The Scalability Problem
Bitcoin processes roughly 7-15 transactions per second post-SegWit on its base layer, depending on transaction mix (simple Taproot transactions take less space than complex multisig). For comparison, Visa handles around 1,700 transactions per second on average and claims a theoretical capacity above 65,000. If Bitcoin aspired to be a global payment network using its base layer alone, it would need to close a gap of several orders of magnitude.
The constraint comes from Bitcoin's block structure. Every ten minutes, on average, miners produce a block with a maximum weight of 4,000,000 weight units (4 MWU), as defined by MAX_BLOCK_WEIGHT in Bitcoin Core's consensus/consensus.h. Every full node on the network downloads, validates, and stores every block. That combination of fixed block intervals and fixed block weight creates a hard ceiling on base-layer throughput.
The Scalability Constraint
Solo payouts arrive too rarely for most miners to rely on them. A typical SegWit transaction weighs roughly 560-680 weight units (140-170 vbytes at the SegWit 4x witness discount). Dividing the 4 MWU block limit by an average transaction weight of about 600 WU yields roughly 6,600 transactions per block. At one block every 600 seconds (ten minutes), that works out to approximately 11 transactions per second (TPS) under ideal conditions. In practice, the mix of transaction types (some larger, some smaller, multi-input/multi-output transactions weighing considerably more) brings the effective throughput closer to 7 TPS.
MAX_BLOCK_WEIGHT was set at 4 MWU during the SegWit soft fork (BIP 141, activated in August 2017) as a deliberate constraint. Increasing it would proportionally increase the bandwidth, storage, and validation costs for every full node on the network. The constraint exists to keep node operation feasible on consumer hardware, which is the mechanism that preserves Bitcoin's decentralization.
For comparison, Visa's average throughput of roughly 1,700 TPS would require Bitcoin blocks approximately 240 times larger, or about 960 MWU per block. Sustained at one block every ten minutes, that would multiply the chain's growth by the same factor, expanding the blockchain by roughly 12 to 17 terabytes per year instead of the current 50-70 GB. No consumer hardware could keep up. The nodes that remained would be data centers, and Bitcoin would have the same trust architecture as the banking system it was designed to replace.
This ceiling exists by design. It protects two properties that make Bitcoin valuable in the first place: decentralization and security.
Why Not Just Make Blocks Bigger?
The most obvious solution to limited throughput is to increase the block size. Larger blocks fit more transactions per block, which raises the transactions-per-second ceiling. This was proposed many times throughout Bitcoin's history, and the debate over it (the "block size wars" of 2015-2017) was one of the most consequential disagreements the Bitcoin community ever faced.
The community rejected large block increases for a clear reason: bigger blocks make it harder to run a full node.
The Full Node Bottleneck
A full node validates every transaction against Bitcoin's consensus rules. It does not trust miners, exchanges, or any third party. Anyone running a full node verifies the entire blockchain independently. This is the mechanism that keeps Bitcoin decentralized.
Larger blocks increase three resource requirements for full nodes:
- Bandwidth: Each block must be downloaded and propagated across the network. At the current 4 MWU limit, a full block is roughly 1-2 MB of actual data (after the SegWit witness discount). Doubling the block weight doubles the bandwidth needed to keep up in real time. Nodes on slow connections or in regions with limited internet infrastructure would fall behind.
- Storage: The full blockchain must be stored locally. As of early 2026, the Bitcoin blockchain exceeds 600 GB. At 4 MWU per block and 144 blocks per day, the chain grows by roughly 50-70 GB per year. Larger blocks would accelerate this growth proportionally, pricing out nodes running on consumer hardware.
- Computation: Every transaction in every block must be validated, including signature verification (ECDSA for legacy transactions, Schnorr for Taproot). More transactions per block means more CPU cycles per block interval. Bitcoin Core's
sigopslimit caps the number of signature operations per block to prevent computational denial-of-service attacks.
