Getting the most out of cloud requires choosing the right resources—both the right kind and the right size. When it comes to storage, one of the choices to consider is object storage. That’s different from the file storage we’re more used to. Here’s a brief look at object storage and why you might decide it’s the appropriate storage solution for your data.

Object Storage Defined

In conventional file storage, the system manages blocks and files in a hierarchical file system. Object storage manages objects in a flat address space. The stored object has two parts, the data and metadata. The metadata associated with files only specifies the common file attributes, such as name, whereas objects can have custom metadata.

Objects are never directly updated; when users make a change to the object, the system stores a new version of the object. Even if there’s only a single version of an object, there may be multiple copies in order to provide redundancy.

The Right Time for Objects

Accessing block or file-based storage doesn’t work well when access is geographically distributed. There may be issues with latency, and the ports needed for this access are usually not publicly exposed. Objects are accessed by get, put, and other commands sent via a RESTful API.

Because access to objects requires using an API, using object storage can be more complicated than using file-based storage. Some gateways have been developed to make using object storage easier.

There are several use cases that are clearly inappropriate for object storage, such as traditional databases that record transactional data and have frequent changes. Shared files are also difficult to store as objects because there are no locks and there are multiple versions of objects. Use cases that are well suited to object storage include unstructured data and data that doesn’t change often. Specific scenarios include providing distributed access to files and streaming audio or video content.

Object Storage in Azure

Object storage in Microsoft Azure is known as blob storage. There are tiered storage options available and blobs are cost-effective for large volumes of data. The storage is scalable and you pay only for what you use. Azure offers multiple kinds of blobs, known as block, page, or append blobs, which are optimized for different operations. Blobs can be replicated to different geographic regions, enabling use across a global organization.

Azure provides multiple means to access blobs, including the Azure Storage REST API and access libraries for languages including .NET, Java, Python, PHP, and Ruby. Azure also provides multiple means of getting started with blobs; tools such as the Azure Storage Data Movement library and Azure Data Factory help you load your blobs with data.

Do you have other questions about object storage in Microsoft Azure? Prescient Solutions is a Microsoft Partner and provides Azure IT consulting and managed services in the Chicago and Schaumburg areas. Our certified experts work with your team to understand your business and provide the routine IT support it needs. Contact us to learn more about Microsoft Azure and how Prescient Solutions can help you make the most of your cloud environments.