DxChain MVP is Live, Wants to Be Decentralized Hadoop
On July 12, after a four-month design and development, DxChain officially released its MVP, which showcases DxChain’s blockchain storage capability as promised. MVP, which stands for Minimum Viable Product, is a development technique in which a new product or website is developed with sufficient features to satisfy early adopters.
DxChain is a big data network for blockchain storage and computation. Based on DxChain’s roadmap, we first develop the storage solution and then go for the computation aspect. For now, our MVP demonstrates some essential functions in blockchain storage, including generating blockchains, uploading and downloading files.
In this MVP, we use five nodes to build a test network, which generates a DxChain block. We set the block generation time to 30 seconds, upload a 15-byte test file, and download it.
Storage and computation are undoubtedly critical to blockchain. Even the most successful blockchain applications — Bitcoin and Ethereum — expose huge bottlenecks. Bitcoin is mainly used for financial transfer, so it does not involve storage and computation at all. Ethereum appeared to address the issue a little somehow, but its computation is still limited to only three million gas per block.The current smart contracts can only perform some primary functions.
Today’s blockchain technology is mostly applied to direct transfer of money and logistics industries. It faces challenges to unleash its full potential and branch out into new areas.
To solve storage and application problems in blockchain, DxChain introduces several innovations:
1. Framework Innovations
DxChain uses “chains-on-chain” architecture to address the issues of data storage, computation, and privacy. The architecture includes a master chain, a data chain, and a computation chain.
The computation chain is used for parallel computing applications to process large volumes of data, which eventually powers machine learning and business intelligence.
The data chain is used for storing data, protecting users’ privacy, and supporting the computation chain.
The master chain is used for operating blockchain transactions while coordinating the computation and data chains.
2. Reference Architecture
Over the past decade, Hadoop has addressed the issue of distributed data storage within an organization or a company. However, the problems of how to achieve trust between different organizations and participants to make distributed storage remain unsolved.
We integrate Hadoop’s forte, which has been validated in industries over the past decade, with blockchain’s unique mechanism to eventually solve the problems of distributed storage and computing in a decentralized environment.
We hope our public chain will become a distributed Hadoop at the age of blockchain — providing the most stable and universal decentralized storage and computation solution.
3. Technological innovation
Here are the highlighted DxChain’s technological innovations:
1. DxChain uses “verification game + provable data computation (PDC) as its consensus mechanism under the computing framework. Verification game ensures the verifiability of the computation process, and PDC provides the verifiability of the computation result.
DxChain uses “Proof of Spacetime (PoSt) + Provable Data Possession(PDP)” as its consensus mechanism for data storage.
DxChain’s data model is built on storage. It gives data definition to make it valuable and make data computation much more convenient. Meanwhile, the data model also helps realize to privacy protection mechanism — encryption and differential privacy.
Next, we will continue our development and follow our schedule to advance the development progress.
With regards to DxChain:
A Decentralized Big Data and Machine Learning Network Powered by a Computing-Centric Blockchain.