If running a full node required a data center instead of a laptop, fewer people would do it. Fewer nodes means fewer independent validators, which means more trust concentrated in fewer hands. That is the opposite of what Bitcoin was built to achieve.
The block size wars ended with a clear consensus: Bitcoin's base layer would stay lean and accessible. Scaling would happen in layers above it.
The Layered Architecture: How the Internet Solved the Same Problem
Bitcoin's approach to scaling follows the same design principle that made the internet work: protocol layering.
The internet's base protocol, TCP/IP, does one thing well: it moves packets of data between computers. TCP/IP has no concept of a webpage, an email, or a video stream. Those capabilities are built in layers above it, where HTTP handles web pages, SMTP handles email, and RTP handles real-time video. Each protocol is specialized for its purpose, and all of them rely on TCP/IP underneath for reliable data delivery.
TCP/IP could have been designed to handle web pages, email, and video natively. But that would have made the base protocol more complex, harder to maintain, and more fragile. By keeping the foundation simple and reliable, the internet's designers allowed an entire ecosystem of specialized protocols to be built on top of it without modifying the base.
Bitcoin uses the same strategy. The base layer does one thing well: it provides a secure, decentralized, censorship-resistant ledger for final settlement. Every other capability, from instant micropayments to confidential transfers to digital asset issuance, is built in layers above it.
The Base Layer: Bitcoin's Settlement Foundation
The Bitcoin base layer (sometimes called "layer 1" or the "settlement layer") provides three guarantees that no other system replicates:
- Finality: Once a transaction is buried under enough blocks, reversing it requires more computational power than any attacker can marshal. Six confirmations (roughly one hour) is the standard threshold for high-value settlements. At that point, the transaction is effectively irreversible.
- Censorship resistance: No single entity can prevent a valid transaction from being included in a block. Miners compete to include transactions, and any miner anywhere in the world can include any valid transaction.
- Trustlessness: Any participant can verify the entire state of the ledger independently. No trust in any third party is required.
These properties make the base layer the strongest settlement system ever built. A large on-chain transaction confirmed in several blocks carries a level of settlement assurance that no bank, clearinghouse, or payment processor can match.
The tradeoff is speed and cost. A base-layer transaction takes an average of ten minutes for the first confirmation, and fees rise when block space demand is high. For a $500,000 settlement between institutions, a $5 fee and a 30-minute wait is trivial. For buying a coffee, it is not.
This is where layers come in.
Layer-2 Solutions: Inheriting Security, Adding Capability
The key insight behind layer-2 design is that not every transaction needs the full weight of the Bitcoin base layer's security. A coffee payment does not need the same settlement assurance as a real estate closing.
Layer-2 networks handle the high-volume, lower-value, or time-sensitive transactions that would be impractical on the base layer. When participants need the full security of the base layer, they settle back to it. The base layer becomes the final arbiter, the court of last resort, while layer-2 networks handle the day-to-day.
The Lightning Network: Instant Payments at Scale
The Lightning Network is a layer-2 payment channel network built on top of Bitcoin. It enables near-instant, low-cost payments by moving transactions off-chain while preserving the ability to settle to the base layer at any time.
Here is how it works at a high level:
- Open a channel: Two parties lock bitcoin into a shared 2-of-2 P2WSH multisig output using a funding transaction (BOLT 2). This is an on-chain transaction that creates a payment channel between them.
- Transact off-chain: Once the channel is open, the two parties can send bitcoin back and forth instantly by exchanging signed commitment transactions between themselves. Each new commitment transaction invalidates the previous one through a revocation key mechanism. These transactions are valid Bitcoin transactions, but they are not broadcast to the blockchain. Thousands of payments can happen in a single channel without touching the base layer.
- Route through the network: Payments can be routed across multiple channels using Hash Time-Locked Contracts (HTLCs), as specified in BOLT 2 and BOLT 3. If Alice has a channel with Bob, and Bob has a channel with Carol, Alice can pay Carol through Bob without needing a direct channel to Carol. Onion routing (BOLT 4) ensures that no single intermediate node knows both the original sender and the final recipient. This creates a network effect: as more channels exist, more payment paths become available.
