Amazon Deep Learning AMIs vs Microsoft Azure

• You can get several common packages including keras, pytorch and tensorflow setup within an environment ready to code on any AWS instance which saves time
• Great for virtual applications that helps communicate between various pieces of software
• Not need to worry about compatibility or any major aspects of setup e.g. GPU configuration

Read full review

• Azure Functions is the easiest serverless service to work with in my experience. Easy to ship Node.js functions without bundling dependencies.
• Proactive and responsive support. We've worked with most other cloud providers - Amazon, Google, Oracle. Amazon Web Services' customer support is a black hole and their documentation is worthless. Google and Oracle are better, but Azure's support is responsive and their docs are pretty good.
• Azure's Web UI is pretty easy to work with

Read full review

• Some aspects of the User Interface are quite confusing and activating packages can be a bit convoluted
• It can be a bit confusing to switch between frameworks for novice users

Read full review

• Could be easier to setup redundancy
• Feels as if there are too many options, bells, and whistles you could add.
• Microsoft Azure functions does not have a "Configure Services" method. Programming Azure functions has a slightly different programming model than does a "normal" API application. I would prefer that the startup process be identical.

Read full review

• It has made our Data Science/ Machine Learning Courses easier to manage/ need less human input therefore allowing us to increase the cohort size for this degree
• It has unified a lot of technologies reducing the load on our IT team

Read full review

• It has taken a few years of planning and growth (which continue), but we've been able to successfully budget and plan for ever increased capacity by starting first with some Virtual Machines and factoring in on-prem maintenance vs cloud constant improvements. VM (server, compute, networking) costs even out quickly over time and will show savings very soon.
• Because of Azure Application Gateways (which cost very little to set up and maintain), we were able to build a state of the art single-sign-on hub for all of our users and other enterprise platforms (mostly non-Microsoft and non Azure systems), reducing the amount of time to switch among these for workers and increasing overall productivity.
• With Azure Active Directory, we do not need to maintain separate identity systems and can more easily integrate our organizations credentials into authorization for access to other systems.

Read full review