4 posts tagged with "linear-leios"

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.

CIP development progress​

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.

Plutus validation experiments​

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:

• Linear Leios successfully supported doubling the Plutus per-transaction budget
• Protocol breakdown occurred at sixfold increases in Plutus budgets due to validation bottlenecks
• Endorser blocks could handle approximately 5,000 execution steps of Plutus computation, representing 250 times the current Praos per-block budget
• This capacity could support either a handful of transactions with 20x greater Plutus budgets or increasing every Plutus transaction budget by 50%
• Late diffusion of Plutus-heavy transactions poses potential risks to EB adoption timing.

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.

Simulation improvements and comparisons​

Cross-simulation validation​

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.

Protocol parameter optimization​

The 2025w32 experiment established a comprehensive set of protocol parameters and throughput scenarios for inclusion in the CIP Evidence section:

• Protocol variant: Linear Leios with conservative resource allocation
• Resource requirements: 4 vCPU per node, 10 Mb/s bandwidth
• Stage configuration: 7 slots each for voting and diffusion stages
• Block limits: Maximum 12 MB transaction references per EB
• Transaction size: 1,500 bytes per transaction with normal Plutus frequency

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.

Implementation enhancements​

Haskell simulation​

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.

Rust simulation​

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.

Next steps​

• Complete remaining network and incentives sections of the CIP specification
• Finalize mini-protocol designs for enhanced simulation accuracy
• Continue investigation of mempool rule adequacy for high-throughput scenarios
• Expand Plutus validation analysis to cover additional execution budget scenarios.

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.

Cross-simulation analysis​

Haskell vs Rust Linear Leios comparison​

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:

• CPU usage patterns differed between implementations
• Network usage showed variations across simulators
• Vote diffusion behavior exhibited inconsistencies
• Active investigation underway to resolve implementation differences.

Protocol parameter experiments​

Stage length analysis for "No IBs" Leios​

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:

• Settlement time remains relatively stable across the tested parameter range
• Stage length has minimal impact on transaction processing times within 5-12 slot range
• Larger stage lengths result in less frequent voting periods
• Protocol performance appears robust to moderate stage length variations.

Network topology validation​

Mini-mainnet vs pseudo-mainnet comparison​

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:

• The 750-node mini-mainnet network proved slightly more stressful to the protocol than the 10,000-node pseudo-mainnet
• No substantial differences in simulation results between the two network configurations
• Both networks produce equivalent findings and conclusions for protocol analysis
• Validation confirms the mini-mainnet topology as a suitable proxy for larger-scale analysis.

Rust simulation enhancements​

Linear Leios attack modeling​

• Added equivocation voting delay functionality to Linear Leios implementation
• Implemented attack scenario modeling where endorser block (EB) diffusion is deliberately delayed until the last moment
• Enhanced simulation capabilities for security analysis and adversarial behavior testing.

Next steps​

• Continue investigation of discrepancies between Haskell and Rust Linear Leios implementations
• Expand parameter sensitivity analysis for additional protocol variants
• Refine attack modeling capabilities for comprehensive security assessment
• Apply lessons from network topology comparison to future experimental design.

This week, the Leios team focused on improving simulation analysis, conducting validation time studies, and working on comprehensive protocol variant experiments. The team developed new analysis tools, conducted fundamental performance studies of Cardano validation times, and completed extensive mapping of Linear Leios protocol performance under various conditions.

Analysis tools and infrastructure​

Generic trace analysis framework​

• Developed a generic analysis script for processing Leios simulator output
• Created a comprehensive R-based analysis pipeline generating diagnostic plots from leios-simulation-trace-processor output
• Enhanced analysis capabilities for systematic evaluation of simulation results
• Documentation and usage instructions are available in the trace processor README.

Validation performance analysis​

Cardano mainnet validation timing study​

• Completed a preliminary analysis of block and transaction validation times for Cardano mainnet since epoch 350
• Key findings using the db-analyser tool include:
  • Median transaction signature verification: 0.53 ms/tx
  • Median validation time per kilobyte: 0.29 ms/kB
  • Joint linear model estimate: 0.066 ms/tx plus 0.221 ms/kB
  • Data suitable for bulk estimates but too noisy for individual transaction predictions
• Identified missing explanatory variables (UTXO set size, input/output counts) extractable from the ledger or cardano-db-sync
• Results provide a foundation for more accurate simulator validation time modeling
• Detailed analysis is available in the validation timing notebook.

