MruDocs
Simulation › Dusk

Dusk

A Monte Carlo of redundancy policies on hardware that only ever decays. It asks what a shrinking quorum buys, and what it costs, on the same simulated hardware.

Rustno dependenciesdeterministic per seeddusk.mru.space ↗

Run it

cargo run --release                 # the default scenario, about 2 s
cargo run --release -- --spares 2   # add salvaged processors
cargo run --release -- --sweep      # sensitivity, about 30 s
cargo run --release -- --ground     # against TMR with a ground team, about 10 s
cargo run --release --bin charts    # redraw the charts and data.json

Times are for an 8-core laptop. Results depend only on the seed, never on the thread count.

Options

OptionDefaultMeaning
--runs10000Missions per policy
--seed1Base seed
--threadsall coresWorker threads
--horizon-years1000Mission horizon
--nodes3Processors at launch
--spares0Salvaged processors
--shape1.5Weibull shape
--scale-years125Weibull scale
--p-corr0.1Chance a death kills another
--p-upset0.001Bad-result upsets per node-day
--coverage0.99Lone self-check coverage
--selfcheck-throughput0.5Lone output vs a vote

The fault model

Each mission is a day-by-day fault-injection run.

  • Processors die on a Weibull lifetime. Shape 1.5 and a 125-year scale give a median life of about 98 years.
  • Deaths can be correlated. With probability p_corr, a death also kills one other processor that day.
  • Spares are salvaged from instruments as they die, already aged by the same radiation.
  • Upsets corrupt a day's result, at the rate that gets past EDAC and scrubbing.
  • The comparison is paired. Every policy sees the same fault history, so any difference is the policy.

Default results

policy              useful      p10      p50      p90      service   wrong days
                    mean y                                  mean y   mean / run
fixed TMR             99.0     37.6     92.7    168.9         99.0        0.000
standby simplex       83.1     38.4     78.3    134.0        166.4        0.603
shrinking quorum     132.6     64.3    126.6    208.3        166.4        0.246

1.34× the useful work of fixed TMR in total (95% CI 1.33× to 1.35×). Never less on any single mission.

Sweeps and ground support

--sweep varies each parameter the model cannot pin down. Across every row, the gain over fixed TMR stays between 1.11× and 1.68×.

--ground adds a ground team that commands the fallback by hand while support lasts. With 49 years of support, Voyager's so far, the gain is still 1.25×. Answer time barely matters. What matters is whether anyone is still there.

What this does not claim

NoteThe parameters are illustrative, not calibrated. The claim is the shape of the trade, and the sweep shows where it holds.
PreviousThe quorum CLINextMru 2049

Updated 29 Sep 2026 · Edit on GitHub · Questions? Request information