DLT Scalability Solutions: Layer 2, Sharding, and Throughput Optimisation for Swiss Infrastructure

Scalability — the ability of a distributed ledger to process increasing transaction volumes without proportional degradation of performance, cost, or decentralisation — remains the most significant technical constraint on institutional DLT adoption. The throughput of first-generation blockchain networks, measured in single-digit to low-double-digit transactions per second, falls far short of the requirements of financial market infrastructure, which may need to process thousands or millions of transactions per day during peak periods. A diverse ecosystem of scalability solutions has emerged to address this constraint, each offering distinct trade-offs that Swiss institutional users must evaluate against their specific requirements.

The Scalability Trilemma

The scalability challenge in DLT is often framed as a trilemma involving three desirable properties: scalability, security, and decentralisation. Increasing any one property typically comes at the cost of one or both of the others. A highly decentralised network with strong security guarantees (like Ethereum Layer 1) sacrifices throughput. A high-throughput network (like a centralised database) sacrifices decentralisation and potentially security. The art of DLT scalability engineering lies in finding architectures that expand the feasible frontier of this trilemma.

For Swiss institutional applications, the trade-offs within the trilemma are evaluated differently than for general-purpose public blockchains. Financial market infrastructure prioritises security and regulatory compliance above decentralisation, and may accept a degree of centralisation in exchange for higher throughput and lower latency. Enterprise applications may tolerate restricted participation (permissioned networks) in exchange for performance that meets commercial requirements. The scalability solutions adopted by Swiss institutions reflect these priorities.

Layer 2 Scaling

Layer 2 solutions process transactions outside the main chain (Layer 1) while inheriting the security guarantees of the underlying ledger. The principal Layer 2 architectures are rollups, state channels, and sidechains, each with distinct properties relevant to institutional applications.

Optimistic rollups execute transactions off-chain and post the transaction data to Layer 1. The validity of off-chain execution is assumed to be correct (optimistic), with a challenge period during which any participant can submit a fraud proof if they detect an invalid state transition. If a fraud proof is verified, the invalid transaction is reversed. Optimistic rollups offer significant throughput improvements over Layer 1 — typically 10 to 100 times higher — and are compatible with existing smart contract languages, enabling straightforward migration of Layer 1 applications.

Zero-knowledge rollups (zk-rollups) execute transactions off-chain and generate a cryptographic proof (a zero-knowledge proof) that the state transition is valid. This proof is posted to Layer 1 along with a compressed summary of the transactions. The Layer 1 contract verifies the proof, confirming the validity of the off-chain execution without the need to process individual transactions. Zk-rollups offer stronger security guarantees than optimistic rollups (no challenge period required) and potentially higher compression ratios, but are more complex to implement and may have higher proof generation costs.

For a more detailed discussion, see our encyclopedia entry on rollups.

State channels enable two or more parties to conduct an unlimited number of transactions off-chain, with only the opening and closing of the channel recorded on Layer 1. State channels are highly efficient for bilateral transaction flows — such as payment channels or repeated interactions between specific counterparties — but are less suited to applications involving many parties or complex state transitions.

Sidechains are independent blockchains connected to a parent chain through a bridge mechanism. Transactions are processed on the sidechain with its own consensus mechanism, and assets can be transferred between the sidechain and the parent chain through the bridge. Sidechains offer full flexibility in their consensus and execution design but do not inherit the security of the parent chain — their security depends on their own validator set.

Sharding

Sharding divides the ledger into multiple partitions (shards), each processing a subset of the network’s transactions in parallel. By distributing the transaction load across multiple shards, the network’s aggregate throughput increases approximately linearly with the number of shards, while each individual node processes only the transactions assigned to its shard.

The implementation of sharding introduces challenges around cross-shard communication, data availability, and shard assignment. Cross-shard transactions — which involve state on multiple shards — require coordination mechanisms that add latency and complexity. Data availability — ensuring that the data for each shard is accessible to validators who need to verify it — requires mechanisms that prevent individual shards from withholding data. Shard assignment — determining which transactions are processed by which shard — must balance load distribution with the need to co-locate related state for efficient processing.

