Transactions Per Second (TPS) is a performance metric measuring how many transactions a network or system can process in one second. In blockchain and cryptocurrency, TPS is a key benchmark for evaluating a network's throughput, scalability, and readiness for real-world adoption. Higher TPS means a network can serve more users and applications simultaneously without slowdowns or congestion.
TPS quantifies a network's processing capacity at any moment. In traditional computing, it measures database performance and server capacity. In blockchain, it has added significance because throughput is limited not only by hardware but also by the consensus mechanism, block size, block time, and decentralization level. Both a simple token transfer and a complex smart contract count as transactions, though their computational demands differ greatly.
Before examining blockchain performance, it helps to understand the scale traditional payment processors operate at. Visa's network can handle about 24,000 transactions per second, while Mastercard processes around 5,000 TPS. These figures represent tested capacity ceilings rather than average daily load but set a widely cited benchmark blockchain developers aim for. The gap between traditional finance and most blockchain networks has been a central argument against using public blockchains for high-volume payments.
The decentralized architecture of blockchain networks is the main reason their throughput lags behind centralized payment processors. In a decentralized system, every transaction is broadcast to the network, verified by multiple independent nodes, and permanently recorded on a shared ledger. This preserves security, transparency, and resistance to censorship but introduces latency that centralized databases do not have.
Several technical factors shape how many transactions a blockchain can handle per second. Block time, the interval between consecutive blocks, determines how often new transactions are confirmed. Shorter block times generally increase TPS. Block size limits how many transactions fit in each block; larger blocks allow more activity per confirmation. The consensus mechanism also plays a key role. Proof of Work (PoW) requires nodes to solve computational puzzles before adding blocks, limiting throughput. Proof of Stake (PoS) selects validators based on staked collateral, enabling faster block production and higher TPS.
Bitcoin, the original blockchain, processes about 5 to 7 transactions per second on its base layer. This limit comes from its 10-minute block time and a conservative block size, both chosen to preserve decentralization and security. Ethereum's base layer handles roughly 15 to 30 TPS after its transition to Proof of Stake in September 2022. Solana, built for high throughput, achieves 1,500 to 4,000 TPS in real-world conditions, with a theoretical ceiling of 65,000 TPS. In June 2024, Solana processed over 91 million transactions in one day, showing its architecture's practical reach. BNB Chain processes around 183 TPS, while newer networks like Sui and Hedera push boundaries with speed-focused designs.
A meaningful distinction exists between theoretical TPS and observed TPS. Networks often publish maximum throughput from test environments, while actual performance under live conditions can be much lower. Analysts and developers recommend consulting real-time tracking tools and multiple data sources when evaluating a network's speed.
Recognizing that base-layer limits constrain mainstream adoption, the blockchain industry has developed several scaling solutions to increase effective TPS without compromising decentralization and security.
Sharding divides the blockchain into smaller parallel segments called shards, each processing a subset of network transactions. Multiple shards operating simultaneously allow the network to handle far more activity than a single sequential chain. Ethereum's long-term roadmap includes danksharding to achieve dramatically higher throughput.
Layer-2 protocols offload transaction processing from the main blockchain. Transactions execute on a secondary layer, then are batched and settled on the base chain in compressed form. Ethereum's Layer-2 ecosystem, including Arbitrum, Optimism, and Base, can exceed 40,000 TPS under optimal conditions. The trade-off is added complexity: users must bridge assets between layers, and liquidity can fragment across environments.
Off-chain transactions process activity entirely outside the main blockchain, recording only the final state on-chain. The Lightning Network, built on Bitcoin, works this way. Participants open payment channels, conduct multiple off-chain transactions, and settle the net result on Bitcoin's base layer, enabling fast, low-cost micropayments the base layer alone cannot support.
Novel consensus mechanisms offer another approach. Solana's Proof of History (PoH) creates a verifiable chronological record before events are added to the blockchain, letting validators process transactions in sequence with minimal coordination. Hedera uses a directed acyclic graph (DAG) called Hashgraph, achieving consensus through a gossip protocol instead of traditional block finalization.
A widely cited principle in blockchain design, called the blockchain trilemma, states it is difficult for any network to maximize decentralization, security, and scalability simultaneously. Increasing TPS by using fewer validators or larger blocks can compromise decentralization. Conversely, maximizing security and permissionless participation tends to limit throughput. Networks make different choices based on their use cases, explaining the wide TPS gap between Bitcoin and Solana despite both being public blockchains.
Transaction throughput shapes which applications a blockchain can support. Low-TPS networks suit high-value, low-frequency activities like Bitcoin settlement and Ethereum smart contracts where security and decentralization matter more than speed. High-TPS networks fit gaming, decentralized exchange trading, micropayments, and NFT minting, where many low-value transactions must be processed quickly to maintain usability.
As on-chain activity grows across DeFi, tokenized real-world assets, and consumer applications, the demand for higher throughput will continue to intensify. Networks that can deliver consistent real-world TPS while preserving decentralization will hold a structural advantage in competing for that activity.
.webp)