By Nick Roquefort-Villeneuve, Global Marketing Director – Amalto Technologies
Specialists have estimated that datacenter storage capacity in 2018 stood at 1,450 exabytes and is expected to grow to a massive 2,300 exabytes by 2021. Yes, that equates respectively to 1.45 and 2.3 zettabytes. I know, you’re wondering why you took Latin in school, instead of ancient Greek… To summarize in plain English: Data storage is huge and is getting huger by the minute. And the rate at which data storage increases cannot be without problem and urgent concerns: New locations, more powerful and scalable hardware, tighter security, significant energy consumption, access facilitation and an exponential need for bigger data, among other variables.
Centralized Data Storage
To store data in the cloud is considered nowadays as the traditional system of data storage. The process is quite straightforward:
- You, as a user, upload a data in the cloud
- The cloud service provider saves your data in one if its datacenters
- Then, you get to access your data from your computer at any time by sending a request to the datacenter
- The datacenter sends you the data
Why isn’t this process not that efficient in itself for the user and his or her environment?
- If you’re in the United States, it’s likely that your data will be stored somewhere abroad, where it’s preferably naturally cold, like Northern Scandinavia, resulting in delays (potentially seamless to you, but not to a server performance tool) in the delivery of your data
- Servers need to be located in temperature-controlled environments, constantly updated and maintained
- The technology that ensures performance must follow with the volume of data stored and therefore must be replaced regularly, which represents a substantial cost
- All of the above cannot be that good for the environment
- The server service provider has access to all stored data at any time, which represents a breach of the data owner’s privacy
- And a web browser, which is the gateway to the cloud, is also the tool of choice for a hacker, hence making cloud computing vulnerable. In other words, one attack from a single point of entry could likely compromise the entire storage system
Decentralized Data Storage
Decentralized storage infers the distribution of data across an extensive network of nodes, very much like in the distributed ledger technology feature of Blockchain. So, what does that mean exactly?
Imagine a peer-to-peer cloud storage technology. In other words, every time you’d elect to store data:
- The data would be split, encrypted and distributed to different nodes
- You’d be assigned a hash (private key of encryption)
- You’d use this hash each time you send a request to access your data
This is the data flow.
So, what about the storage servers? The storage is powered by Blockchain. And the Blockchain platform that enables decentralized storage provides a marketplace for hosts and users through the use of its native token. Thus, the participants (hosts) to a decentralized data storage network would, for example, rent extra space on their computer out to the Blockchain platform and in exchange for their service get paid in the Blockchain’s native digital token. Users, who need to store their data, would pay a storage fee in the system’s native token to a host. This is the financial model.
The business model is quite simple and what follows is an example among many others possible. The Blockchain’s native token is utilized to execute a file storage smart contract on the platform. A host commits to storing the data for an agreed upon period of time in an agreed upon space and is required to set aside some of the Blockchain’s native tokens as collateral. If the user requests access to his or her data and the host is offline, then the host will lose a portion of the token set aside as collateral. If the host loses the user’s data, then the former not only loses the entire collateral but also doesn’t get paid. The host solely gets paid upon completion of the contract.
The 3 S's: Security x 3
Decentralized data storage infers enhanced security and privacy. Files in a decentralized network are broken apart and spread across many nodes (also called the “sharding” process). As mentioned earlier, the files are encrypted with a private key (hash), which makes it impossible for other participating nodes to view a user’s file. Sharding also makes it impossible for anyone to access the entire content as the files are torn apart.