CAStor™ Content File Server

Block-based vs. Object-based File Storage

Traditional file systems and NAS use block storage
The file system metaphor and folder/sub-folder interface has a long history and is firmly entrenched with end users and in applications. While the file folder metaphor has proven the test of time, the underlying storage infrastructure has not and is struggling under its own weight. Traditional file systems and NAS technologies built on block storage infrastructure are facing significant challenges for managing the massive volume of files and capacity required today.

In the block storage method, files are broken down and stored in pieces or 'blocks' with multiple 'blocks' per file. Each address of each block must be managed. Not only does this method add complexity but it constrains the number of files that can be stored and has limitations on the overall storage capacity.

Imagine a file system with 4 million files and each of those being broken into 5 blocks each meaning the file system must manage 20 million block addresses. As requirements grow to the 100s of millions to billions of files it's easy to see how these file systems are challenged to scale.

CAStor CFS: Innovation in File Serving
Unlike other file systems that must manage multiple individual blocks for every file, CAStor CFS stores a file contiguously as a single, complete object. It only needs to manage a single address per file allowing it to easily manage file stores at any scale.

CAStor CFS is built on top of a solid foundation in the CAStor content storage clusters and enables massively scalable storage capacity of 1000s of Terabytes. CAStor clusters provide a vast, flat address space that allows the storage of millions to billions of files. Caringo's content file server delivers a new approach to enable companies to store files more efficiently, more economically, and with less effort.

Download CAStor