Object storage explaination required

Saahib

I am sorry if its already discussed but I am looking for common layman explanation for object storage services provided by Amazon and some others. What exactly it is and how it is different from regular hosting / backup servers ?

Traffic

Object Storage (also known as object-based storage) is a 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.

Source: http://en.wikipedia.org/wiki/Object_storage

deadbeef

@Saahib said:

Object storage: Sinkhole where you dump files which are then retrievable via an endpoint, usually an HTTP API.

Programmers love that kind of stuff, much easier to build scalable apps around an API than a filesystem.

MarkTurner

@Saahib - its a way of having an online data storage without having to worry about setting up, scaling or maintaining of the underlying platform.

For example, Delimiter's (Yomura) S3 platform lets you create a certain amount of hostnames in the package for example:

images.domain.com
downloads.domain.com

So you can segment your data into 'buckets'. Then set parameters to make buckets public, private and so on.

Lets say you were producing a new game, you wanted to make available an upgrade but you knew you would be hammered by 500,000 people trying to download it. So you could either scale out 20-30 servers to handle the peak traffic or just stick it on an object storage service eg downloads.domain.com then point customers to download it from:

http://downloads.domain.com/myupdate.zip

All the download traffic would be served from the object store. Its then down to the operator of the object store to scale sufficient heads to handle the traffic.

You pay storage + bandwidth which is a lot cheaper for dynamic workloads.