associated with the simulated utilization of the infrastructure components is determined and a supply of resources by a plurality of available sources to meet the determined resource demand is simulated. Moreover, at least one metric associated with the determined resource demand and the simulated supply of resources by the plurality of available sources is determined.

Abstract: In an embodiment, a device characterization system includes a sensor to sense an attribute of a device, a processor, and an algorithm executable on the processor to collect time series data of the attribute from the sensor, detect edges in the data, identify clusters from the edges, label the clusters based on input from a supervisor, and estimate device characterization parameters from the clusters.

Abstract: In a method for generating a capacity schedule for a facility to meet a predefined operational goal, in which the facility is to receive resources from a first resource supply, a prediction of a supply of resources available from the first resource supply for a predetermined period of time is accessed. In addition, a prediction of a demand for resources in the facility during the predetermined period of time is accessed. The prediction of the supply of resources and the prediction of the demand for resources are inputted into a constrained optimization problem and the capacity schedule for the facility is generated by finding a solution to the constrained optimization problem.

Abstract: A service sustainability system and method. An example method may include identifying service-level factors for providing a service. The method may also include identifying dependencies between the service-level factors. The method may also include determining service-level resource consumption based on information corresponding to the service-level factors and dependencies between the service-level factors.