- Close the channel: When either party wants to settle, they publish the final state of the channel to the Bitcoin base layer. The on-chain transaction distributes the bitcoin according to the most recent balance. If one party broadcasts an outdated state, the other can claim all channel funds via a penalty (justice) transaction (BOLT 5).
The result is a payment network whose throughput scales with channel liquidity and routing capacity rather than a fixed protocol-level block interval, with fees measured in fractions of a sat and payments that settle in seconds. Lightning is ideal for everyday payments, micropayments (tipping content creators in sats, paying per-article or per-minute), and point-of-sale transactions.
Blockstream maintains Core Lightning (CLN), one of the major open-source Lightning Network implementations. CLN is designed for reliability and extensibility, with a plugin architecture that allows developers to build custom functionality on top of the Lightning protocol.
The Liquid Network: Faster Settlements, Privacy, and Digital Assets
The Liquid Network is a Bitcoin sidechain that addresses a different set of limitations than Lightning. Unlike Lightning, which is a true layer-2 that derives its security from the Bitcoin base layer, Liquid is a separate blockchain with its own consensus, connected to Bitcoin by a two-way peg. Where Lightning optimizes for instant small payments, Liquid optimizes for faster settlements, transaction privacy, and the ability to issue new digital assets anchored to Bitcoin.
Key properties of the Liquid Network:
- Faster block times: Liquid produces blocks approximately every minute, compared to Bitcoin's ten-minute average. For traders and businesses that need faster confirmation, this reduces settlement time from an hour (six Bitcoin confirmations) to about two minutes: Liquid blocks are final when signed, and two Liquid confirmations (~2 minutes) is the conservative settlement standard.
- Confidential Transactions: By default, Liquid transactions use Confidential Transactions (CT), a cryptographic protocol that hides both transaction amounts and asset types from outside observers. Only the sender and receiver know the details. This gives businesses and individuals privacy that is unavailable on the transparent Bitcoin base layer.
- Asset issuance: Anyone can issue digital assets on Liquid, including stablecoins (like USDT on Liquid), security tokens, and other financial instruments. These assets benefit from the same Confidential Transactions privacy and the same one-minute block times as LBTC.
- Federated security: Liquid is secured by a distributed federation of major Bitcoin industry participants (exchanges, financial institutions, and infrastructure providers). This federation model trades some of Bitcoin's fully decentralized security for significantly faster confirmation times and additional features.
To move bitcoin onto the Liquid Network, a user performs a peg-in: locking bitcoin on the mainchain and receiving an equivalent amount of LBTC on Liquid. To move it back, they perform a peg-out. LBTC is pegged 1:1 to bitcoin, so one LBTC always represents one bitcoin held in the federation's custody.
Beyond Lightning and Liquid: Other Scaling Approaches
Lightning and Liquid represent two distinct architectural approaches to scaling Bitcoin in layers: state channels and sidechains, respectively. State channels like Lightning are true layer-2s secured by the base layer, while sidechains like Liquid run their own consensus and connect through a peg. The broader set of designs includes additional approaches, each with different tradeoff profiles.
State Channels
Lightning is the most prominent example of a state channel network. State channels work by locking funds on-chain, conducting transactions off-chain between participants, and settling the final state back to the base layer. The base layer is the enforcement mechanism: if one party tries to cheat, the other can publish the correct state to the blockchain and claim the funds.
State channels offer strong security guarantees because any dispute can be resolved on-chain. The tradeoff is that participants must lock capital in channels, and routing payments across many hops can require significant network liquidity.
Sidechains
The Liquid Network is the most widely used Bitcoin sidechain. A sidechain is a separate blockchain that runs alongside the main Bitcoin blockchain, with a mechanism for moving bitcoin between the two. Sidechains can implement different consensus rules, block times, and features without modifying Bitcoin's base layer.
The sidechain concept was first described in a 2014 whitepaper co-authored by Blockstream co-founders. The core insight: by allowing bitcoin to move between chains with different properties, developers can experiment with new features and optimize for specific use cases without risking the stability of the base protocol.
