There are many advantages to using Microsoft Azure. One big reason people look at Azure is the ability to separate storage from compute. This separation of storage and compute allows for cost reduction. If you don’t need to use a service, why should you have to have it turned on? And more importantly, why would you want to pay for it if it is not being used.
Storage and Compute in Azure
Storage is just that, storage. If you store data (and we all do), you need to pay for it the whole time. But if you can separate that from compute, you can turn on compute resources when you need them to interact with that data or scale up or down on those resources. Compute is the most expensive part of the cloud. The more you can reduce compute, the more you can reduce your overall cost.
Scaling Compute AND Storage
But cost is not the only reason to separate compute and storage. Oftentimes, storage and compute needs do not scale the same way. With Azure, you can increase your storage and the amount of data that you need without having to increase your compute resources. If you were looking at supporting more data inside a VM or database, you’d typically have to add more compute to support more storage. In contrast, with Azure and the structures that support it, you can begin to separate those needs and tune and manage accordingly. You can also increase your compute without increasing storage.
Check out Azure Data Week coming in October 2018.
Additionally, multiple compute resources can leverage the same storage assets. You can store one version of your data and using the Azure storage structure (such as Azure Blob Storage, file storage or Azure Data Lake), users can interact with that data with other resources. For example, if you have data stored in Azure Data Lake, you can turn on Spark ML to do some machine learning with Spark against that data set, and at the same time, turn on Interactive Hive, allowing users to run reports using that same data set.
Of course, the cost reduction is a key factor to start using Azure, but even more, it’s about enhancing your ability to manage the overall experience for your team, your data and your usage.