Customer challenges
1
Simulation workloads are bursty — heavy at simulation peaks, idle otherwise — so dedicated GPU sits under-used between runs.
Customer benefits
Access to full-network compute at simulation peaks
Resources released to other units off-peak
Higher twin throughput without dedicated idle GPU
Key features
Simulation-Demand-Aware Scheduling
Recognizes simulation and digital-twin workloads and schedules GPU resources according to their demand.
Supports Smart City / Factory / Airport / Port / Energy twins
Runs digital twins for smart city, factory, airport, port, and energy environments on the same fabric.
Draws network-wide compute at simulation peak, releases it off-peak
Draws compute from across the network when simulation demand peaks and releases it off-peak.
Before & after
| Before | After — with Linker AI Nexa |
|---|---|
| Before Dedicated GPU per twin lab, idle between simulation runs. | After — with Linker AI Nexa Simulation demand draws network-wide compute at peak and releases it off-peak. |