[OSS] Run Azure Cognitive Services in Containers

Maybe not a real OSS topic today … but as it is about freeing a service from the lock-in into a platform, I think it is still a fit. So did you know that it is possible to run Cognitive Services, as they are available in Azure, also ON-Premises, on Edge or in any other Cloud?!

Did you know that you can run those containers everywhere no matter of Container Host platform. So that you can even build an open-source based solution on Linux running docker … hosting your Cognitive Service?

And did you know that it is available for more than a year already?!

So if you missed that … keep on reading.

Container Support for Azure Cognitive Services

Back in November 2018 Microsoft announced the initial availability of Container Support for Azure Cognitive Services. If you want to know more have a look into the original posts:

• Bringing AI to the edge
• Getting started with Azure Cognitive Services in containers
• Running Cognitive Service containers

In the meantime those services have evolved and you will also find more than just a few blog posts around this.

The key benefits to run Cognitive Services outside of Azure:

• Consistency – develop, test and run the workloads on the same fundament, so you can assure consistent outcomes
• Low Latency – put your service close to where it is needed, so you do not have to consider network latency to cloud locations
• High Scalability – Scale Out your solution easily, no matter where … just spin up more containers on the platform of your choice
• Portable Architecture – you can move your service wherever you want and even run it locally if required
• Data Governance – keep data inside of required boundaries, make sure that critical data never leaves your premises.

So as you can see… there are some reasons to consider running those Services in Containers.

Available Containers

Right now Container Support for cognitive Services is just available for a subset of all existing services. So it is still worth having a look into Azure 🙂

Available Services (or parts of):

• Anomaly Detector
• Computer Vision
• Face
• Form Recognizer
• Language Understanding (LUIS)
• Speech Service API
• Text Analytics

More on this can be found in the official documentation.

Dieser Post ist auch verfügbar auf: Englisch