Tutorial Lengkap AWS Lambda + SageMaker: Serverless ML Inference

Menggabungkan AWS Lambda dengan SageMaker memungkinkan inference ML yang serverless, scalable, dan hemat biaya. Tutorial ini mencakup pattern untuk mengintegrasikan fungsi Lambda dengan endpoint SageMaker untuk aplikasi ML production.

Mengapa Lambda + SageMaker?

• Serverless: Tidak perlu mengelola infrastruktur
• Hemat biaya: Bayar per invocation
• Scalable: Scaling otomatis sesuai demand
• Fleksibel: Multiple pattern integrasi
• Event-driven: Trigger inference dari event AWS apapun

• Real-time API inference
• Trigger batch processing
• Event-driven ML pipelines
• Inference hemat biaya
• Multi-model serving

Prerequisites

pip install boto3 sagemaker

AWS CLI sudah dikonfigurasi
aws configure

Pattern 1: Lambda Memanggil SageMaker Endpoint

1. Deploy SageMaker Endpoint

import sagemaker
from sagemaker.sklearn import SKLearnModel

session = sagemaker.Session()
role = sagemaker.getexecutionrole()

Deploy model
model = SKLearnModel(
    modeldata="s3://bucket/model/model.tar.gz",

    role=role,
    frameworkversion="1.0-1",
    pyversion="py3"

)

predictor = model.deploy(
    initialinstancecount=1,

    instancetype="ml.m5.large",
    endpointname="sklearn-endpoint"

)

# lambdafunction.py
import json
import boto3
import os

Initialize SageMaker runtime client
sagemakerruntime = boto3.client("sagemaker-runtime")


ENDPOINTNAME = os.environ.get("SAGEMAKERENDPOINT", "sklearn-endpoint")


def lambdahandler(event, context):
    """Panggil SageMaker endpoint untuk inference."""
    try:
        # Parse input
        if "body" in event:
            body = json.loads(event["body"])
        else:
            body = event

        features = body.get("features", [])

        # Siapkan payload
        payload = json.dumps({"features": features})

        # Panggil endpoint
        response = sagemakerruntime.invokeendpoint(
            EndpointName=ENDPOINTNAME,

            ContentType="application/json",
            Body=payload
        )

        # Parse response
        result = json.loads(response["Body"].read().decode())

        return {
            "statusCode": 200,
            "headers": {"Content-Type": "application/json"},
            "body": json.dumps({
                "prediction": result,
                "endpoint": ENDPOINTNAME
            })
        }

    except Exception as e:
        return {
            "statusCode": 500,
            "body": json.dumps({"error": str(e)})
        }

3. Deploy Lambda dengan SAM

# template.yaml
AWSTemplateFormatVersion: '2010-09-09'
Transform: AWS::Serverless-2016-10-31

Resources:
  InferenceFunction:
    Type: AWS::Serverless::Function
    Properties:
      Handler: lambdafunction.lambdahandler
      Runtime: python3.9
      Timeout: 30
      MemorySize: 256
      Environment:
        Variables:
          SAGEMAKERENDPOINT: sklearn-endpoint

      Policies:
Version: '2012-10-17'
          Statement:

Effect: Allow
              Action:

sagemaker:InvokeEndpoint
              Resource: '*'

      Events:
        Api:
          Type: Api
          Properties:
            Path: /predict
            Method: post

# Deploy
sam build
sam deploy --guided

# Dockerfile
FROM public.ecr.aws/lambda/python:3.9

Install dependencies
COPY requirements.txt .
RUN pip install -r requirements.txt