Live System

Turn up the traffic. Find the bottleneck. Bring a service back.

Traffic controls

Incoming traffic 400 req/s Cache hit rate 75%

Running. Simulation time: 0.0s.

Edge appearance

Animation pattern Base speed 60 units/s Pattern size 5 Gap 22

Pattern color Line color

Label background Background opacity 80% Scale speed with traffic Reverse visual direction Freeze animation only

Appearance does not change the simulation. Color bands alternate between the line and pattern colors.

System snapshot

Completing now: 0 req/s
Queued now: 0
Completed total: 0
Dropped total: 0

Try this

Keep traffic at 400 req/s and lower cache hits to 25%. The database receives 300 req/s, but can serve only 180. Its queue fills, then requests drop.

Recover by raising the cache hit rate. Or take API / A offline and watch the remaining instance absorb the traffic.

Inspired by Breakscale. This is a deterministic fluid queue model, not a discrete-event simulator or production benchmark. It does not model latency, retries, or network delay.

Request pathDrag nodes / scroll to zoom

ClientsIdle

Incoming requests

0 req/s

Load balancerIdle

Route to healthy instances

0 req/s

API / AIdle

300 req/s capacity

0 req/s

API / BIdle

300 req/s capacity

0 req/s

CacheIdle

Hits complete; misses go to storage

0 req/s

DatabaseIdle

180 req/s capacity

0 req/s