Utilitarian is a multi utility AMR system built for the Cloud.

architecture image

Multi utility#

Utilitarian is built to manage any energy type, meter and communications protocol.


Utilitarian is a decentralized system that can be deployed across as many servers as needed to fit your use case.

By using a message broker workloads are split across asynchronous worker processes for time consuming like polling for meter data. Workers can be scaled independently for fit your load.

Receiving of push meter data is made in separate services that can all be individually scaled for the needed load.

Multi tenant#

Utilitarian is built with multi tenancy using a same schema approach. We organize meters and meter information around the concept of Organizations. A User can be part of several Organizations.


Several instances can be run of every service and they can be run in different environments, for example if you want to split load between different data centers or availability zones Utilitarian does not limit you.

Services can be load balanced using standard technologies.

The main points of failure are the Postgres database and the RabbitMQ message broker, but these risks can be mitigated using High Availability (HA) setups and standard database reliability techniques.


Depending on internal and regulatory requirements you can run Utilitarian with a range of security levels and settings so that you can find the best fit for your organization.


You can just run the services needed for your specific use case.


Utilitarian is designed to enable users to easily build their own integrations and subsystems if needed. It is possible to hook into the message streams from any application using the message payload documentation.

Utilitarian is built using open source technologies and libraries. We have also open sourced many of our own libraries that we use within Utilitarian.

Cloud Native#

Utilitarian is built for the cloud. We deliver all services via docker containers so you can decide how and where to run the application. You can run it on a single machine or deploy using Kubernetes in a hybrid cloud with multiple data centers.