Rollups
Rollups are a layer-2 approach that bundles (or "rolls up") many off-chain transactions into a single on-chain proof. The base layer verifies the proof rather than each individual transaction, allowing much higher throughput while anchoring security to Bitcoin.
Rollups are more established in other blockchain ecosystems, but research is active on bringing rollup-style constructions to Bitcoin. Potential approaches include validity rollups (which use cryptographic proofs to verify off-chain computation) and sovereign rollups (which use Bitcoin purely for data availability). These designs are still early-stage for Bitcoin, but they represent a growing area of layer-2 research.
The Broader Bitcoin Scaling Ecosystem
Beyond Lightning and Liquid, the wider Bitcoin scaling ecosystem includes several other projects that take their own approaches. Rootstock (RSK) is a smart-contract sidechain that runs an Ethereum-compatible virtual machine and is secured by merge-mining with Bitcoin. Stacks is a separate layer that anchors its own blocks to Bitcoin and adds a smart-contract environment for Bitcoin-related applications. Other emerging designs explore drivechains, statechains, and BitVM-based constructions. These projects make different tradeoffs around programmability, trust model, and how tightly they couple to Bitcoin's security. Among them, the Lightning Network and the Liquid Network are the two most widely deployed and production-ready, which is why this guide focuses on them.
How the Layers Work Together
Each layer in Bitcoin's architecture handles a different class of transactions, and they complement each other rather than compete.
| Use Case | Best Layer | Why |
|---|---|---|
| Large settlement ($100K+) | Base layer (on-chain) | Maximum security and finality. The fee and wait time are trivial relative to the value being settled. |
| Daily payment (coffee, groceries) | Lightning Network | Instant confirmation, sub-sat fees. No reason to use block space for a $5 payment. |
| Micropayment (content tip, API call) | Lightning Network | Lightning can handle payments smaller than the minimum on-chain transaction fee. |
| Trading between exchanges | Liquid Network | One-minute blocks and Confidential Transactions. Faster and more private than on-chain settlements. |
| Private transfer | Liquid Network | Confidential Transactions hide both amounts and asset types. On-chain Bitcoin transactions are publicly visible. |
| Issuing a stablecoin or security token | Liquid Network | Liquid supports native asset issuance. The Bitcoin base layer does not. |
| Long-term savings (cold storage) | Base layer (on-chain) | Maximum security for long-term holdings. No counterparty risk, no federation trust assumptions. |
In practice, users often move bitcoin between layers depending on what they need. A user might hold long-term savings on-chain, move some bitcoin to Lightning for daily spending, and use Liquid for a private transfer or to access USDT. The Blockstream app supports Bitcoin, Lightning, and Liquid from a single self-custodial interface, making it straightforward to manage bitcoin across all three layers without switching between different wallets. For hardware wallet security, Jade Plus ($149-169) signs both on-chain and Liquid transactions with air-gapped QR codes, so users can keep their long-term savings and Liquid holdings protected by the same device. Entry-level users can start with Jade Classic ($79) for on-chain hardware wallet security.
The Tradeoff Triangle: Security, Speed, and Decentralization
Every transaction system, whether digital or physical, faces a fundamental engineering constraint. Such a system can optimize for security, speed, and decentralization, but strengthening one property typically requires weakening another. This is sometimes called the "blockchain trilemma".
Bitcoin's layered architecture works because each layer chooses a different position on the tradeoff triangle:
| Layer | Security | Speed | Decentralization |
|---|---|---|---|
| Base layer (on-chain) | Maximum (proof-of-work, full node verification) | Slow (~10-minute blocks) | Maximum (anyone can run a node) |
| Lightning Network | High (disputes settled on-chain) | Near-instant | High (peer-to-peer channels) |
| Liquid Network | High (federated consensus, signing by federation members) | Fast (~1-minute blocks) | Moderate (federation of vetted members) |
No single layer needs to be perfect at everything. The base layer anchors the entire system with maximum security and decentralization. Lightning adds speed without sacrificing much security. Liquid adds speed and privacy with a different trust model. Users choose the layer that fits their specific needs for each transaction.