Simulation optimization studies​

Timestep resolution analysis​

• Conducted a comparative study of simulation timestep effects at 1,000 TPS Full Leios scenarios
• Compared 0.100 ms and 0.025 ms time resolutions with no significant differences in results
• Validated the use of coarser timesteps for improved parallelism and reduced simulation runtime
• Supporting analysis and evidence are available in the timestep study notebook.

Protocol variant experiments​

Mid-throughput protocol validation​

• Completed 100 TPS experiments for Stracciatella and Linear Leios variants using 1,400 B/tx over 900 seconds
• Key findings include:
  • 5 slots/stages are insufficient for Linear Leios at 100 tx/s
  • Including transactions in EBs causes congestion compared to transaction references
  • 10 MB/EB is required for 100 tx/s performance (5 MB/EB is insufficient)
  • EB-sortition unluckiness in Stracciatella extends the transaction lifecycle but can be mitigated
  • CPU and network peaks occur when transactions are embedded in EBs
• Analysis artifacts are available in 100 TPS experiment documentation and the analysis notebook.

Comprehensive Linear Leios performance mapping​

• Completed an extensive simulation study mapping Linear Leios performance under various loads and configurations
• Compared multiple Linear Leios variants against the Stracciatella baseline
• Generated comprehensive performance characterizations across different throughput scenarios
• Results are documented in the Linear Leios summary with variant-specific analysis:
  • Linear Leios with embedded transactions
  • Linear Leios with transaction references
  • Linear Leios without transactions
  • Stracciatella baseline comparison.

Simulation model refinements​

• Revised the Linear Leios model based on analysis findings, particularly regarding partial EB validation before peer propagation
• Reimplemented Stracciatella as a separate simulation to identify specification deviations
• Identified and resolved multiple implementation inconsistencies during the specification verification process.

Next steps​

• Continue developing analysis infrastructure for systematic protocol evaluation
• Integrate improved validation timing models into simulation configurations
• Expand protocol variant testing based on performance mapping results
• Refine simulation models for enhanced accuracy and specification compliance.

This week, the Leios team achieved significant milestones in protocol development and analysis, successfully demonstrating high-throughput capabilities and exploring new protocol variants. The team conducted comprehensive experiments with the Stracciatella variant, analyzed the Linear Leios throughput efficiency, and implemented new simulation capabilities.

High-throughput demonstration​

• Completed experiments demonstrating over 1,000 TPS capability with the Stracciatella variant of Leios
• Achieved spatial efficiency above 95% with transaction lifecycle times under two minutes
• Validated protocol performance under extreme throughput conditions well beyond the current Cardano capacity
• Documented detailed findings in the Stracciatella analysis notebook.

Protocol variants analysis​

Stracciatella variant​

• Completed a comprehensive analysis of the Stracciatella variant (no IBs, transaction references in EBs, two-stage pipeline)
• Key findings:
  • 5-slot/stage performs less well but scales better than 8-slot/stage
  • Only a minimal fraction of transactions fail to reach the ledger, likely due to EB expiration
  • Network usage is slightly heavy, while CPU usage appears suspiciously light
  • Congestion begins to appear at throughput levels above 1,000 TPS.

Linear Leios throughput efficiency​

• Conducted an analysis of the Linear Leios variant's probability of including certified EBs on-chain
• Results show Linear Leios could achieve approximately 500 times the throughput of Praos at over 50% network resource efficiency
• 500 times Praos throughput would exceed 1,000 historically typical transactions per second
• Generated comprehensive throughput and efficiency visualizations available in the analysis repository.

CDDL specifications​

• Added CDDL specifications for Linear and Stracciatella protocol variants.

Simulation improvements​

Rust simulation​

• Implemented a first pass of the Linear Leios variant in the Rust simulation
• Enhanced simulation capabilities to support protocol variant testing and analysis
• Continued optimizing simulation performance for high-throughput scenarios.

Small transaction experiments​

• Completed the analysis of small-transaction, high-throughput experiments with 300-byte non-Plutus transactions
• Key findings:
  • 1,000 tx/s with 300 B/tx is feasible in Leios variants
  • Clear time–space tradeoff between variants
  • full-with-ib-references uses space more efficiently than full-without-ibs
  • full-without-ibs has a shorter transaction lifecycle than full-with-ib-references
  • 2 CPU cores are sufficient for high-throughput operation
  • Network usage remains modest under high load
• Supporting materials are available in analysis documentation and analysis notebook.

Next steps​

• Continue investigation of protocol variants for CIP convergence
• Expand simulation capabilities for additional protocol variants
• Refine performance optimization strategies for high-throughput scenarios
• Complete documentation of protocol variant comparisons and recommendations.