The Ethereum network has come a long way in the past few years. Everything from the rise of decentralized funding (DeFi) to the recent upgrade in London has made the network the most compelling attempt at creating a “world computer”, but there is still much to be done.

For global acceptance to become the backbone of Web 3.0, the network needs the benefits promised by the Eth 2.0 upgrade. However, more will be needed to scale to a new wave of decentralized applications (DApps), and it looks like Layer 2 solutions may be the only answer.

Related: Do you want to improve the blockchain infrastructure? Working on second layer solutions

The Promises of Ethereum 2.0

In August, Ethereum saw the upgrade being implemented in London. This hard fork marks the first stop on the road to Ethereum 2.0 and has introduced many significant updates to the network in preparation for the transition. London has come as Ethereum continues to struggle under the weight of recent booms in the DeFi and Unavailable Token (NFT) markets. Sometimes the speed and cost of transactions have resulted in many DApps being banned outright, undermining the benefits for which decentralized systems were created.

One of the most notable features implemented by London is EIP-1559, which aims to improve inflation rates and stabilize transaction fees on the network. To do this, it implements a system where the base fees for transactions are burned instead of paid to the miners. Miners still receive block rewards and users can voluntarily add “tips” to their transactions to incentivize priority, but now each block has a certain amount of Ether (ETH) removed from the net forever.

In contrast to Bitcoin, Ethereum does not have a fixed upper limit, so its total supply increases with each block. This has led many to worry about long-term inflation because of the open growth. While EIP-1559 won’t dump Ethereum, it certainly controls how quickly the supply can expand.

While the first step is important, London is only the tip of the iceberg when it comes to scaling Ethereum.

Calls 2.0

Many of Ethereum’s operational problems stem from the fact that the network’s native transaction speed is limited due to its inherent lack of scalability. For comparison: the Ethereum network can currently process around 30 transactions per second (tx / s). For comparison: a traditional payment system like Visa is designed for 1,700 tx / s.

Ethereum has to catch up and that is what Ethereum 2.0 is all about. For one, the network will transition from Proof of Work (PoW) to Proof of Stake (PoS), which will mean moving computers competing to solve complex math problems to one where nodes set assets for block confirmation. Although PoS is much more efficient than PoW and improves network speed to around 50 tx / s, it is still far from the requirements of a global payment system.

This is where another important development of Ethereum 2.0 comes into play: sharding. Sharding is a process of dividing each block into 64 “pieces” that can be processed in parallel. Essentially, this means we can take an estimate of 50 tx / s and multiply by 64, which should give us a little over 3,000 tx / s – way ahead of Visa and more than enough to work.

Related: Ethereum 2.0 upgrades are not a game changer that can bring in more users

Hitting the visa is not enough

While sharding will allow Ethereum to hit or even beat the old payments infrastructure, it may not be good enough. Traditional payment systems mainly deal with relatively simple transactions. This has been fine for years, but the internet, and now DeFi, is pushing things beyond what we could ever have imagined.

Now let’s look at decentralized exchanges around the clock, NFT marketplaces, NFT-powered virtual worlds, and blockchain games. All of this inherently requires a much higher frequency of complex transactions than most traditional payment systems can handle. For example, in a blockchain game, a single player can make multiple transactions per minute, and pausing gameplay to wait for each transaction to complete won’t work. Combine that with DeFi’s ambitious vision of overthrowing the traditional financial sector and you begin to understand how much weight the Ethereum network could carry.

The problem is that even 3,000 Tx / s cannot support these services if they reach worldwide acceptance figures.

However, by combining additional scaling solutions – such as “roll-ups” and “sidechains” – Ethereum has the potential to achieve 100,000 transactions per second. This is great for the high throughput applications DeFi promises, but what are these answers?

Scaling for tomorrow

First of all, there are compilations. They come in many forms, including Optimistic, Validium, Plasma, and ZK. Rollups is a solution for scaling transaction loads by running it off-chain and writing a cryptographic validation on the chain when complete. This frees up resources in the main thread and can increase the overall speed.

Next are sidechains, sometimes referred to as “second-layer” solutions. These are essentially parallel secondary blockchains connected to the main chain. They can be used multiple times and process different processes, which in turn significantly relieves the base layer. The added benefit of sidechains is that they also act as interoperable “bridges” across multiple underlying networks, providing additional liquidity, throughput, and cross-compatibility for connected chains.

Imagine a crypto future where there is an entire ecosystem of main chains like Ethereum, all of which interact with each other through a number of sub-chains. Different networks can be used for your specific solutions, but cryptographic techniques ensure that the data can be verified securely anywhere. Ultimately, this could provide the speed required at low enough cost to ultimately realize the true vision of DeFi, an accessible and affordable financial system for everyone.