This is the power of the layered approach: rather than forcing one protocol to do everything (and doing everything poorly), Bitcoin lets each layer excel at what it does best.
Why Layers Matter for Bitcoin's Future
Bitcoin's layered architecture is a deliberate design choice. The decision to keep the base layer simple, secure, and accessible to anyone with a consumer computer is what makes Bitcoin a credible foundation for a global financial system.
Every successful piece of infrastructure in computing history follows this pattern. TCP/IP did not try to be HTTP. SMTP did not try to stream video.
Bitcoin's base layer settles transactions with a level of assurance no other system provides. Lightning makes bitcoin practical for daily spending. Liquid makes bitcoin practical for trading, privacy, and asset issuance. Future layer-2 solutions will add capabilities that do not exist yet, all without requiring changes to the base protocol.
The layered design is what lets Bitcoin stay decentralized at the base while still scaling to everyday use.
Frequently Asked Questions
Is Lightning Network a separate cryptocurrency from Bitcoin?
No. Lightning is a protocol layer built on top of Bitcoin, defined by the BOLT specification (Basis of Lightning Technology). The bitcoin on Lightning is real bitcoin, locked in on-chain 2-of-2 multisig payment channels. When a Lightning channel is closed, the final balances are settled back to the Bitcoin base layer as a standard on-chain transaction. Lightning does not have its own token or separate blockchain.
Why can't Bitcoin just increase its block size to handle more transactions?
Larger blocks increase the bandwidth, storage, and computation required to run a full node. The current MAX_BLOCK_WEIGHT of 4 MWU (set by BIP 141 during the SegWit activation in August 2017) was chosen to keep full node operation feasible on consumer hardware. Doubling the block weight would double the storage growth rate (currently roughly 50-70 GB per year) and the bandwidth needed to relay blocks in real time. If running a node required expensive server hardware instead of a consumer laptop, fewer people would do it. Fewer independent nodes means more centralization, which undermines Bitcoin's core value proposition. The Bitcoin community chose to keep the base layer accessible and scale through layers instead.
Is Liquid Network as secure as the Bitcoin base layer?
Liquid uses a different security model. Instead of proof-of-work mining, Liquid is secured by a distributed federation of vetted Bitcoin industry participants. This federated model enables faster block times (~1 minute vs. ~10 minutes) and features like Confidential Transactions, but it requires trusting the federation to operate honestly. For most use cases involving faster settlements, trading, and privacy, this tradeoff is appropriate. For maximum-security long-term storage, the base layer remains the strongest option.
Do I need to choose one layer, or can I use multiple?
You can use all of them. Many Bitcoin users hold long-term savings on-chain, use Lightning for everyday payments, and use Liquid for faster settlements or privacy. The Blockstream app supports all three layers from a single self-custodial interface, so managing bitcoin across layers does not require separate wallets. Lightning ↔ Liquid swaps within the app let you move between instant payments and private settlements in seconds.
What happens if a layer fails or shuts down?
How your funds are protected depends on the layer. Lightning is a true layer-2: you can close your payment channels at any time and settle your funds back to the Bitcoin base layer without anyone's permission, because the base layer enforces the channel's latest state. Liquid is a sidechain, so the protection works differently. Your LBTC is backed 1:1 by bitcoin held in the federation's multisig, and you peg out to reclaim on-chain bitcoin. That peg depends on the federation honoring it rather than on base-layer enforcement alone, which is the tradeoff a sidechain makes for its added features.
Are rollups coming to Bitcoin?
Research is active. Validity rollups and sovereign rollups are being explored as potential layer-2 solutions for Bitcoin. These are still in early stages compared to Lightning and Liquid, which are production-ready and widely used today. Rollups could eventually add another option to Bitcoin's layer-2 toolkit, but they are not yet deployed at scale on Bitcoin.