“3 sentences” will allow Ethereum to expand to 1 million TPS in 6 years?

As we all know, blockchain networks are very powerful. They can provide users with a shared distributed digital ledger, allowing participants in a peer-to-peer network to directly exchange information or assets without relying on an intermediary. However, due to architectural reasons, blockchain cannot simultaneously and efficiently achieve the three goals of decentralization, security, and scalability, but can only achieve two of the three, which is also known as the "blockchain impossible triangle" theory.

For most blockchains (such as Bitcoin and Ethereum), they have chosen decentralization and security at the expense of scalability. However, with the continuous breakthroughs in on-chain technology, "all three" will soon become possible. Please follow this article to see how Ethereum can be expanded without affecting the other two properties.

For many years, the "impossible triangle" theory has plagued the blockchain industry, and many alternatives have even chosen to sacrifice decentralization in exchange for transaction speed. Although this helps improve the user experience, once centralization is chosen, the innovation, culture, or benefits related to decentralization will not be realized. Although some expansion solutions including plasma and side chains have appeared on Ethereum, they still have not achieved the design goals.

Modular Ethereum

In general, blockchains are responsible for three main tasks: execution, consensus, and data availability. Execution refers to the occurrence of all transactions, consensus refers to the agreement of network participants on what happened, and data availability ensures that everyone can access the data. Since its inception, blockchains have always been monolithic, which means that they need to perform all three tasks in a single chain. However, it is exciting that more than ten years later, modular blockchain architecture has finally emerged, which uses the concept of specialization to separate these tasks into different independent chains. It is expected that by September 2022, when Ethereum upgrades to Proof of Stake (PoS), each task will have its own different chain: consensus will be handled by the PoS beacon chain, data availability will be handled by the current Ethereum chain, and execution will be handled by Lay2. It is worth mentioning that Lay2 is an independent blockchain built on Ethereum (Lay1). Compared with Lay1, Lay2 has faster throughput and lower fees, while also taking into account decentralization and security.

Nowadays, although many Lay2s have emerged on Ethereum, it seems to be far from the expansion goal. Ethereum's current processing speed is about 10 TPS (transactions per second), but to achieve its expansion goal, the speed must be increased by at least 100,000 times (Visa's processing speed is about 65,000 TPS). In order to expand to 1 million TPS, Ethereum needs to further complete the following three things-improve rollups, sharding, and increase network bandwidth.

1. Rollups — 100x throughput increase​

Rollups are independent blockchains that can take on the task of executing transactions, so Ethereum can focus on consensus and data availability. Executing transactions on Rollups instead of the main chain can make the Ethereum network 10 times more scalable. In addition, more optimizations can be achieved through upgraded data compression technology. Data compression refers to reducing the number of details required to represent the same basic information. Publishing compressed data to Ethereum will further reduce costs and increase scalability by another 10 times.

2. Data Sharding - Increases Throughput 100 Times

Currently, each consensus participant must download all the data from the block and independently verify the data before verifying the block. The whole process is very inefficient, making it a major bottleneck of the blockchain network. Sharding, as a network architecture that expands data availability, allows network participants to only download a sample of all data while ensuring data availability, increasing Ethereum scalability by 10 times. In addition, sharding introduces a new transaction type - "blob-carrying transactions". Compared with current transaction types, the transaction cost of carrying blobs on Lay2 is much lower and can increase Ethereum scalability by another 10 times.

3. Network bandwidth – Increase throughput by 10 times

The last part of Ethereum that needs improvement, “network bandwidth”, is different from the previous two parts in that it relies on external technological advances to expand the network. Typically, consumer bandwidth increases by about 50% per year (similar to Moore’s Law), so it takes about five to six years to achieve a 10x bandwidth increase. The following figure shows the growth of network bandwidth from 1983 to 2019.

Summarize

If Ethereum wants to gain mass adoption, it is imperative to improve scalability. Once the Ethereum merger is successful (expected to be completed in September 2022), the expansion technologies mentioned in this article will become a priority. Considering the immaturity of new technologies and the growth limitations of network bandwidth, Ethereum is expected to expand to 1 million TPS in the next six years.