The Chunking Protocol
The process of splitting data (Part I of data security model)
The security of data is essential to any individual or organization. For this reason, we attempted to provide an unhackable model that ensures data protection for all renters. The first element of the model is the Vuuple Chunking Protocol.
When a renter uploads a file to the Vuuple network, it is divided into multiple pieces, and stored on the drives of multiple lenders. This procedure is called chunking, which includes scanning the file and splitting it into short elements. Each element is called a chunk and is a unit of redundancy detection. To establish the most efficient chunking and de-duplication method (the process of eliminating redundant chunks within the same storage location), we decided to implement a Content-Defined Chunking (CDC) protocol that utilizes a Rabin’s fingerprint algorithm. CDC creates boundaries based on the content of the files and then splits those files based on the boundaries created. Each chunk within a given boundary is assigned a fingerprint via the Rabin algorithm. When a file is updated and uploaded multiple times on the Vuuple network, only the portion of the file that was changed will be saved. All redundant data within the bounded window of byte stream will be eliminated. This allows the network to achieve quick uploading times and save an abundance of storage space.
Since all uploaded files are split amongst multiple lenders, even if a harmful actor were to access a Vuuple file, they would only have one piece to a larger puzzle. This individual would only have access to a small byte of a larger data piece and would not have any knowledge on the location of the other bytes. Furthermore, each file chunk is replicated in different locations, meaning if one portion of data were lost due to inadequate lending practices, the entire file could still be recalled.
