Experiment network topology.
Four virtual machines.
1 controller & data collector, 3 workers.
Each with 8 CPU cores and 32GB RAM.
Experiment benchmark services.
Two service topologies: chain and branch.
Two service placements: with and without resource contention.
Relationship among CPU core limit, RAM limit, and per-service latency.
Chain services.
Without resource contention.
Relationship among CPU core limit, RAM limit, and per-service latency.
Chain services.
With resource contention.