Introduction to Folder Protocol
Introduction
The Cloud Computing has revolutionized the IT sector globally by enabling the businesses and developers to move away from on-premise IT infrastructure to an on-demand infrastructure. Some of the driving factors for this radical change are reduced costs, seemingly infinite development resources as an on-demand pricing model, scalability of IT networks and storage, ease of maintenance, etc. Though these cloud computing services are designed under fundamental principles of distributed computing, they centralize the trust and fail to offer any kind of decentralization and better security and privacy for the organizations.
Recently as blockchain technologies are gaining popularity, there are a lot of startups proposed blockchain based infrastructure projects to decentralize computing and storage. The projects like Swarm, StorJ, IPFS/Filecoin, etc. are working on developing decentralized storage infrastructure to decentralized the trust. Despite growing popularity of these networks, they suffer from latency and scalability issues for data intensive applications and services on blockchain. Their design and architectural limitations don’t allow them to be a good fit solution for enterprise scale storage options in decentralized digital economy.
Motivation
At Folderlabs, we are creating the innovation to achieve decentralized storage for enterprises and developers who are planning to create data intensive blockchain based applications or decentralized applications (Dapps). We are focusing on a business-to-business or business-to-developer Dapps ecosystems related use-cases where the organizations and developers need to handle petabytes (PB) level data for their Dapps and blockchain infrastructural needs.
Overview of Data Layer of Blockchain Architecture
Every blockchain is different from each other in some design/architectural details. But all blockchain networks have some core architectural components in common. In general, a blockchain architecture has at least five important components (also called layers): data layer, network layer, consensus layer, incentive layer, application layer.
Data layer
The data layer of the blockchain contains the data structure and the related details that created the blockchain network.
Block is the fundamental building block of blockchain. Each block contains records of transactions generated during a given time period. Each current block also records the hash (cryptographic signature) of the previous block. The blocks are linked linearly in chronological order (increasing in time). This chain structure provides immutability on the blockchain network.
A block is composed of block header and block body. The exact content of the block differs from every blockchain. For example, in the case of Bitcoin, the block header includes fields of block version, merkle tree root hash, timestamp, nBits, nonce and previous block hash. Block version indicates the validation rules nodes to follow. Merkle root hash is computed on all the transactions in the block. Timestamp records current time since January 1, 1970. NBits indicates the validation of a block hash. Nonce is adjusted by the miner to validate a block. Previous block hash is a 256-bit hash value that points to the previous block.
Merkle Tree is created by pairing each transaction with another transaction and creating a hash on them. The computed hashes are also paired with one other hash and hashed again. Such process repeats until only one hash remains, i.e., the merkle root.
Overview of StorJ, SiaCoin, and IPFS/Filecoin
StorJ
StorJ is a p2p cloud storage network built on Ethereum with Proof of Stake protocol. Customers send encrypted data to the network. File block encryption protects the confidentiality of the files. The storage provider provides proof of data integrity, and when the verification is passed, the customer pays the storage provider a storage fee.
SiaCoin
SiaCoin is a decentralized storage platform with Proof of Work protocol that uses smart contract technology to create a document intelligence contract between the storage provider (miner) and the storage user (customer). The contract requires that the storage provider provides the customer with a data storage certificate during the certification window. If the proof is valid, the smart contract will automatically pay the customer’s provider the negotiated storage fee.
IPFS/Filecoin
Filecoin is an incentive model built on top of IPFS (Inter Planetary File System) Network. IPFS is a decentralized distributed storage network where the customers and the storage providers (miners) request services and submit orders to the storage and retrieval markets. The miner provides a service to view matching quotes to initiate a transaction. The protocol guarantees the data integrity by copying proofs and space-time certificates. The Filecoin protocol writes the order book, transactions, and integrity challenge response records to the blockchain.
IPFS addresses a file by a content identifier (CID), instead of location-based, which is a cryptographic hash of the content of the file. IPFS uses distributed hash table (DHT) to support the process of routing and retrieving content from the nodes in the IPFS network, while using a Merkle DAG data structure to describe the file as a whole and reconstruct any file from its chunks. IPFS uses InterPlanetary Name Space (IPNS) as a content creation and update system, Interplanetary Linked Data (IPLD) for data management and Bitswap to send and receive notification between nodes.
Challenges Specific to Scalable Decentralized Storage
There are five fundamental challenges facing decentralizing the storage. They are listed below:
- Data Integrity — the ability to retain our data intact
- Data Availability — the ability to access our data at anytime
- Data Recovery — the ability to prevent data loss
- Privacy — to keep others from accessing our data
- Performance — speed and stability required for enterprises to create and access the data at petabytes scale
Folder Protocol — Bringing Enterprise Scale Data Storage to Decentralized Blockchains
Given the above challenges of decentralized storage, Folderlabs is focusing on creating a blockchain protocol for the large data intensive needs of a business user or the developer communities to bring enterprise level scalability and performance to the decentralized storage. When our algorithm is completed, Folder Network will offer a technical maturity and inherent capacity to support high-throughput, low-latency decentralized storage and retrieval markets for the data intensive blockchain based applications.
Folder Protocol Use Cases
For the decentralized storage market to be more than theoretical or experimental project, it must support the mainstream and enterprise business use cases. This means being compatible and competitive in the current Cloud storage options. Folder Network will support seamless integrations into the Cloud Storage markets like AWS, Google, Private Clouds, etc. to support from both supply and demand side of storage applications.
The needs of a developer community and business user with large data intensive application in the decentralized storage are un-served and under-served with the existing decentralized storage options listed above. The one size fits all monolithic approach to security, privacy and performance are limiting factors for adoption of decentralized storage in Web3.0 ecosystem. Folder Protocol fills the missing gaps and directly addresses needs of these customers with its second layer solution to the decentralized storage platforms. We engage with many customers in understanding their business requirements who want a better privacy, data availability, speed, performance in our proposed solution.
Here are some business use cases of high throughput and low-latency decentralized storage using Folder Protocol:
- Industrial IOT network — secure data sharing among the first and third party industrial IOT devices network in a smart manufacturing system
- Privacy preserving decentralized healthcare data commons — Securing sharing of medical and health records and clinical data for pharmaceutical R&D supply chains, healthcare payers & providers, Distributed Clinical Research, empowering patients with data control, integrating data from the outside the payer & provider sectors
- Decentralized Privacy Preserving AI and Deep Learning — secure data sharing of distributed datasets during the AI and Deep Learning model training and inference and enabling AI edge Computing.
- Private LTE Clouds and 5G based Internet Communication Technologies
