In the first two parts of this topic, I discussed how data is managed in SQL Azure and what the cost would be to an organization. In this installment, I wanted to propose some data solutions that might be a fit for a SQL Azure database.
Here is some quick history that is shaping my thoughts on this. While at Magenic, my focus is on business intelligence solutions on the Microsoft SQL Server product stack. Prior to returning to Magenic, I worked at XATA Corporation as the data architect for a hosted, SaaS solution serving the trucking industry. So, out of this background, I base my first suggestion. In SaaS (Software as a Service), customers often use tools provided by the vendor for specific tasks which tasks generate data. Most SaaS solutions offer some form of reporting or query based analysis for the data that is collected. However, some users require a greater level of interaction with the data. The issue is that the data is stored in the SaaS vendor’s multi-tenant data store which is usually not conducive to having ad hoc queries run against it. This has led to one of the most common solutions to this problem – export the data to customer. The problem is that the customer must now host the data on premise and is often responsible for merging new data as it comes from the vendor. In this solution, SQL Azure could act as the “go-between” between the multi-tenant data store and the customer. This will allow the vendor to provide a value-added BI service that the customer can leverage in a number of ways including reports, Excel, and even Access. The vendor can keep the data synchronized and the customer can leverage the data as needed.
Beyond the SaaS BI ad hoc solution, SQL Azure can be used to support development of solutions that require a shared data store without putting the data on each desktop. In keeping with the concept of the cloud being “anywhere”, SQL Azure can also be used to support distributed solutions that require a SQL data store to function.
Obviously, there are still issues with using SQL Azure as a primary, production data store due to the lower SLAs from Microsoft. However, it is not too early to investigate creative ways that will help your business leverage a relational database in the cloud. I would encourage you to check out the free trial options from Microsoft to experiment with the platform.