Watch our September 2025 monthly review livestream covering the latest progress on Leios as the project is getting handed over from IOR to IOE. This month we cover a summary of the CIP-164 reception, early prototyping by Blink Labs, the impact analysis report we had been writing and a look on the roadmap on our journey to a public Leios testnet.
Resources:
This week marked a significant milestone with the formal publication of the Ouroboros Leios CIP proposal, complemented by the release of the second comprehensive technical report and extensive network performance analysis. The team successfully delivered the definitive protocol specification and supporting evidence to the Cardano Foundation for formal review.
The team published the Ouroboros Leios CIP proposal to the Cardano Foundation CIPs repository, representing the culmination of extensive research, analysis, and specification development. The proposal has been submitted for formal review and is pending assignment of a CIP number at the next CIP meeting. This submission provides the definitive technical specification for the proposed Leios protocol implementation, establishing the foundation for community review and potential integration into the Cardano ecosystem.
The team completed and released the second technical report, capturing comprehensive modeling, simulation, and analysis findings from March to August 2025. This substantial document covers network protocols, threat model analysis, simulation experiments, test network topologies, empirical Cardano network measurements, analytic studies of proposed Leios behavior and performance, and extensive technical observations. The report provides provisional findings and conclusions acknowledging the evolving nature of the protocol design during the study period.
The team conducted extensive inter-datacenter bandwidth measurements using iperf3 for bidirectional testing between locations across North America and Europe. The comprehensive analysis examined bandwidth between OVH, AWS, and CenturyLink infrastructure across multiple geographic regions, establishing empirical baselines for network performance expectations in distributed deployment scenarios.
Key findings from the bandwidth analysis indicate that 100 Mbps represents a conservative lower bound for inter-datacenter connectivity, with significant variation based on geographic distance and infrastructure provider combinations. The measurements revealed bandwidth ranges from 95 Mbps (CenturyLink Colorado to OVH Canada) to 973 Mbps (AWS Oregon to OVH Oregon), providing critical input for proposed Leios deployment planning and parameter optimization.
The team completed comprehensive network degradation experiments, examining proposed Leios behavior under severely constrained network topologies. The analysis systematically reduced network connections by up to 87% from original mainnet-like topology configurations, testing protocol resilience under extreme network degradation scenarios.
The degradation experiments demonstrated that the proposed Leios protocol continues to operate correctly even when 87% of network connections are lost, with the network diameter increasing from 5 to 8 hops and the average connections per node dropping from 23.5 to 6.0. The protocol maintained functionality under both honest scenarios and adversarial conditions where attackers delay transaction and EB releases, indicating robust operation under degraded network conditions.
The team extended ledger operation analysis with comprehensive quantile regressions at 50th, 95th, and 99th percentiles for ledger 'apply' and 'reapply' operations. This analysis addressed concerns about validation time interference between EB reapplication and Praos block release timing.
The quantile analysis provides predictions for full EBs with varying Plutus script intensities, demonstrating that 'apply' operations occur distributed across multiple slots and computational threads. At the same time, 'reapply' operations must complete before new reference blocks (RBs) and EBs can build upon newly certified EBs.
The team regenerated comprehensive figures for CIP inclusion using the upgraded sim-cli version 1.3.0, which included updated diffusion and voting duration configurations. The regression experiment analysis compared performance across simulator versions, revealing minimal discrepancies but slightly reduced performance with version 1.3.0 compared to previous versions.
Watch our August 2025 monthly review livestream covering the latest Leios progress, including the CIP proposal publication and second technical report release.
Resources:
This week, the Leios team refined attack analysis methodologies, conducted critical bandwidth limitation experiments, and updated simulation components for CIP documentation. The team successfully demonstrated proposed Leios behavior under constrained network conditions and validated attack experiment findings with improved simulation models.
The team completed validation of the late-release attack experiment using the latest version of the Rust simulation infrastructure. The rerun confirmed previous findings regarding adversarial impact on proposed Leios throughput, strengthening confidence in the protocol's characterized attack resistance properties under the tested scenarios.
