Object Storage
lastmod 2026-03-12created 2018-07-3010 edits
Object storage (also known as object-based storage) is a computer data storage architecture that manages data as objects, as opposed to other storage architectures like file systems which manage data as a file hierarchy, and block storage which manages data as blocks within sectors and tracks. Each object typically includes the data itself, a variable amount of metadata, and a globally unique identifier. Object storage can be implemented at multiple levels, including the device level (object-storage device), the system level, and the interface level. In each case, object storage seeks to enable capabilities not addressed by other storage architectures, like interfaces that can be directly programmable by the application, a namespace that can span multiple instances of physical hardware, and data-management functions like data replication and data distribution at object-level granularity.
Object-storage systems allow retention of massive amounts of unstructured data. Object storage is used for purposes such as storing photos on Facebook, songs on Spotify, or files in online collaboration services, such as Dropbox.
What is Object Storage?
Object Storage is as follows.
- Emerged to store and manage unstructured data reliably
- Object storage stores data by automatically creating, keeping, and managing metadata that contains specific information about each file
- Storage that manages objects by assigning each one an ID that users or servers can identify, regardless of the data's physical location
- Bundles the data, an identifiable ID, and metadata into a single object, stores it in a container, and manages everything at the same level
- With this management method, when a file is stored, the file and its metadata are created and managed together
- By giving structure to unstructured data through this management method, it makes it possible to store and manage far larger amounts of content more efficiently and intelligently
- Through this management method, even unstructured data can be accessed in new ways, such as gaining additional value through fast file search and analysis using metadata, or sharing files via link addresses
- A drawback is the Throughput problem
Every time a file is updated, you have to wait until all replicas are updated, so it is not recommended for work where data changes frequently
- Suitable for storing data such as backups, archived video files, and virtual machine image files
- Think of object storage as being like valet parking
With valet parking, you hand over your key for valet service before entering a restaurant, the attendant parks the car for you, and when you come out after eating and hand over your ticket, they bring the car back to the front, so you never need to know where the car was parked. In the same way, the user does not need to know where a file is stored; think of it as a structure where you only need to access the file by its ID
List of Object Storage Softwares
- Cloud Service