What is 'Sharding' and UTXO Scaling for Blockchain?

Ishan Pandey: Hi Cheng, welcome to our series “Behind the Startup”. Please tell us about yourself and the story behind Alephium?

Cheng Wang: Hi Ishan, first of all, thank you for having me! I have been involved in blockchain research and application for many years. I came to Switzerland in 2013 for my PhD studies.

My research topic was consensus algorithms, long before consensus algorithms became of interest.

During my research, I proposed the first linear-time asynchronous Byzantine consensus algorithm. It’s what drove me to work in blockchain.

Since the release of the Bitcoin white paper, blockchain has gone through its first important decade.

As you know, some of the most popular blockchains are currently facing important scalability and security challenges. Blockchain technology is still at an early stage and the natural question to ask is:

What is the blockchain infrastructure we need for the next 10 years for DeFi (but more generally Dapps)?

We obviously need a blockchain that is scalable with high throughput and low transaction fees. In addition, you want a high level of programmability like on Ethereum.

Last but not least, you want a blockchain that is as reliable and secure as Bitcoin. The Lindy Effect says that the future life expectancy of technologies or ideas is proportional to their current ages. If we apply this theory to cryptocurrency, it means Bitcoin will still be the most robust and promising approach for blockchain.

While all the current layer-1 scaling solutions are in favor of PoS consensus, we believe that sharding + PoW is still the most robust and decentralized way to build a scalable blockchain. As a result, the Company was developed on the idea to build a scalable Bitcoin with a reliable DeFi solution.

Our company is the first operational sharded blockchain scaling and improving on Bitcoin core technologies, Proof of Work & UTXO, to deliver a highly performant, secure DeFi and Dapps platform with enhanced energy efficiency.

It is built on a novel and complete sharding algorithm called BlockFlow. It improves on the UTXO model of BTC to make it scalable and uses DAG data structure to reach consensus between different shards.

Our company’s stateful UTXO model offers both layer-1 scalability and the same level of programmability as the account model. It introduces the concepts of shared state as seen in the Etherum account model.

The stateful UTXO model is leveraged by a custom virtual machine and resolves many of the critical issues of the current dApps platforms with groundbreaking improvements on security, development experience and new paradigms such as trustless P2P smart contracts transactions.

We propose an improved algorithm called proof-of-less-work that uses a clever combination of physical work and token economics to dynamically adjust the work required to mine new blocks, ensuring a reduced energy footprint compared to classic Nakamoto PoW mining.

Putting all these innovations together, our platform delivers a highly demanded solution in the industry: A scalable blockchain improving on mature ideas from Bitcoin to deliver reliable, powerful and secure DeFi and dApps capabilities.

Ishan Pandey: What is sharding? Further, please explain what is BlockFlow sharding architecture? (Detailed technical explanation)

Cheng Wang: Sharding is a technique that enables the splitting of a blockchain into multiple smaller chains called “shards".

Each shard consists of its own data, making it distinctive and independent when compared to other shards. A sharded chain benefits from better scalability, and provides a higher transaction throughput.

The company is built on the BlockFlow algorithm which delivers efficient and practical UTXO sharding. Inspired by distributed computing models, it combines DAG (Directed Acyclic Graph) and sharding.

Much more lightweight than other sharding methods, BlockFlow also natively supports single-step cross-shard transactions making the company user experience just as convenient as any single chain platform.

The addresses are randomly divided into G groups and transactions are divided into G x G shards based on the input and output addresses. As a result, we have G x G chains in total. The transactions from an address belonging to group B to an address belonging to group A are committed to shard (B, A).

For the address in group B, one only needs to download the transaction data of 2G - 1 shards related to itself: namely (X,B) and (B,Y), where X, Y are part of the G groups. As a result, the amount of data that a single node needs to save is reduced from G^2 to 2G-1.

For consensus, we chose the heaviest fork based on the underlying data structure. Each block selects several block hashes as dependencies. Transitively, each new block determines a unique fork for each shard. Miners find the best dependencies and pack valid transactions while nodes validate both dependencies and transactions.

