DBRaven
Start Analysis →

Composite Simulation

Composite Simulation Engine

Chain two simulation engines together. Output signals from the first engine (connection utilization, WAL throughput, queue data delay) are extracted and fed as parameter overrides into the second engine via deterministic transfer functions. Models causal cross-domain interactions that no single engine can capture alone.

Models a read-heavy workload load ramp, extracts peak connection utilization, then feeds that utilization as reduced pool headroom into a DAU growth scaling projection. Shows how existing baseline load stress compounds as users scale.

pool_headroom = pool × (1 − peak_util%) · M/M/1 · N = λW (Little's Law)

Workload Engine Inputs

1–500,000

Steady-state request rate.

1.1–20

Peak traffic as ×baseline.

5–5,000

DB connection pool size (shared).

0.5–5,000

Steady-state query latency.

0–65,536

In-memory cache capacity.

1–65,536

Hot dataset the cache must hold.

Scaling Engine Inputs

1,000–1,000,000,000

Starting daily active users.

0–200

DAU compound monthly growth.

3–36

How many months to project.

0–20

Read replicas for the scaling engine.

No simulation run yet

Select a template, set your parameters, and click “Run Composite Simulation” to see chained engine results.