Understanding Blockchain Scalability

One of the biggest problems that public blockchains face today, is scalability. While products are trying to tackle this. Scalability is not a one size fits all solution. In this blog, we look at advancements, technologies, and the next steps!?

First, What is Blockchain Scalability? The ability of a blockchain to process and store a large number of transactions is blockchain scalability, where (TPS) transactions per second measures transaction throughput & bytes of storage required measures blockchain size.

With rapidly growing time-bound use-cases beyond payments. We need blockchains to maintain a target throughput & latency in the face of rising workload without jeopardizing the central tenet of decentralization.?

While scaling some factors to watch include:

1. Transaction throughput: Total no. of transactions processed per second.

2. Cost issue: Total cost for verifying a transaction on the blockchain.

3. Block size: Total storage capacity of block utilized by transactions.

4. Latency: Time between submission & acceptance of transaction by network.

5. Consensus model: Process of approving/verifying blockchain transactions.

6. Storage: Total space/capacity of a blockchain network required to store data.

7. Network Load: No. of transactions carried by network.

8. Computation Energy: Amount of energy required for block mining/validation.

9. No. of Nodes: Total no. of nodes available in blockchain network.

Amidst the above factors, transaction throughput is the prominent cause of scalability issues because low transaction throughput leads to high cost, slow processing & network congestions.?

Before we go deeper into scalability, let’s briefly touch upon Blockchain Trilemma. In a nutshell, the blockchain trilemma asserts that trade-offs between scalability, security, and decentralization are unavoidable while building a blockchain.?

If we look at the existing networks, Bitcoin, Ethereum - Decentralized, Secure, not scalable BSC, Solana, Terra, Fantom, Avalanche - Scalable, Secure but Less Decentralization Hyperledger, Corda - Scalable, Secure but Centralized

Achieving all three characteristics is said to be impossible. However, the innovation across the ecosystem has resulted in a diverse range of Layer-1 and Layer-2 solutions that claim to have found the solution to circumvent the blockchain trilemma.

Layer 1 solutions are innovations or changes to the underlying parent chain making the network scalable. Applied to layer 1 blockchains, the solutions range from simple tweaks to complex upgrades. Some innovations include:

1. Increase the block size limit: By increasing the block size, the number of transactions that can fit into one block increases, and thus performance can be scaled. Solana, for example, has a block size of 10MB whereas Bitcoin has a block size of 1MB.
2. Consensus protocol improvements: Using efficient consensus mechanisms to enable fast consensus and block validation. e.g ETH is now moving to Proof-of-stake from Proof-of-work.
3. Sharding: The concept of breaking the blockchain network into smaller pieces (shards) is called sharding. The shards aim to execute in parallel and process group transactions increasing throughput. Ethereum 2.0 is exploring shards, along with Near, Zilliqa, Qtum, Tezos.

4. Segregated Witness: A Bitcoin extension that focuses on changing the structure of data storage. It removes signature data from each transaction, increasing transaction capacity & storage space.

Fact: Digital signature (tx data) takes upto 70% of the total block space.

Now, let’s look at layer 2 solutions. Layer 2 offloads a portion of the protocol's tx data to adjacent systems and manages the processing before reporting to the parent blockchain. Hence, improving TPS and reducing layer 1 congestion & gas fee.

Some products include:?

State Channels: It allows users to make many off-chain transactions while only submitting 2 transactions to the Ethereum network. E.g. users can deposit funds into contract & sign state updates for moves in a chess game. Later, the only game result is, updated.

Sidechains: A separate blockchain that runs parallel to Layer 1 & operates independently with its consensus mechanisms & security. It is connected to Layer 1 via a 2-way bridge & uses alternate validator selection & consensus mechanisms for faster TPS.

Plasma: They enable fast transactions by offloading them into a small chain (referred to as child chain) anchored to Layer 1. Uses fraud proofs to arbitrate disputes.5 Each child chain has its mechanisms for validating blocks and particular fraud-proof implementation.

Roll-ups: Executes transaction off-chain mainly on a rollup specific chain & then batch transaction data, compresses it & sends it to Layer 1 chain. Its called 'Rollups' because transactions are committed to the main chain in bundles i.e (they are rolled up)

ZK roll-ups: (Validity proofs, On-chain data): Uses Zero-knowledge cryptographic proofs to ensure correctness. Optimistic roll-ups (Fraud proofs, On-chain data): Provides minimum info with no proofs, assumes everything is correct & only provides proof in case of fraud.

Validiums (Validity proofs, off-chain data): Uses validity proofs but data is not stored on Ethereum chain. 10k TPS per Validium chain.

Volitions (Validity proofs, on/off-chain data): Let's easily switch between ZK-rollup & Validium. Interoperable with apps on ZK-rollup.

Examples

? Rollups - Aribitum ( @offchainlabs ) Optimism @MatterLabs @Boba Network

? Plasma - @Omise Co., Ltd. @OMG Network @Polygon Network

? Sidechain - @Polygon Network @Skale Labs @Gnosis

? State Channels - @connext Labs inc @Raiden Network

? Validium/Volition - @Immutable @Starkware

In conclusion, today, there are multiple approaches to scalability each with its one benefit and approach to solve the issue at hand.

In the following blogs, we will unwrap the various scaling technologies to give you a big picture.

Follow @GravityX Capital to stay tuned.?