The simulation experiment for CIP figures received comprehensive updates with the latest Rust simulation version, incorporating several critical improvements. The enhanced experiments feature semi-optimal protocol parameter settings, improved assumptions for validation costs, and expanded exploration of increased Plutus execution step effects. These refinements provide more accurate performance predictions for CIP documentation.
The team conducted comprehensive bandwidth constraint analysis examining proposed Leios operation under severely constrained network conditions. The experiments systematically reduced inter-node bandwidth to values as low as 1 Mb/s to determine minimum network requirements for protocol viability.
Key findings demonstrate that proposed Leios operates successfully at 0.250 TxkB/s throughput even with constrained 2 Mb/s bandwidth between nodes. However, the protocol experiences breakdown at 1 Mb/s bandwidth limitations, establishing critical minimum network requirements for deployment scenarios. These results provide essential constraints for infrastructure planning and protocol parameter optimization.
The team expanded the regression analysis framework to systematically compare all tagged versions of the Rust simulator against consistent network topology and configuration parameters. This approach accelerates detection of behavioral changes across simulator versions and ensures consistent experimental foundations for protocol analysis.
Simulation infrastructure received updated transaction validation CPU models that improve accuracy of computational cost predictions. These enhanced models provide more reliable estimates for proposed Leios performance under various workload scenarios and support more precise parameter optimization for different deployment environments.
This week, the Leios team conducted comprehensive attack parameter analysis, released updated simulation tools, and performed detailed analysis of Cardano mainnet validation performance. The team successfully characterized the effectiveness of late-release attacks across different parameter ranges and established baseline performance metrics for ledger operations critical to proposed Leios implementation.
The team conducted a comprehensive parameter-sweep experiment to determine optimal adversarial strategies for late Endorser Block (EB) attacks. The experiment systematically varied EB delay timing from six to eight seconds with adversaries controlling 33% of stake, examining both diffusion parameter configurations (L_diff on/off) under the txs-received propagation scheme.
Critical findings from the parameter sweep analysis demonstrate that efficiency degradation begins when EBs and transactions are delayed 6.5 seconds, with minimal additional impact beyond 7-second delays. The analysis revealed that L_diff = 0s configurations perform better than L_diff = 7s under adversarial conditions. Importantly, none of the tested scenarios using txs-received propagation resulted in transaction loss or protocol breakdown, indicating robust behavior under these attack conditions.
The team released versions 1.1.0 and 1.2.0 of the Rust simulation infrastructure, incorporating updated transaction validation CPU models that improve accuracy of performance predictions. These releases enhance the fidelity of proposed Leios simulations and provide more reliable computational cost estimates for protocol analysis.
The team implemented a comprehensive regression analysis framework comparing behavior across all tagged versions of the Rust simulator sim-cli. This systematic approach enables rapid detection of behavioral changes in the simulator across development iterations, ensuring consistency and reliability in experimental results.
The team extended the analysis of ledger operations with comprehensive linear statistical models predicting Apply and Reapply phases of ledger updates. These regressions provide crucial input for proposed Leios performance studies, particularly scenarios involving higher Plutus execution limits.
The team completed analysis of empty block diffusion on Cardano mainnet using data from pooltool.io. This analysis establishes empirical estimates for the Δ_hdr parameter required by the proposed Leios protocol. The findings demonstrate that 94.0% of empty Praos blocks arrive at nodes within one second of their slot start time, providing critical timing constraints for protocol parameter selection.
This week, the Leios team advanced the Cardano Improvement Proposal (CIP) documentation, conducted comprehensive attack analysis experiments, and continued cross-validation between simulation implementations. The team successfully demonstrated resilience characteristics of proposed Leios under adversarial conditions and refined protocol specification components for the formal CIP submission.
The team made substantial progress on the Ouroboros Leios CIP proposal. The draft specification section is now complete except for the network and incentives components. The motivation and abstract sections have been refined for clarity, enhancing the document's accessibility and technical precision. The first of four main rationale subsections has been fully drafted, providing evidence-based arguments for the necessity and viability of the proposed Leios protocol enhancement.
The team conducted the first simulation experiment for attacks, examining late Endorser Block (EB) and transaction diffusion scenarios where adversaries control timing of critical protocol messages. The experiment varied the diffusion parameter L_diff and EB propagation schemes to assess protocol robustness under different adversarial strategies.
