hw.host.power/energy versus hw.power/energy

[sebastien-rosset]

How does one determine whether a hardware component should report power utilization with hw.power or hw.host.power?

The spec states:

Host energy usage MUST be reported using the specific hw.host.energy and hw.host.power metrics only, instead of the generic hw.energy and hw.power described in the previous section, to prevent summing up overlapping values.

This means one need to decide whether a hardware component is:

1. A Physical system as opposed to a virtual system or a container. Examples: physical server, switch or disk array., OR
2. any type of hardware component

The definitions are quite vague. Examples are given for hw.host.energy, e.g., physical server, switch, disk array; presumably this can be extended to include IOT devices, printers, cooling system, APC units, PDU, routers, phone, scanner, projector, dishwasher, furnace, etc. I'm not so sure about sub-components and systems of systems.

Since a physical system can have multiple parts that each consume power and can be measured, should hw.power be used for the sub-components? For example, a physical server has power supply units (PSUs), CPUs, DIMMs, disks, GPU, PCI components, etc. Each of these consume energy and typically have sensors that can report power utilization. The power supply units can report the total energy consumed by the host, and each sub-component can have an instrument that reports the power utilization of that component. I see hw.gpu.power has specifically been defined for GPUs, but many other types of components consume power.

What about complex systems such as a factory, car, plane, house or industrial furnace? For example, what should a house smart meter report?

Some hardware systems consume all the energy they receive, e.g. a physical server. Some devices transfer some of the energy they receive. For example:

1. A smart PDU can report the input/output power, energy, voltage, current. The PDU itself consumes very little energy, most of the power is transferred to the connected devices. Suppose the PDU has 10 connected devices, each consuming 500 Watts. The PDU may consume 20 Watts, so overall the PDU "consumes" 5,020 Watts. How should the PDU report its power? If hw.host.power reports 5,020 Watts, and each connected device reports hw.host.power with 500 Watts, then in aggregate the power is double counted.
2. A network device that supports power over Ethernet. The device may consume 500W and some of that power is transferred over Ethernet to connected devices, which themselves may report their own power utilization.

[sebastien-rosset]

Opened #3141