Blockchain Research

Blockchain scaling

TL;DR — Scaling is the effort to raise a blockchain's transaction capacity without sacrificing decentralisation or security. The dominant approach today moves execution off the base layer (layer 2 rollups) while keeping verification and data on it.

Scaling is the problem of letting a blockchain handle far more activity without giving up the properties that make it worth using in the first place — decentralisation and security. It is harder than it sounds, because the most obvious fixes quietly undermine those properties.

Why base layers are limited

A public blockchain asks every full node to independently re-execute and store every transaction. That redundancy is exactly what makes the chain trustless: anyone can verify the whole history themselves. It is also the throughput ceiling. Raising capacity by enlarging blocks or shortening block times increases the hardware, bandwidth, and storage needed to run a node — and as those costs rise, fewer independent operators remain, pushing the network toward centralisation.

This is the blockchain trilemma in action: decentralisation, security, and scalability resist being maximised together. Serious scaling work tries to ease that tension rather than pretend it does not exist.

The main approaches

  • Layer 2 (rollups). Execute transactions on a separate system and post compressed data — and, for some designs, a proof — back to the base layer, which keeps the result honest. This is the dominant strategy today; the base layer becomes a settlement and data-availability anchor while execution happens off-chain. See what are layer 2 solutions.
  • Sharding. Split the base chain into parallel partitions (“shards”) so that not every node has to process everything. Modern designs increasingly use sharding to scale data availability for rollups rather than to execute transactions directly.
  • Modular vs monolithic designs. A monolithic chain does execution, settlement, consensus, and data availability in one layer. A modular design separates those concerns across specialised layers, letting each scale independently — the architectural direction much of the ecosystem is moving in.
  • Sidechains and state channels. Older approaches that move activity off-chain with weaker or different security guarantees; useful in specific cases but not security-equivalent to rollups.

The recurring question

For any scaling design, the most revealing questions are the same: where does the transaction data live, and what does the base layer actually guarantee? The answers determine whether a “scaling solution” genuinely inherits base-layer security or quietly reintroduces trust. The cluster articles below dig into layer 2 and zero-knowledge rollups, the designs the Ethereum scaling docs treat as the primary path forward.

In this guide

Frequently asked questions

Why can't a blockchain just use bigger blocks?
Bigger blocks raise throughput but also raise the cost of running a full node, since every node must download, execute, and store every transaction. As node requirements climb, fewer people can participate independently, which weakens decentralisation — the core tension scaling work tries to avoid.
What is the difference between layer 1 and layer 2 scaling?
Layer 1 scaling changes the base chain itself (e.g. sharding, faster consensus). Layer 2 scaling keeps the base chain as-is and moves execution to a separate system that settles back to it, inheriting the base chain's security while processing transactions more cheaply.
What is the blockchain trilemma?
It is the observation that decentralisation, security, and scalability pull against each other, so naive attempts to improve one tend to weaken another. Most modern scaling research is about easing that tension rather than ignoring it.

Sources

Published 2026-06-16 · An educational project of the Mashiyu Foundation.