Windows Azure and SQL Azure Use Cases

The key to effectively leveraging “Cloud Computing” or more accurately, Distributed Computing architectures like Windows and SQL Azure is to implement them where they make the most sense. This is actually good advice for any computing paradigm, but some folks believe that a particular tool should be used in all circumstances. Microsoft does not recommend that you take all of the computing resources you have on-premise and move them to a distributed architecture.

A Use Case is defined as “when or where you use a technology” and a Pattern/Practice is defined as “how you implement a technology”.  In this series of posts, I’ll cover the use cases, and I’ll also give you resources to leverage to implement them. If you are looking for a way to learn “cloud” or Distributed Computing, check out the Azure Learning Plan here.

Windows Azure Use-Cases


Elastic Scale – Bursting workloads up and down in use patterns 

Agility – The ability to quickly develop and deploy an application 

New Development- Code option for new applications

Web-Centric Applications – Applications that are developed for a web paradigm

Hybrid Applications and Data – Applications and data that need to be both on-premise and in a distributed environment 

High-Performance Computing – Apllications that require multiple processing nodes, such as scientific, research or financial data (also known as Technical Computing) 

Infrastructure Limits – Inability or unwillingness to add more physical computers to the environment 

Fast Acquisitions – The ability to quickly migrate a newly acquired business to the current computing environment

SQL Azure Use-Cases  

Application Data Store – Uses SQL Azure to provide a central location for department-level or other collaborative applications 

Shared Data Hub  – Provides a common data store for multiple organizations to handle data between them 

Web-Based Applications – Using SQL Azure as a data store type for web applications 

Shared Storage Application – Using SQL Azure as a hybrid data store or as part of a HA/DR plan.

Comments (6)

  1. Simon Galbraith says:

    In general this is a great blog and I'm a fan but this post is mostly missing. All the posts seem to have 5 stars including this one…


  2. BuckWoody says:

    Hello Simon – this is just a "Table of Contents" Post. And Whoever reads the posts makes the stars. You can star it however you like – I don't control that.

  3. SQLCowboyUP says:

    I think of Azure as a solution in search of a problem.  It reminds me of Cheez Whiz: it's cheese in a very related-by-marriage-thrice-removed sort of way to cheese in the same way that Azure is related to the day to day operations of our data infrastructure. Security, Scalability, performance–I just don't see how this could be a better solution. If the other issues could be eliminated then comes the cost — which probably eliminates it as a solution from the business-case perspective.  

    Of course, I could be wrong (as I have been so many other times)–do tell.

  4. Buck Woody says:

    SQL Cowboy – Actually, I disagree. The cases shown here work for our clients every day. Of course, if you're not faced with any of these issues, and if what you're doing works well, as I mentioned here, stick with that.

    Thanks for reading!

  5. geodetics says:

    I listened to your Deep Fried Bytes podcast – great show. Brought me here today – I think you made an excellent case for where we are going.

  6. BuckWoody says:

    geodetics – thanks! I think this will be a great new way to work. I really do believe that distributed computing is the future.