How to expand the blockchain so that it can process hundreds or even thousands of transactions per second without affecting its decentralization characteristics? This problem not only makes some of the smartest people in the field of encryption can't be embarrassed at night, but also makes some of the more fanatical cryptocurrency factions have frequent wars over the years.
In this introductory article on blockchain expansion, news.Bitcoin.com uses very few cryptocurrency projects – Bitcoin Core, Bitcoin Cash and Zilliqa – Different methods of expansion are described.
- Digital economist Chen Xiaohua: Actively exploring the road to the "epidemic" of the blockchain war, the blockchain will usher in a new round of development climax after the epidemic
- The Law of Simplicity: McKinsey's Three Advices for Blockchain Innovators
- Interview with Zheng Zhiming, Academician of the Chinese Academy of Sciences: Blockchain is a relatively mature and reliable evaluation system built for the first time in human history
- Expert comment: Development blockchain should pay attention to open source governance experience
- Operators accelerate the deployment of blockchain, and the 1 billion output value in the telecommunications blockchain field is about to be released?
- Draw a blockchain social portrait
Image source: pixabay
Brief history of expansion
In fact, it is difficult to increase transaction throughput without affecting decentralization. In the early days of Bitcoin, people rarely discussed the expansion problem because it was not a problem at all. At that time, Bitcoin transaction volume per transaction (TPS) was very low and transaction costs were low. Bitcoin has more important issues to address, such as fixing critical vulnerabilities, creating an ecosystem, supporting an active user community, and ensuring Bitcoin survives long enough to scale. This is not to say that expansion has never been mentioned; in fact, this topic has been discussed many times in the Bitcointalk forum, IRC chats and emails, but it is not discussed as an imminent issue.
For example, in an e-mail sent to Bitcoinj developer Mike Hearn, Nakamoto proposed an idea of a payment channel in which “the parties to the transaction can agree to keep updating the transaction data (tx), only the final result will be Recorded by the network." Supporters of the Lightning Network (LN) seized this as evidence of a second-tier solution based on the Bitcoin protocol envisioned by Nakamoto. However, considering that Nakamoto has “disappeared” for nearly nine years, we cannot say with certainty what his vision of expanding Bitcoin along these methods is.
However, it is worth noting that Bitcoin does not have a block size limit when it is initially launched. A year later, Nakamoto introduced a 1MB block size limit to prevent spam attacks. Obviously, the creators of Bitcoin conceived that the future bitcoin network could introduce larger blocks. He wrote in October 2010 that larger blocks could be introduced in stages, such as: if (number of blocks) 115000, the maximum block size = a larger limit [if (blocknumber > 115000), maxblocksize = largerlimit]" Therefore, the expansion method adopted by Bitcoin (Layer 2 / LN) and BCH (larger block) Supporters can claim that their solution meets Nakamoto's early vision. After understanding a very different blockchain expansion approach, we will quickly study the pros and cons of each approach.
If you compare a larger block to building a larger plant, then sharding is similar to creating an assembly line within the plant, so that each worker (miner) assigns a dedicated task, the verification area. Part of the block, not the entire block. Zilliqa is an encryption network known for promoting this expansion technology, and its founder published a paper in 2016 on the feasibility of introducing fragmentation into the public blockchain. By splitting the transaction verification process into multiple parts (or fragments), you can significantly increase throughput and achieve higher TPS.
Zilliqa used 2,400 nodes in its recent test network, claiming that the recorded throughput reached approximately 2,800 tx/s. When asked about the feasibility of applying fragmentation technology to UTXO-based blockchains such as Bitcoin, Zilliqa's chief scientific officer Amrit Kumar told news.Bitcoin.com: "Slice technology can indeed be used in UTXO chains. In fact, there have been some academic works devoted to the UTXO chain-based fragmentation technology. "Shards will play a key role in Ethereum's expansion strategy. Although this method can significantly improve throughput, it also has drawbacks, we will soon study this, but first, let us return to the fierce bifurcation incident caused by the different concept of expansion of Bitcoin.
Bitcoin expansion battle
As Aaron van Wirdum wrote in The Long Road to Segwit:
“Technically, since October 2010, the issue of block size limitation has been highlighted. More specifically, since February 2013, this issue has been openly discussed. By the spring of 2015, this issue It finally broke out."
In this debate, developers such as Gavin Andresen and Mike Hearn are in favor of expanding the bitcoin block size to more than 1MB, while Bitcoin Core developers who are loyal to Blockstream are in favor of retaining the 1MB limit and suggesting other ways to expand. Initially, this was done through Segwit, which reduced the average size of each transaction, but the final solution would be to use a Layer 2 solution lightning network where multiple transactions took place under the chain, only the initial And the final transaction is recorded on the chain.
Due to its complexity, the development of Lightning Networks has been plagued by repeated delays and user experience (UX) issues, and the network is not fully prepared. However, the network has been growing steadily in 2019 and currently has nearly 1100 BTC capacity, more than 8,300 nodes and more than 38,000 channels. The advantages of LN include the ability to settle transactions immediately, and transaction costs are negligible. However, Lightning Networks also has critics who question its complexity and reliance on liquidity providers, and the various intents and purposes of liquidity providers make the network a relatively central and hosted solution. .
Many people who think that the problem should be solved in the chain have turned to Bitcoin cash, and Bitcoin cash is about to celebrate its second anniversary. The miners handled the 8MB block without difficulty, and the cost was still low, making the BCH delivery price less than a penny. Critics of large block expansion methods often seize two points of attack: large blocks need to travel longer in the network, and in the next few years, when block rewards decrease, transaction costs will not be enough to encourage miners to maintain cyber security.
Proponents of the big block claim that increasing data storage costs and download speeds can deny the first criticism, and the reduction in block rewards can be resolved in the future. At least at this time, Bitcoin cash is working as intended by its supporters. However, due to deep-rooted differences and constant criticism, supporters of the chain and chain expansion programs are likely to never agree. The differences between the two sides are very big. The supporters of Bitcoin and Bitcoin cash on Twitter have been making "crossfires" all the time.
The future of blockchain expansion
Although large blocks and lightning networks are seen as the second choice for extending bitcoin, they are actually only two mainstream methods that BCH and BTC value respectively. In fact, there are many other proposals to increase throughput, including sidechains, which can alleviate network stress without having to switch to a hosted solution like Lightning Networks.
For example, although sharding is associated with so-called second generation blockchains, it is also feasible for UTXO based blockchains such as bitcoin and bitcoin cash. However, it also has the disadvantage of having to be adjusted. As explained by Amrit Kumar of Zilliqa:
“In a non-sharded network, each node independently verifies each transaction. However, in a fragmented network, only a subset of the network verifies each transaction. So, in theory, the security you get is indeed slightly Weak. However, in practice, if the slice size (the number of nodes in each slice) is large enough (such as 600 or so) and randomly selected from the initial network, then any bad things happen in the slice. Sex will be very small."
Although different blockchains have different visions on how to achieve high throughput, the differences between projects are increasing. From the birth of Bitcoin to the present, thousands of encrypted networks have been created, and each network is constantly pursuing its own expansion path.