SystemGrid

Load generator

Client requests / second

Effective client request rate: ~120 req/s (response traffic is derived hop-by-hop). Use the field for exact values or the slider for coarse steps (large fan-out scenarios).

Client requests enter through API gateways and User → service edges (split evenly per channel), propagate downstream with capacity-aware forwarding, then responses propagate back toward the client. Stress reflects either direction. Edge animation follows paired load; nodes show req/resp rates and pulse when stressed.

Select a predefined node to edit it in the floating panel above the block, or a traffic edge for flow direction. Hover a node for note, duplicate, or delete.