Abstract: A method includes generating a first data center graph including a plurality of host graphs, the host graphs representing resources of hosts of a data center, combining the first data center graph with an application signature graph to produce a second data center graph, the second data center graph representing the resource utilization of an application when running on the data center, predicting an energy consumption of the application when running on the data center by processing the second data center graph using a graph neural network (GNN), and scheduling the application on the hosts of the data center based on the predicted energy consumption of the application.

Abstract: Systems, methods, and devices for rightsizing data center resources to applications and consolidating server nodes including receiving a definition of goals for rightsizing and consolidating one or more server nodes of a server system and receiving a specification of a trigger for optimization. In the optimization, finding one or more sets of application runtime configurations for the server system and performing a trend analysis on a first set of the one or more application runtime configurations. In response to the trend analysis, predicting performance with consolidation of the one or more server nodes based at least in part on the trend analysis and confirming that the predicted performance matches the goals to verify that the first set satisfies the goals. In response to the performance matching the goals, implementing, by one or more processing resources, the first set of the application runtime configurations in the server system.

Abstract: In one example implementation, a computer-implemented method includes sending instructions from a scheduler directly to a plurality of compute resources to obtain sensor data from the plurality of compute resources. The sensor data is received directly from the compute resources. The scheduler develops a workload distribution plan based on information related to applications waiting to be executed and the sensor data received from the compute resources. The applications are assigned to the compute resources according to the workload distribution plan.

Abstract: A method includes generating a first data center graph including a plurality of host graphs, the host graphs representing resources of hosts of a data center, combining the first data center graph with an application signature graph to produce a second data center graph, the second data center graph representing the resource utilization of an application when running on the data center, predicting an energy consumption of the application when running on the data center by processing the second data center graph using a graph neural network (GNN), and scheduling the application on the hosts of the data center based on the predicted energy consumption of the application.