Emulator can generate multiple fake Nodes and Containers for the benchmark testing in the Alameda environment.
[global]
emulator_listen_address = ":9200"
emulator_namespace = ""
emulator_web_path = "/metrics"
emulator_label_name = "emulator"
emulator_label_value = "emulator"
emulator_node_name = "emulator"
emulator_node_resource_cpu_cores = 8
emulator_node_resource_memory_bytes = 32000000000
emulator_node_resource_cpu_usage_range = "0"
emulator_node_resource_memory_usage_range = "0"
emulator_node_metadata_template = "/etc/emulator/node.json"
emulator_container_metadata_template = "/etc/emulator/pod.json"
emulator_container_ignore_creation = false
emulator_performance_scrap_seconds = 900
[datahub]
address = "alameda-datahub.alameda.svc:50050"
[datahub."retry-interval"]
default = 3 # second
[containers]
container_count = 100
container_prefix_name = "emulator"
container_cpu_csv_filepath = "/etc/emulator/metric_cpu.csv"
container_memory_csv_filepath = "/etc/emulator/metric_memory.csv"
container_pulled_start_hour = "0"
container_cpu_random = true
container_memory_random = true
container_cpu_random_range = "0.1-0.9"
container_memory_random_range = "0.1-0.9"
container_data_step = 3600
[global]
emulator_listen_address: emulator service daemon listen port is 9200
emulator_node_resource_cpu_cores: Set the fake node cpu cores
emulator_node_resource_memory_bytes: Set the fake node memory bytes
emulator_performance_scrap_seconds: Same as the Prometheus scrape
configuration
[datahub]
address: Config the DataHub service ip and port
[containers]
container_count: Config the container(pod) generate count
container_cpu_csv_filepath: Reference the pod CPU performance metrics file
container_memory_csv_filepath: Reference the pod Memory performance metrics file
container_data_step: Performance data generate time seconds step
make docker-build
Output docker image name is "federatorai-agent:latest"