A scalable protocol agnostic backup server

In the modern digital age, there is a deluge of data The data and associated information affects nearly every aspect of human life. Ensuring availability and durability of data is critical Availability of data is addressed by various file system technologies that invariably provide a distributed and redundant storage infrastructure. Durability of data is a different dimension and is ensured by providing backups of critical data While there are many backup solutions for personal data, ranging from using devices like pen-drive, external USB drives, tape drives to cloud based solutions like Dropbox and equivalent services, our interests is backup solutions for enterprise data There are two dimensions of variability for backup servers. The protocol that backup clients use to communicate with backup servers and the file systems where data resides that need to be backed up. Our intent is to develop a backup server that abstracts away both of these variations to provide a unified framework into which specific implementations can be plugged in Applying good software engineering principles of abstraction, separation of concerns and loose coupling, we present a design for such a protocol agnostic server that is also scalable. We discuss a specific implementation of this server. Though the design is for specific application of backup servers for different environments, the approach is generic and is useful in many other contexts and applications.