http://localhost:1313/chantico/technical/custom-resources/index.html This section describes the custom resource definitions (CRDs) that are defined by and used in the Chantico project. These CRDs describe the structure and behavior of the custom resources that are managed by the Chantico controllers. Currently, we have CRDs that are specifications for declaration of components that are part of a data center, how to connect to them for energy monitoring, how to obtain the energy usage metrics of such DC components, and how relations are formed between those devices. Hugo en-us Copyright 2025-2026 TNO. Chantico is licensed under the <a href="https://www.apache.org/licenses/LICENSE-2.0">Apache License, Version 2.0</a>. Published documentation is available under <a href="https://creativecommons.org/licenses/by/4.0/deed.en">CC-BY 4.0</a>. http://localhost:1313/chantico/technical/custom-resources/snmp-config/index.html Mon, 01 Jan 0001 00:00:00 +0000 http://localhost:1313/chantico/technical/custom-resources/snmp-config/index.html This document describes the MeasurementDevice custom resource definition (CRD) that is managed by the Chantico controller. It explains the design and data model, how the resource is structured and used by the controller to perform updates to the SNMP exporter configuration. Introduction Chantico is built on the foundation that most information useful for chantico is provided in custom resources. This allows for a declarative approach to managing the devices in a data center that are relevant for the energy usage domain. A core data source for chantico are devices that expose their metrics via the Simple Network Management Protocol (SNMP), which is a widely used protocol for monitoring and managing physical devices connected to a network. http://localhost:1313/chantico/technical/custom-resources/monitor/index.html Mon, 01 Jan 0001 00:00:00 +0000 http://localhost:1313/chantico/technical/custom-resources/monitor/index.html This document describes the role of the PhysicalMeasurement custom resource in the configuration of Chantico components. It explains the design and data model, particularly how it differs from other custom resources. Whereas the MeasurementDevice custom resource describes a type of device and how it could be monitored (with SNMP authentication and walks), the PhysicalMeasurement custom resource describes the actual endpoint that is to be monitored, i.e., where the device is located. This is important to make use of the defined SNMP exporter authentications and modules to point at a specific IP address to actually obtain metrics. This is what Prometheus does to scrape the metrics: provide the parameters to the SNMP exporter to obtain the metrics from a specific device, with a selected authentication setting and module. http://localhost:1313/chantico/technical/custom-resources/graph/index.html Mon, 01 Jan 0001 00:00:00 +0000 http://localhost:1313/chantico/technical/custom-resources/graph/index.html This document describes the data center resource graph managed and validated by the DataCenterResource controller. It explains the design, the data model, how the graph is constructed, validated and some of the fields that are used to describe the graph. Chantico is built on the foundation that physical and virtual components that make up a data center can be related to each other. Each component that can be monitored, aggregated or allocated energy-wise is represented as a DataCenterResource custom resource. The relationships between these resources define how energy flows from one resource to another. Together, the resources and their relationships form a graph. Once the graph is completely constructed, it allows for accounting all the energy usage in a data center. http://localhost:1313/chantico/technical/custom-resources/energy-accounting/index.html Mon, 01 Jan 0001 00:00:00 +0000 http://localhost:1313/chantico/technical/custom-resources/energy-accounting/index.html This document describes the energy accounting feature implemented in the DataCenterResource controller. It explains the design, the data model, how Prometheus recording rules are generated, and how to test the full pipeline end-to-end with a local kind cluster. Overview Data center resources (PDUs, bare metals, VMs, …) form a tree. Energy flows from the root nodes (PDUs whose power is measured by hardware) down to leaf nodes (servers, VMs, pods). Each edge in the tree carries a coefficient that describes what fraction of the parent’s energy is attributable to the child.