For more on sharding, see our encyclopedia definition.

Ethereum’s roadmap includes a form of sharding focused on data availability (danksharding), which provides additional data bandwidth for rollups rather than executing transactions directly on shards. This approach, known as rollup-centric scaling, positions Layer 1 as a data availability and settlement layer while delegating execution to Layer 2 rollups.

Modular Blockchain Architectures

The modular blockchain thesis disaggregates the functions of a blockchain — execution, data availability, consensus, and settlement — into separate layers, each optimised for its specific function. This architectural approach enables specialisation and composability, allowing different layers to be selected and combined according to the requirements of specific applications.

Execution layers process transactions and update state. Data availability layers ensure that transaction data is accessible for verification. Consensus layers order transactions and agree on the canonical state. Settlement layers provide finality and dispute resolution.

For Swiss institutional applications, modular architectures offer the flexibility to optimise each layer for the specific requirements of the use case. A tokenised securities application might use a high-performance execution layer for trade processing, a robust data availability layer for regulatory audit purposes, and a secure settlement layer with legal finality guarantees. The modularity enables the selection of components that meet institutional requirements — including regulatory compliance, operational resilience, and data residency — without being constrained by the design choices of a monolithic blockchain.

Throughput Requirements for Swiss Financial Infrastructure

The scalability requirements of Swiss financial market infrastructure vary by application but are generally well-defined.

Securities trading and settlement on SDX requires the capacity to process the daily transaction volume of the Swiss securities market, with headroom for peak periods (such as index rebalancing days, corporate action deadlines, and market stress events). The current daily transaction volume on SIX Exchange is in the range of hundreds of thousands of transactions, and any DLT-based replacement must match or exceed this capacity with acceptable latency.

Payment systems require even higher throughput. The SIC system processes millions of transactions per day, with real-time settlement requirements that demand sub-second confirmation times. DLT-based payment systems must meet these performance standards to be viable alternatives to existing infrastructure.

Enterprise applications have diverse throughput requirements. Supply chain tracking systems may need to process tens of thousands of events per day from IoT sensors distributed across a global logistics network. Trade finance platforms may handle thousands of document exchanges and payment instructions per day. The scalability solution must be matched to the specific throughput requirements of each application.

Performance vs Decentralisation: Institutional Perspective

Swiss institutional DLT users typically prioritise performance, reliability, and regulatory compliance over the maximisation of decentralisation. This priority ordering has implications for the scalability solutions that are adopted.

Permissioned networks with a limited number of high-performance nodes can achieve throughput levels that are orders of magnitude higher than public blockchain networks, because the consensus process is simplified by the known identity and trust relationships of the validators. SDX, for example, operates a permissioned DLT with a small number of trusted validators, enabling it to achieve the throughput and latency required for securities settlement without the scalability constraints of public blockchain consensus.

However, the trade-off between performance and decentralisation is not without cost. Permissioned networks with a small validator set are more vulnerable to validator collusion, single-point-of-failure risks, and governance capture than broadly decentralised public networks. The governance and operational resilience requirements imposed by FINMA on financial market infrastructure are designed to mitigate these risks, but they represent additional operational costs that must be factored into the economic assessment of DLT-based solutions.

Outlook

The scalability of DLT is improving rapidly, driven by advances in cryptographic proof systems, hardware capabilities, consensus algorithm design, and network architecture. The rollup-centric scaling roadmap, particularly the maturation of zk-rollup technology, promises to deliver throughput improvements that bring DLT performance into the range required for most financial market applications.

For Swiss institutional users, the practical question is not whether DLT can scale to meet their requirements — current solutions are demonstrating the necessary capabilities — but rather which scaling approach best fits their specific requirements regarding performance, security, regulatory compliance, and operational complexity.

For related analysis, see our coverage of Swiss DLT node operators and interoperability protocols.

Donovan Vanderbilt is a contributing editor at ZUG DLT, covering distributed ledger technology law, regulation, and institutional adoption from Zurich. The Vanderbilt Portfolio AG provides research and analysis on Swiss digital asset infrastructure.