Key findings from the initial attack analysis include successful demonstration that late-release attacks can impact proposed Leios throughput under specific conditions. The analysis revealed that transaction loss occurred in some scenarios due to memory pool rule formulations in the simulator, leading to important insights for protocol hardening and implementation requirements.
The team completed another comprehensive comparison between Haskell and Rust simulation implementations at analysis/sims/202532b. Results indicate successful resolution of previous discrepancies in vote diffusion behavior between the two simulation environments. This validation ensures both implementations produce consistent results for protocol analysis and strengthens confidence in the simulation-based evidence supporting the CIP.
The experiment for draft figures for CIP progressed with updated simulation runs designed to generate publication-quality performance charts and protocol behavior visualizations for inclusion in the formal CIP documentation.
This week, the Leios team made significant progress on the Cardano Improvement Proposal (CIP) documentation, conducted extensive Plutus validation experiments, and resolved outstanding discrepancies between simulation implementations. The team successfully demonstrated Linear Leios performance under various Plutus workloads and completed comprehensive protocol parameter analysis for CIP inclusion.
The team completed substantial portions of the Leios CIP draft, bringing it closer to submission readiness. The specification section is now complete except for network and incentives components, with the motivation and abstract refined for clarity. The first of four main rationale subsections has been fully drafted, providing evidence-based arguments for Leios necessity and viability.
The team conducted comprehensive experiments examining Linear Leios performance under varying Plutus computation loads using 6-vCPU nodes at 100 TPS. Key findings from the Plutus validation analysis include:
The analysis revealed significant variability in CPU time requirements for Plutus scripts relative to their execution steps, highlighting the need for careful resource planning in high-throughput scenarios.
The team completed another comprehensive comparison between Haskell and Rust simulations at analysis/sims/202532b, successfully resolving previous discrepancies in vote diffusion behavior. This validation ensures both simulation implementations produce consistent results for protocol analysis.
The 2025w32 experiment established a comprehensive set of protocol parameters and throughput scenarios for inclusion in the CIP Evidence section:
Key performance findings demonstrate that modest computational resources adequately support throughput up to 0.3 TxMB/s, with 7-slot stages providing sufficient diffusion time while minimizing EB discard probability. The 12 MB EB limit allows occasional peak utilization to compensate for unlucky sortition periods.
The team addressed head-of-line blocking issues in Linear Leios by implementing message slicing capabilities in the mini-protocol multiplexer, eliminating unexpected delays in vote diffusion. Additional work focused on developing new mini-protocols for enhanced Linear Leios simulation fidelity, with ongoing refinements to balance protocol granularity and sophistication.
Implementation of transaction withholding attack scenarios for Linear Leios, where EB producers delay transaction publication until EB release. The simulation also received updates to improve handling of late transactions, EBs, and RBs in Linear Leios scenarios.
Watch our July 2025 monthly review livestream covering the latest Leios progress and research developments.
Resources:
This week, the Leios team focused on comparative analysis between simulation implementations and conducted experiments examining the impact of protocol parameters and network topology on Linear Leios performance. The team made progress in cross-validation between Haskell and Rust simulations while investigating protocol behavior under different network conditions.
The team completed a comprehensive comparison between the early draft Haskell simulator and the more mature Rust simulator for Linear Leios. Key findings from the analysis documented in analysis/sims/2025w31b/analysis.ipynb revealed several discrepancies requiring investigation:
The team conducted experiments varying the stage-length protocol parameter from 5 to 12 slots per stage in "No IBs" Leios. Results documented in analysis/sims/2025w31/analysis.ipynb showed:
The team completed a comparative analysis of Linear Leios performance across different network topologies using the Rust simulator. Key findings from analysis/sims/2025w30b/analysis.ipynb include:
This week, the Leios team made significant progress on CIP development, refined validation timing analysis with improved methodologies, and achieved high-throughput validation milestones. The team completed major components of the CIP specification, proposed improved throughput metrics for better comparability, and demonstrated a 1,000 TPS capability with specific protocol variants.