The Stellar Development Foundation introduced Project Thruput on October 28 as a technology preview designed to change how a blockchain execution engine handles competing transactions. The project is available on FutureNet with open-source material and reproducible benchmarks.
In reported tests, single-validator execution capacity rose from roughly 1.18 million payments per second on one machine to approximately 7.02 million on eight machines. The foundation explicitly states these figures represent controlled component-level measurements and do not represent end-to-end public network throughput.
Technical Design and Approach
The proposed model aims to reduce ordering bottlenecks that occur when many transactions touch shared state. Conventional systems may execute independent instructions in parallel but still serialize work involving the same account or data. Project Thruput is designed to scale execution by adding machines even under highly contended workloads.
The benchmark tests used a limited set of accounts and measured execution, signatures and relevant engine work.
Distinctions Between Measurement and Deployment
Public-blockchain throughput requires more than rapid local computation. It also demands timely consensus and durable data propagation across independent machines. Execution inside a single validator, communication between validators, consensus finality and the performance seen by applications are related but distinct metrics.
Real networks face unreliable links, competing transactions, storage growth and adversarial behavior. A breakthrough in one component can enable broader progress without proving that an entire network is ready for the same demand.
Resource Allocation Model
Project Thruput operates with an alternative resource-allocation design in which accounts receive a scheduling share related to unlocked native-token balances when demand exceeds capacity, while computational limits still apply to individual transactions. This differs from conventional per-transaction execution fees. The model raises questions about fairness, account concentration and protection against abuse.
Next Steps
Independent reproduction of the benchmark, wider workloads and eventual public-network integration are the next meaningful steps. Operators should wait for complete deployment specifications before making commitments to users about transaction speed or fee structures.
