Why did it take so long to create endpoint with AWS Sagemaker using Boto3?

user_1673784 0
Viewed 373

It took 45 minutes to create my endpoint from the stored endpoint configuration. (I tested it and it works too). This is the first time that I've used boto3 to do this, whereas previously I just used the Sagemaker web GUI to create an endpoint from endpoint configuration. Suggestions to my code are appreciated:

import boto3

sagemaker_client = boto3.client('sagemaker')

response = sagemaker_client.create_endpoint(
    EndpointName='sagemaker-tensorflow-x',
    EndpointConfigName='sagemaker-tensorflow-x'
)

Note: I've replaced the last part of my endpoint name with x.

amazon-sagemaker boto3 tensorflow amazon-web-services python
1 Answers

AWS has currently issues with Sagemaker:

Increased Error Rates and Latencies for Multiple API operations

5:33 PM PDT We are investigating increased error rates and latencies for CreateTrainingJob, CreateHyperParameterTuningJob, and CreateEndpoint API operations in the US-EAST-1 Region. Previously created jobs and endpoints are unaffected.

6:04 PM PDT We are continuing to investigate increased error rates and latencies for CreateTrainingJob, CreateHyperParameterTuningJob, and CreateEndpoint API operations in the US-EAST-1 Region. Previously created jobs and endpoints are unaffected.

enter image description here

user_2488230
Related