• As of May 18, 2021, Ethereum was ranked 18th worldwide by a market capitalization of over $391 billion, only after Walmar.

• Among the top 300 ICO projects by market capitalization, over 80% are located at Ethereum according to 36kr.com.(https://www.36kr.com/p/1722438500353)

• Ethereum ranks first by DeFi TVL, reaching $45 billion on Q1 2021.

• Ethereum has the most core protocol developers throughout the blockchain ecosystem, with the number of active developers in January 2021 reaching 240. In addition, the number of codes on ETH is eight as much as Bitcoin, showing the vitality of Ethereum eco. (Developer Report released by encrypted asset management company Electric Capital)

• Moreover, the number of daily Twitters tagged Ethereum, and the number of active addresses on Ethereum are both the most among all public chains. There is no doubt that Ethereum is the most powerful ecosystem in the area of block chain.

Last year has witnessed the rise of DeFi. The TVL on Ethereum has surged by 29 times within only one year of 2020, with a 5x market value of stable coins. On the one hand, the outbreak of DeFi has caught the mass public’s eyes to cryptocurrencies; on the other, the demand for network capacity also soared, represented by the number of transactions. Moreover, due to the computational complexity of smart contracts, the number of transactions that a single block can execute has dropped. However, the block generation rate remains constant, thus transactions are processed in an ever-slow speed. This can be demonstrated in the following simple formula:

The time that general transactions have to wait = the ‘length’ of the line speed

The congestion of the network is reflected in a longer line (larger number of transactions waiting to be processed), and a slower speed. As a joint result, the time that a user waits for his/her transaction to be executed is extended significantly when not cutting in line.

The high utilization rate of Ethereum also proves this point. As the most important blockchain in the DeFi ecosystem, DeFi has exploded in Ethereum since 2020, and the utilization rate of Ethereum network has been at a level as high as 95%, even reaching 99% recently, in an almost saturated status.

（data source: etherscan.io）

How is the congestion perceived by users?

It is decided by the mining mechanism. In Ethereum, each transaction needs to be executed and packed by the miners, who are incentivized by the gas fee that users pay. Miners are chiefly motivated by profit, so they tend to execute transactions with higher gas fee.

However, the capacity of each block is limited. When there are too many transactions waiting to be executed, those with low fee will be halted into miners’ local memory; if the offer is still below the ‘bid’ in the next block, the transaction has to keep waiting. This is the so-called ‘execution queue’. Currently there has been a huge backlog; that is to say, if the user sets a low fee, the transaction may be held for more than a few hours or even never be queued.

As shown in the figure, the number of pending transactions on Ethereum is stabilized at around 160k, peaking at over 18k.

（data source: etherscan.io）

This causes users to pay more: time or fee

Congestion forces users to choose between time and money. If users don’t want to pay high gas fees, the only thing that they can do is to wait — trying again and again in a time with less traffic. Or they can just increase gas fee to avoid long and frustrating waiting. The following figure shows the fluctuation of the average transaction fee on Ethereum.

（data source: etherscan.io）

Before the year 2020, transaction fees almost remained below $1 with little ups and downs, peaking at only $5.58 on 23th June. However, with the pouring in of DeFi applications and their interactions with users, transaction fees have experienced a period of fierce fluctuation. So far this year, it has rocketed out of control, even escalated to $68.74 on May 14, meaning that on that day, users paid $68.74 on average for only one transaction.

This overwhelmed a large number of long-tail users and continuously raised the threshold of new players to join in, putting serious obstacles to the long-term development of the whole ecology. As a matter of fact, a lot of Dapps (including some of the bellwethers) have already begun to migrate from Ethereum to the other blockchains, making a wake-up call for Ethereum.

Scaling is urgent

To retain players, this congestion issue must be addressed. The core is to improve TPS, and Ethereum has been upgrading toward its 2.0 vision, even though it is quite a long process.

Comparing to 1.0, Ethereum 2.0 changes in the following to aspects: consensus mechanism (from POW to POS), and sharding. Under traditional POW, all nodes do the exact same executions at the same time, meaning that the amount of transactions the whole network can handle is subject to the upper limit of a single node. However, ‘POS + sharding’ enables Ethereum2.0 to achieve parallel processing, with different sets of nodes verifying different datasets. Ethereum2.0 will be launched with 64 shards, meaning 64 times more throughput on the network, just like expanding a one-lane road into a 64-lane road, lifting TPS from 15 to thousands or more.

It is certain that Ethereum2.0 will improve the base throughput. However, it also means the requirement for more professional and dominant nodes, which is against decentralization; moreover, the upgrades will be put in place over a period of at least three years, so it’s more a long-term revolution rather than an instant remedy.

A quicker solution is in need without spoiling its decentralization and providing a more friendly user experience. When sticking to the main chain is infeasible, a brand new off-chain solution is born, namely Layer2.

Layer 2 refers to a secondary framework or protocol that is built on top of an existing blockchain system. — — Biance Academy

A great portion of computation can be moved from the main chain to this second layer, so the main chain only takes the responsibility of data verification and storage, which considerably improves the efficiency of the main chain.

As a matter of fact, the concept of layer2 has already been applied as early as the Bitcoin period: the Lightning Network built a payment channel, moving the process of multiple transactions between two peers off the chain, and only returning the final account status to the main chain. In August 2017, Joseph Poon, the author of Lightning Network White Paper, and Vitalik, the co-creator of Ethereum, jointly proposed the Plasma scaling idea, which introduced the technique of Fraud Proof to guarantee the authenticity and correctness of the result returned back to the main chian.

In 2018, the rollups came up, putting forward the idea to compress the information of a batch of transactions including state updates and signature verifications, and return this packed data back to the main chain. The idea of rollups compensates for the on-chain data unavailability since Plasma only returns the result of transactions to the main chain, not including other necessary information. There are two kinds of popular rollups methods, namely ZK-Rollup and Optimistic Rollup, using validity proofs (a zero-knowledge proof) and fraud proofs respectively, to ensure on-chain data validity.

Although the concept of Layer2 predates Ethereum 2.0 (not until November 2017 had Vitalik published the roadmap of ETH 2.0), the former had been developing at a very low pace before 2020. Last year, when the explosion of DeFi greatly affected user experience, Layer2 scaling has been refocused.At present, a large number of teams are choosing different Layer2 approaches and compete in both R&D speed and project quality. Each project makes different compromises on either security, decentralization, user experiences, and usability, but it is certain that the public will see plenty number of applications based on different Layer2 solutions in the near future.

In summary, Ethereum2.0 will be a better ecosystem with higher performance in the long run, and on top of that, Layer2 will add to its efficiency. Both scaling methods will complement each other, creating a bigger picture for Ethereum together.