Object storage, also called object-based storage, is an approach to addressing and manipulating data storage as discrete units, called objects. Objects are kept inside a single repository, and are not nested as files inside a folder inside other folders.
To better understand object storage, it helps to understand file storage and block storage, which are two common ways data can be stored.
File storage stores data as a single piece of information inside a folder to help organize it with other data. This method is also called hierarchical storage, and it emulates the way we store paper files. When you need to access data, your computer system just needs to know the path to find it.
Block storage breaks a file into individual blocks of data and then stores the blocks as separate pieces of data. A storage system can do this without having a file-folder structure because each block of data has a unique address. This allows a storage system to spread the smaller blocks of data wherever in the storage system it finds most efficient. The storage system software pulls the necessary blocks back together to assemble the file whenever it is accessed.
Object storage keeps the blocks of data that make up a file together and adds all of its associated metadata to that file. But object storage also adds extended metadata to the file and eliminates the hierarchical structure used in file storage, placing everything into a flat address space, called a storage pool. The storage system software uses a unique identifier assigned to the object to find any particular object, like a video or a photo.
This metadata is key to the value of object storage versus other methods of storing data. Because metadata can be defined by a user, it allows for all manner of analytics on the use and function of data within the storage pool.
Advantages and disadvantages
Greater possibilities for data analytics, and the ability to store an object anywhere within a distributed data pool, makes object storage technology particularly enticing for companies that provide storage services. Object storage is used by Amazon Simple Storage Service, or Amazon S3 -- the largest provider of cloud storage -- as well as most of its competitors.
In addition to the ability to perform custom analytics on data use, object storage's flat address space allows it to be easily scaled by adding more storage to the pool.
But object storage does have its disadvantages. It is generally slower than a file or block storage system, like scale-out network attached storage (NAS). This makes object storage a poor choice for those applications that require rapid and frequent access to data, like financial systems.
Recently, vendors have been making the difference between file or block storage on NAS and object storage less of an issue with universal, or unified object, storage. A vendor uses a NAS-like software front end that presents an object storage pool as though it was NAS to the user. As the commodity hardware commonly used in object storage improves in speed, and more vendors adopt universal storage, object storage will find more uses beyond the cloud.
An object storage architecture addresses the problem of growing unstructured data and helps organizations reduce the complexity of their storage systems. Unlike traditional block- and file-based storage, object storage is highly scalable, and it can help IT organizations operationalize their storage costs by enabling them to easily add more storage as needed rather than buying in bulk upfront.