BlockFlow is agnostic to finality algorithms but the company is currently using Proof of Less Work.

Ishan Pandey: What is “Proof of Less Work”, further how does it bring energy efficiency to the bitcoin technology stack?

Cheng Wang: Proof of Less Work is a variant of PoW.

It reduces the energy consumption of the classic proof-of-work by shifting part of the external energy cost to the internal network cost via coin burning when the hashrate and energy consumption are significantly high. It does not sacrifice security as the same level of cost is required for 51% attack.

The exact way in which the PoLW algorithm works depends on the parameters set. Currently, the parameter is set so that ⅞ of block rewards will be required to mine a new block when the energy consumption is significantly high.

The equilibrium between the miners and the protocol changes dynamically. Although this equilibrium cannot be determined in advance, given the same amount of block rewards and hashrate conditions the company will only use ⅛ of the energy compared to Bitcoin.

Ishan Pandey: Recently, BadgerDAO got hacked where $120 million was stolen. Celsius network lost more than $50 million of investors assets. What are your views on cybersecurity and smart contract vulnerability?

Cheng Wang: Every now and then, we hear huge hacks happening on Electronic Voting Machine (EVM) projects.

Many of the issues allowing for these hacks could have been avoided with a different, safer virtual machine design. For example, we designed a brand new virtual machine and language for smart contracts with security in mind.

We resolve many of the issues like unlimited authorization, reentrancy, etc. As Flashloan is an essential tool of DeFi attacks, the company chose to disable it by design.

Ishan Pandey: What are your views on the rising Ethereum transactions fees?

Cheng Wang: The high transaction fees are mainly due to two factors: limited on-chain transaction throughput and almost risk-free arbitrage.

The low throughput can be improved by sharding (ETH2) on layer1, but it has a long road to go before it’s fully implemented. The high number of arbitrage transactions on Ethereum are due to EVM’s design. We improved both parts with our sharding algorithm BlockFlow and our custom virtual machine.

Ishan Pandey: What are your views on Metaverse? Do you think it will break societal constructs like social media?

Cheng Wang: Metaverses are an interesting phenomenon.

They’ve existed for a long time with platforms such as SecondLife. What blockchain could bring to them is decentralization. It could also enable ownership of virtual real estate. And perhaps other things we cannot yet imagine.

This might have an impact on society, but it’s hard to say if it will break societal constructs. To own something in a cryptographically backed virtual space, you still need to own a device, a connection and some currency to buy it.

Meaning so far, they will remain very dependent on how things work in society already. But they might help people to connect in new and unexpected ways, which in turn would definitely have effects away from the devices.

The future will tell us if these effects will break things. Hopefully, they will bring new turns in societal development which will raise new questions and a different look upon what it is to live and coexist in an interconnected world.

Ishan Pandey: What are your views on Web 3.0? How will it disrupt our society?

Cheng Wang: Web 3.0 may disrupt society, but the hope with our platform is that web3.0 will improve society. To disrupt means to break apart and throw into disorder. Rupturing things is a very different process than building.

While it is true that building sometimes leads to abandoning old paradigms, building poses different types of challenges than disruption.

Our company likes to focus on the constructive effects of emerging technologies and how they can provide the foundations for a permissionless, peer-to-peer enabled future.

Ishan Pandey: According to you, what new trends are we going to see in the blockchain industry?

Cheng Wang: The first trend I expect we’ll see would be for some of the alternative layer 1 protocols to stand out, especially after the success of Solana. New projects could provide better and cleaner solutions to scalability, security and user experience.

The second trend would be that interoperability projects will become more production-ready. Many projects have been working on bridging different layer1 and layer2 protocols for a long time, and the best practices will be tested out by the industry in the coming 1-2 years.

The last trend I would predict is that fungible tokens and non-fungible tokens will continue to grow and will be integrated into games, social networks, and more. The potential of tokens is still undervalued.

Disclaimer: The purpose of this article is to remove informational asymmetry existing today in our digital markets by performing due diligence, asking the right questions, and equipping readers with better opinions to make informed decisions.