Skip to content

Commit 8ff5484

Browse files
committed
site: weekly update
1 parent e4a9fca commit 8ff5484

File tree

1 file changed

+51
-0
lines changed

1 file changed

+51
-0
lines changed
Lines changed: 51 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,51 @@
1+
---
2+
title: Weekly Summary – June 17, 2025
3+
authors:
4+
- will
5+
tags: [progress, update, weekly, bandwidth, experiments, throughput, analysis, simulation, formal-methods, cddl, specifications]
6+
---
7+
8+
This week, the Leios team conducted extensive experiments using the previously developed network topologies to study transaction and input block throughput limits under realistic conditions. The team also conducted empirical bandwidth measurements between data centers, advanced formal methods capabilities, and created initial CDDL specifications for core Leios components.
9+
10+
## CDDL specification draft
11+
12+
- Created initial CDDL specifications for core Leios components:
13+
- Input Blocks with VRF lottery and single IB/slot limits
14+
- Endorser Blocks as new aggregation block type
15+
- Ranking Blocks as Conway extension with optional certificates
16+
- BLS voting system with persistent/non-persistent voters and key registration
17+
- Follows crypto-benchmarks implementation approach while maintaining Conway CDDL compatibility
18+
- Established foundational structures in first draft covering common base components
19+
- Future iterations will add detailed specifications for design variants including full sharding, overcollateralization, and protocol extensions.
20+
21+
## Formal methods
22+
23+
- Added support for `Late IB inclusion` to the formal specification of Full-Short Leios
24+
- Profiled leios-trace-verifier performance, identifying that approximately 60% of execution time is spent in garbage collection
25+
- Improved performance significantly by switching to `--nonmoving-gc` garbage collection strategy.
26+
27+
## Bandwidth measurements
28+
29+
- Conducted empirical bandwidth measurements using `iperf3` between data centers in North America and Europe
30+
- Measured bidirectional connections across multiple cloud providers (OVH, AWS, CenturyLink)
31+
- Results ranged from 95 Mbps to 973 Mbps depending on geographic distance and provider
32+
- Identified 100 Mbps as a conservative lower bound for inter-datacenter connections
33+
- Observed 5-20% reduction in individual link speeds when multiple simultaneous connections are active.
34+
35+
## Large-scale network experiments
36+
37+
- Conducted comprehensive experiments using both the 750-node and 10,000-node network topologies with Haskell and Rust simulations
38+
- Studied transaction and IB throughput limits for realistic scenarios up to 300 TPS and 32 IB/s
39+
- Key findings from the 750-node mini-mainnet experiments documented in [analysis results](https://github.com/input-output-hk/ouroboros-leios/blob/main/analysis/sims/2025w24/analysis.ipynb) and [summary slides](https://github.com/input-output-hk/ouroboros-leios/blob/main/analysis/sims/2025w24/summary.pdf):
40+
- The 750-node mini-mainnet serves as a suitable replacement for the 10,000-node pseudo mainnet for performance measurements
41+
- Substantial agreement between Haskell and Rust simulations for mini-mainnet scenarios
42+
- Block propagation times under one second, consistent with empirical observations from pooltool.io
43+
- Protocol can support 25 MB/s throughput with 1 Gb/s links before degradation
44+
- Mean transaction time from mempool to ledger approximately 150 seconds
45+
- Achieved 80% disk-space efficiency with ~20% network traffic overhead
46+
- Six-core VM sufficient for peak demand at 300 TPS, with average demand under two cores
47+
- Results from 10,000-node pseudo-mainnet experiments available in [analysis documentation](https://github.com/input-output-hk/ouroboros-leios/blob/main/analysis/sims/2025w23/analysis.ipynb) and [presentation slides](https://github.com/input-output-hk/ouroboros-leios/blob/main/analysis/sims/2025w23/summary.pdf):
48+
- Average transaction lifecycle of 100 seconds from mempool to ledger
49+
- Approximately 80% efficiency for both disk and network usage
50+
- Six CPU cores sufficient for peak load handling even at high TPS rates
51+
- Block propagation time averaged under one second across the large network.

0 commit comments

Comments
 (0)