In order to have a truly decentralized blockchain ecosystem we need to have decentralized storage otherwise we’re still relying on a centralized server with one single point of failure. The Equifax data breach and numerous smaller data breaches could have had different outcomes if even while using a centralized server, the individual users had control over their data.
The current way to use cloud storage allows you to store and instantly access your data on remote servers, but these servers are just hardware sitting in a server warehouse. This differs from decentralized storage that has nodes running on multiple computers in multiple locations.
The cloud storage market is estimated to reach $89 billion by 2022, quite a jump from $31 billion in 2017. Microsoft, Amazon, IBM, Salesforce.com and SAP are the top five cloud storage vendors with Google and Oracle right behind.
So even though using Google Drive is super convenient centralized storage, the hosts can monitor, censor or disclose data to third parties and your data could be lost, altered or scrambled.
Decentralized Storage vs. Federated Storage
The alternatives are decentralized storage or federated storage. Decentralized storage networks run open-source software so anyone can look into how they operate. These options could potentially employ millions of server devices instead of a few hundred or thousand. By increasing the amount of competition, this decreases the price.
Federated storage is the collection of autonomous storage resources governed by a common management system that provides rules about how data is stored, managed, and migrated throughout the storage network.
The benefit of federated storage is that you can diversify on centralized servers as a placeholder for decentralization. Most people upload their files on Google Drive, but what if something happens to the file? If you had it uploaded on 5 different platforms you would be able to prove the information was tampered with based on it not being tampered with on the other platforms. This is just a short term strategy as these companies could get absorbed by larger companies and then you are no longer decentralized.
Metanet is a unique project that is trying to power the internet on Bitcoin SV’s Blockchain. It’s a little different than the other decentralized storage projects in the space because it aims to have the internet become a sidechain to the Bitcoin SV blockchain.
Craig Wright and nChain aim to change the way the internet works for streaming, storage and copyrights. Although Craig Wright and the SV community are seen as controversial, some think they will take over this space. nChain has been granted patents so these applications so they cannot be used anywhere other than the Bitcoin SV chain. This is a different strategy in an open-source mindset community.
Craig has a Medium post elaborating the details but in the system they propose, a separate key is calculated for each file, and yet, the user does not need to fear losing the file nor any of his keys. When the key and associated Bitcoin address are used to calculate a file address in the Bitcoin blockchain, the file ownership remains pseudonymous. Users can also create firewalls and partitions. Using the method-42 process, they can encrypt each file separately, and share each file (in whole or part) and even sell access to files.
More importantly, unlike a drive-encryption system where all files are encrypted using a single key, the user has a separate key for all files. If a single key is shared and compromised, it does not endanger the security and privacy of the other files.
Essentially this requires building search engines as well to be able to find this data. Not just simple linked communications but everything embedded. It’s a lofty goal, but some feel this may be the best solution.
MaidSafe stands for Massive Array of Internet Disks – Secure Access For Everyone. Quite the acronym! These are the OGs in the data storage space. This project has been under development for more than a decade now and is the world’s first autonomous data network. The safe network has clients and farmers. Farmers take care of the data on the network by providing disk space, computing power, and uptime. Clients can use the network uploading data, using Dapps, or browsing.
The SAFE network aims to create the world’s first fully autonomous decentralized data network preventing the network from being blocked, controlled or turned off. Many are excited about MaidSafe, but as it has not launched yet so they are forced to look at alternative solutions short term.
Note: MaidSafeCoin is a proxy token that was released during MaidSafe’s crowd sale and will be swapped for Safecoin on a 1:1 basis when Safecoin is released. MaidSafeCoin is a token that is listed on the bitcoin blockchain and can be purchased on a number of exchanges.
Clients use Safecoin for uploading data on the network which is broken, encrypted and distributed on the peer-to-peer network. Upon request, proof of resource consensus will be used to retrieve your data and also pay Safecoins to the farmer who found it first for you.
FILECOIN & IPFS
Filecoin has their own blockchain and utilizes mining. They are working on creating a free marketplace for storage of data and rent of extra disk space.
Filecoin is built upon some very complex and new tech pieces that haven’t been thoroughly tested yet as the project has not been launched. You can apply on their website to become an early miner when the blockchain launches and buy futures on various exchanges.
IPFS is the backbone of the Filecoin system. Currently the internet works off of location based addressing, these URLs are pointed to certain servers around the world. IPFS serves information based on what the information is as opposed to where it is located. Their routing algorithms allow you to choose where you get your content from and you can set your privacy of the peers / nodes you trust.
Swarm is part of the Ethereum stack which is made up from Ethereum, Swarm, and Whisper which will fit together to make up a decentralized web. Ethereum provides the computation power, Swarm the storage layer and Whisper a messaging layer.
Swarm runs on top of the existing smart contract infrastructure, but uses an additional proof of custody operation to scan and repair data over time when needed. Rather than use a new protocol such as IPFS, Swarm offers federated HTTPS endpoints to allow users to make calls to a familiar API interface and upload and download files. Swarm is still in its Proof of Concept stages and is not fully operational. You can try out their command line utilities in a simulated environment here.
Sia actually predates much of the buzz in this market. It was launched in early 2013 out of an MIT Hack event, and uses a Proof-of-Storage algorithm and smart contracts for the storage rental and Proof-of-Work for validating transactions and produce mining hardware for mining Siacoins.
In the Sia ecosystem, storage contracts are used which require the provider to stake a fixed amount of tokens in exchange for the privilege of storing files. While they store the file, the providers must consistently upload Proofs of Storage to the blockchain to verify their participation. Basically you have to prove that on your 250 GB computer you kept the 30 GB available for storage that you said you would. You’re basically being a landlord for storage and if you fulfill your data lease you are rewarded and receive the fees from the end user. If the host loses the files, they won’t get paid plus their collateral is gone too.
Sia recently unveiled a new form of ASIC miners designed specifically for this task, which has resulted in some criticism as to the actual decentralization of the project since it now requires specific hardware to join.
Anyone can rent out extra space on their PC to the Sia decentralized network and get paid in Siacoin. Similarly, any user who wants to use Sia storage needs to pay Siacoins to a host.
Storing 1TB of files on Sia costs about $2 per month, compared to $23 on Amazon S3. I love the free market, you get a better price, control your data and get the peace of mind that your info isn’t being sold to third parties. Issue lies with your files being stored in the cloud are not searchable.
The Storj cloud platform lets users rent storage from their peers on the network. All network transactions are conducted in Storj’s crypto asset Storj, which is a token launched on the Ethereum blockchain in 2014.
Storj’s technology is different, it revolves around file sharing similar to torrents. It separates parts of the files to users in the network, when a user wants the file, they request it and Storj uses a distributed hash table to locate all the shards and piece them together. The files are encrypted before sharing and the user uploading it has their own private key to confirm ownership.
Although Storj also allows users to rent storage from their peers on the network their implementation is different. Storj doesn’t create a free market space for lending your extra space because the prices are already fixed by the for-profit entity company Storj Labs. This is a more centralized model that competes more with Dropbox or Google Drive. The other interesting aspect to Storj is you can buy their services in BTC making others question the necessity for a Storj token. Their tokens are ERC20 tokens so the incentive to mine is being phased out with the changes to Ethereum. With no collateral involved the host is not incentivized to continue hosting. Storj is also a closed silo where your files stored in the cloud are not searchable.
Decentralized storage still has a way to go before it is sustainable and easy enough for mainstream users to participate.