Help Center/ ModelArts/ SDK Reference/ Service Management/ Obtaining Service Monitoring Information

Updated on 2024-03-21 GMT+08:00

View PDF

Obtaining Service Monitoring Information

You can use the API to obtain the monitoring information about a service.

Sample Code

In ModelArts notebook, you do not need to enter authentication parameters for session authentication. For details about session authentication of other development environments, see Session Authentication.

Method 1: Obtain the monitoring information of a service based on the service object created in Deploying a Real-Time Service.

      
           from modelarts.session import Session
from modelarts.model import Predictor

session = Session()
predictor_instance = Predictor(session, service_id="your_service_id")
predictor_monitor = predictor_instance.get_service_monitor() 
print(predictor_monitor)

Method 2: Obtain the monitoring information of a service based on the service object returned in Obtaining Service Objects.

      
           from modelarts.session import Session
from modelarts.model import Predictor

session = Session()
predictor_object_list = Predictor.get_service_object_list(session)
predictor_instance = predictor_object_list[0]                
predictor_monitor = predictor_instance.get_service_monitor()
print(predictor_monitor)

Parameters

**Table 1** **get_service_monitor** response parameters
Parameter	Type	Description
service_id	String	Service ID
service_name	String	Service name
monitors	monitor array corresponding to infer_type of a service	Monitoring details

**Table 2** **monitor** parameters corresponding to **real-time**
Parameter	Type	Description
model_id	String	Model ID
model_name	String	Model name
model_version	String	Model version
invocation_times	Number	Total number of model instance calls
failed_times	Number	Number of failed model instance calls
cpu_core_usage	Float	Number of used CPUs
cpu_core_total	Float	Total number of CPUs
cpu_memory_usage	Integer	Used memory, in MB
cpu_memory_total	Integer	Total memory, in MB
gpu_usage	Float	Number of used GPUs
gpu_total	Float	Total number of GPUs