Decentralized storage refers to the concept of storing files online by splitting them into encrypted fragments and delegating these fragments to multiple nodes on a distributed network, e.g. a blockchain.
The concept of decentralized storage was pioneered in 2013-2015 by projects such as the (InterPlanetary File System) IPFS, Storj and Siacoin. The main idea behind the concept is to leverage the advantages associated with decentralized networks in order to improve privacy, security, censorship resistance, cost and availability of file storage systems.
Decentralized storage should not be confused with cloud storage, which is the storing of files remotely on servers operated by a single service provider like Google Drive or Amazon Drive.
A cloud storage provider’s data stores are centralized, meaning that they also act as a single point of failure: they can experience downtime or be attacked by hackers, leading to lower availability or even loss of stored data. They are also managed by a private entity, allowing it to restrict access to specific customers and censor data.
Decentralized storage systems, on the other hand, split user files into several encrypted fragments and delegate their storage to multiple participants of a geographically and organizationally distributed network, such as a blockchain. This is theorized to lead to a number of advantages, e.g.:
- The distribution of multiple redundant copies of file fragments leads to increased security and privacy: even if some nodes on the network become compromised, files can still be retrieved from the remaining nodes;
- Increased censorship resistance due to the fact that no single entity is in charge of the network;
- Increased uptime resulting from data download being available from multiple network nodes, most of which are online at any given moment.