Ethereum Scalability: How to Scale Ethereum
Ethereum is among the most ambitious blockchains when it comes to scaling. As a matter of fact, no other blockchain might have a scaling roadmap that is as profound as that of Ethereum. However, scaling is not unique to Ethereum, as every blockchain needs to be scaled to become relevant in a future blockchain-enabled market.
What is Blockchain scalability?
Blockchain scalability is the ability of a blockchain network to support a high-transaction throughput and process many transactions. Hence, a blockchain network that can process many transactions per second (TPS) is said to be scalable. On the contrary, blockchains with low throughputs have poor scalability.
Decentralized systems achieve their decentralization by distributing the decision-making process to all network participants. That makes them superior to centralized systems that rely on central authorities to make decisions. However, requiring many people to make decisions has its drawbacks.
In blockchain systems, decentralization slows down the speed of processing transactions because every node on the network has to validate the transactions. What essentially happens is that all nodes must participate in the consensus process. Each blockchain has a consensus protocol (a set of rules) to guide transaction validation. Nodes also relay block transactions and validations to other nodes to keep the entire network updated. Because this type of validation work is done by each and every node, blockchains like Bitcoin and Ethereum have a low TPS rate, leading to scalability issues.
However, the easiest way for developers to solve scalability issues is to sacrifice decentralization and security. The number of nodes contributes to a blockchain’s decentralization. The more nodes participate in a blockchain’s consensus finding, the more decentralized this blockchain is. The trade-off is known as the “blockchain scalability trilemma.” Trent McConaghy and Ethereum’s co-founder Vitalik Buterin coined this term to refer to the fact that blockchain developers can’t achieve decentralization, scalability, and security all at once. That makes blockchains limited despite their merits.
Understand the Future of Everything
Join the Crypto Revolution
Start Your Free Membership Now
100% Free. Yep, You Heard Us
Why scale a blockchain?
Visa processes about 1,700 transactions per second, while Ethereum only processes around 30 transactions at the same time. Bitcoin is even worse in that regard, as its network processes three to seven transactions per second. Consequently, the two blockchains must scale to become mainstream and service a mass market. To achieve this, they have to find solutions to the following factors that affect blockchain scalability:
Blocks contain batches of transactions forming the blockchain. The number of transactions in a block and a block’s generation time determine the blockchain’s TPS rate. Therefore, block size is crucial in improving a blockchain’s TPS. A large block will hold more transactions, thus improving the TPS. However, increasing block sizes infringes on a blockchain’s decentralization as it becomes harder to run a node that is bound to validate more transactions and therefore needs to be able to process bigger data loads.
Nodes have to download every new block. That means they need space to hold these blocks. As the network expands, however, nodes need more disk space, which can be expensive. This refers back to the problem above. With increased block size, nodes will require bigger storage capacity.
Network congestion due to a high influx of transactions means users have to pay more fees. High demand for space on a block forces users to pay more to “skip” the queue. As a result, transaction fees spike. Ethereum users often pay high gas fees because of network congestion.
Response time refers to the duration users have to wait for nodes to validate their transactions. For instance, the Bitcoin network takes about 10 minutes to confirm a transaction. In Ethereum, it can take up to anywhere between 15 seconds to five minutes. The higher the response time, the longer it takes for a transaction to settle on a blockchain.
How to scale a blockchain
There are three scaling approaches:
Internal scaling involves improving the internal data structures of the node and the consensus protocol. The goal is to optimize resources like storage and the associated costs.
Node splitting entails splitting the responsibilities of the full nodes into sub-nodes. Each sub-node gets a more specific role and is operated by specialized parties. This scaling option seeks to create a verification node that is more affordable for the user to run.
Also known as off-chain scaling, this method entails adding Layer 2 (L2) chains on top of the main chain and can generally be referred to as modular scaling. Bitcoin’s Lightning Network or Ethereum’s various Layer 2 solutions are an example of this.
The first two scaling categories can result in the centralization of the blockchain. It’s for this reason that blockchains like Ethereum (and Bitcoin) are opting for external scaling in the form of Layer 2 solutions or sidechains.
Ethereum’s scaling approach is part of its several upgrades, which used to be called the ETH 2.0 upgrade. As such Ethereum’s scaling efforts are still going on as we write this. As of now, the Beacon Chain that runs on Proof-of-Stake (PoS) has already been introduced (update here: how to stake Ethereum). The next step is integrating this new consensus layer chain with Ethereum’s current mainnet. This will also mark the end of Proof-of-Work (PoW) mining for Ethereum. The third crucial step is implementing Ethereum’s actual scalability roadmap. At the end, Ethereum’s future scalability depends on the fulfillment of these solutions.
Ethereum developers make changes on the base layer through the implementation of a mechanism known as sharding. Sharding involves creating new chains (shards) to boost transaction speed and ease network congestion.
Ethereum is applying off-chain scaling outside its base layer. The most popular off-chain solutions (Layer 2 solutions) are described below:
- Rollups: Rollups execute transactions outside the base layer. More precisely, transactions are bundled and compressed off-chain before being verified at the consensus base layer. The two types of rollups are optimistic and zero-knowledge.
- State channels: They use multisig contracts, allowing users to transact faster off-chain.
- Sidechains: They’re independent blockchains running parallel to the mainnet. Ideally, they get their own blockchain security from the underlying base layer chain.
- Plasma: This is a separate blockchain that settles disputes using optimistic rollups.
- Validium: It uses validity proofs, and data isn’t stored on the base layer.
Is Ethereum more scalable than Bitcoin?
Well, this is a tough question to answer conclusively. As of now, Ethereum could be seen to be more scalable than Bitcoin because it processes slightly more transactions per second. At the same time, because Ethereum’s hashrate is lower and its network might not be as decentralized as Bitcoin, some would consider Bitcoin to be more secure than Ethereum. Moreover, Ethereum is expanding faster than Bitcoin, making it prohibitive for the average user to run a full node. With Ethereum in the process of moving to Proof-of-Stake and adopting an entire modular scalability edifice, we are yet to see whether the higher scalability that comes with this change will also be secure enough.
At the end of the day it can be said: Both Bitcoin and Ethereum are scaling in stages as they work towards widespread adoption. Nevertheless, the two blockchains have varying goals, which mean they might not be comparable in the future as adoption increases. To illustrate, Bitcoin has a clear monetary policy, and it could become a global currency when it achieves mass adoption. Ethereum, on the other hand, has a less clear monetary policy and is more likely to lean towards its functions as a platform for decentralized applications (DApps).