We frequently get asked how Bluzelle is different from a number of data storage solutions, including but not limited to Filecoin, Storj, Sia, and IPFS. The comparison is not ideal because, though all of these services can be used to store information, they answer different problems. For example, bitcoin itself can be used to store data, but using it as the data storage for a database would result in a comically slow and expensive application. For a more realistic example, consider IPFS: it can in theory be used to store data, but it provides no guarantee about how long the data will be available unless you’re hosting it yourself. IPFS is more about addressing and transporting data. It can be thought of as a transport protocol somewhere between bittorrent and HTTP.
To answer the question, we’ll talk about what Bluzelle aims to accomplish compared to these other services and data storage services in general. We avoid directly discussing those other architectures in detail because the details are nuanced and often in flux. As the team behind Bluzelle, we’re much better positioned to discuss the advantage of our own architecture than we are to describe the shortcomings of others.
Database
Bluzelle is analogous to a database, not to a filesystem or a hard drive. Though client data is broken down, encrypted, replicated and so on, we abstract the details of all these processes away. A client of Bluzelle does not have to think about how many replicas of the data there should be or where they should be located. It does not have to think about what how to arrange the data into files or how to encrypt those files. It does not have to make sure that the physical devices storing the data do so faithfully or recover from replicas when they inevitably fail; the swarm handles these tasks. Bluzelle exposes a simple key-value store and abstracts away all these details.
Trustless
Bluzelle is a trustless architecture. The protocol ensures that, from the perspective of a client, their data is secured against both careless and malicious handling by third parties. This is in contrast to a traditional cloud storage provider (while the reputation of such a business is too valuable for them to have motive to actively interfere with client data, they could still refuse service to a client at any time, especially if that client becomes unpopular) or a “storage market” style system, where clients and miner senter into individual contracts to store data (the poor behaviour of a single miner is not outside the realm of possibility). In Bluzelle, the data is secured against all but a systematic and expensive attack on the database itself.
Scalable
Bluzelle is designed from the ground up for scalability. Traditional blockchain models (such as Bitcoin and Ethereum) rely on every node processing every transaction in order to validate new blocks. The consequence is that adding more nodes to the network does not increase it’s computing power — in fact a larger network is slower because it suffers a higher overhead. Bluzelle’s novel swarm-of-swarms architecture enables it to effortlessly scale up to handle arbitrarily large amounts of data, because each shard of data is replicated only within a single swarm.
These are the draw of Bluzelle: a truly scalable database built on the wasted storage media of the world, with the security to make it usable for serious projects and the scalability to grow beyond a curiosity.