Way to run container groups without any cluster, billed per second. It offers neither autoscaling nor spreading across zones, and Azure Machine Learning treats it as a legacy v1 deployment target.
Also called ACI, ACI.
Read more: Microsoft Learn
In the Ultra Transcenders books
Each book explains Azure Container Instances in context, with comparison tables and the common traps.
Terms in this definition
- Container
Something that groups data. Blob Storage containers sit inside a storage account and hold blobs much as folders hold files; Cosmos DB containers hold items and set the scope for partitioning and throughput.
- Autoscaling
A classic compute option where you give a minimum and maximum worker count and Azure Databricks adjusts the number of workers within that range to match the load.
- Azure Machine Learning
Azure platform where you train, deploy, monitor and retrain models of your own, practising MLOps through tools such as pipelines, online endpoints, model monitoring and prompt flow.
- Model deployment
Inside a resource, each deployment is a named copy of a Foundry model with its own TPM quota and deployment type. Requests identify the model by this deployment name.
Related terms
- az container create
Azure CLI command that deploys a container group to Azure Container Instances. Its options cover the image, exposed ports, a VNet subnet, the restart policy, and registry access through credentials or
--acr-identity. - Container group
ACI's top-level unit, comparable to a Kubernetes pod: one or more containers placed on the same host, sharing ports, an IP, volumes and a lifecycle; if it holds several containers, it must run Linux.
- Microsoft.ContainerInstance
Azure Container Instances resources fall under this provider namespace, so holding Microsoft.Compute permissions gives no rights over them.
- Virtual nodes
Nodes in AKS that run on Azure Container Instances, so Linux pods get compute within seconds. Azure CNI networking is a prerequisite.