PRE-TRAINED SOFTWARE CONTAINERS FOR DATACENTER ANALYSIS

Examples described relate to using pre-trained software containers for datacenter analysis. In an example, an onsite computing device may discover devices that are part of a datacenter. The onsite computing device may create a relationship amongst a group of devices, based on a criterion. The onsite computing device may download a pre-trained software container to each device of the group from a cloud system. A pre-trained software container is specific to each device the pre-trained software container is downloaded to, and is pre-trained to collect telemetry data from the respective device. A pre-trained software container may identify an anomaly related to the respective device from the telemetry data, and generate an inference in a pre-defined format. The onsite computing device may receive the inference from respective pre-trained software container and analyze the inference.