what choices do I have for storage within Azure?

Question

what choices do I have for storage within Azure?

Dion, Stephen 0

I'm a leader of a new data and analytics CoE. We are contracted with MCA Connect to implement their Inspire platform to have a common data dictionary and reporting environment. To get started with building the environment to support the Inspire platform, I need to understand the options available to me for storage in Azure since we need to move a considerable amount of legacy AS400 data off the IBM box and into the Azure cloud. Please focus on cost and effort to deploy as critical needs.

0 comments

1 answer

Your answer

Answer 1

For the lowest deployment effort, Azure Blob Storage would be the simplest and least expensive option for landing large volumes of AS400 data. It provides highly durable object storage at a very low cost, scales automatically, and integrates well with Azure Data Factory, Synapse, Microsoft Fabric, and third-party analytics tools. However, because it stores files rather than relational tables, additional processing would likely be needed before the data can be properly queried.

Azure Data Lake Storage Gen2 (ADLS Gen2) offers balance of cost and long-term analytics capability. It is built on Blob Storage, it adds a hierarchical namespace that improves organization and performance for analytics workloads and maintains relatively low storage costs. It is suited as the central repository for historical AS400 data and is a common choice for modern Azure data platforms.

If the data needs to be queried immediately using SQL with minimal transformation, consider Azure SQL Database or Azure SQL Managed Instance, which simplify reporting and application integration. These services require less effort than managing SQL Server on virtual machines but have significantly higher storage and compute costs than Blob Storage or ADLS Gen2. This generally makes them better suited for curated or operational datasets rather than large raw archives.

If you are looking to minimize upfront implementation efforts, Microsoft Fabric OneLake is another option because it provides a unified storage layer tightly integrated with reporting, data engineering, and analytics. It can accelerate deployment but note that costs are generally higher than using native Azure storage services alone.

If the above response helps answer your question, remember to "Accept Answer" so that others in the community facing similar issues can easily find the solution. Your contribution is highly appreciated.

hth

Marcin