Using Amazon Global Accelerator to improve cross board request improvement

Know How Guide and Hands on Guide for AWS

Using Amazon Global Accelerator to improve cross board request improvement

  1. Create the lambda funtion echo_hello in eu-west-1 region ```python import json import urllib import boto3 from botocore.vendored import requests

def lambda_handler(event, context): myip = requests.get(‘http://checkip.amazonaws.com’).text.rstrip() print(myip) return { ‘statusCode’: 200, ‘body’: json.dumps(‘Hello from Lambda!’ + myip) }


2. Configure the Lambda functions as a target for an Application Load Balancer
![echo_hello](/aws-is-how/network/aga/media/echo_hello.png)

3. Configure the Amazon Global Accelerator with endpoint to Application Load Balancer
![aga](/aws-is-how/network/aga/media/aga.png)

![aga-lambda](/aws-is-how/network/aga/media/aga-lambda.png)

4. Create the lambda funtion `invokeGlobalLambda` in cn-north-1 region

The [source code](/aws-is-how/network/aga/script/lambda_function.py)

![invokeGlobalLambda](/aws-is-how/network/aga/media/invokeGlobalLambda.png)

```bash
pip install -r requirements.txt --target ./package
cd package
zip -r ../my-deployment-package.zip .
cd ..
zip -g my-deployment-package.zip lambda_function.py

aws lambda update-function-code --function-name invokeGlobalLambda --zip-file fileb://my-deployment-package.zip --region cn-north-1

aws lambda invoke --function-name invokeGlobalLambda \
    --region cn-north-1 --log-type Tail \
    --query 'LogResult' --output text hello-output.txt | base64 --decode
  1. Enable AWS Lambda with AWS X-Ray 
    aws lambda update-function-configuration --function-name invokeGlobalLambda \
--tracing-config Mode=Active --region cn-north-1
  2. Run performance testing with Apache AB ```bash ab -n 10000 -c 10 -s 360 https://myapi.execute-api.cn-north-1.amazonaws.com.cn/dev/invokeGlobalLambda

Concurrency Level: 10 Time taken for tests: 1596.488 seconds Complete requests: 10000 Failed requests: 8400 (Connect: 0, Receive: 0, Length: 8400, Exceptions: 0) Total transferred: 3985579 bytes HTML transferred: 1115579 bytes Requests per second: 6.26 [#/sec] (mean) Time per request: 1596.488 [ms] (mean) Time per request: 159.649 [ms] (mean, across all concurrent requests) Transfer rate: 2.44 [Kbytes/sec] received

Connection Times (ms) min mean[+/-sd] median max Connect: 24 44 80.8 31 2210 Processing: 1347 1549 394.4 1516 16160 Waiting: 1347 1549 394.3 1515 16160 Total: 1373 1594 409.9 1551 16190

Percentage of the requests served within a certain time (ms) 50% 1551 66% 1567 75% 1579 80% 1589 90% 1625 95% 1676 98% 1845 99% 2536 100% 16190 (longest request) ```

  1. Check the invocation performance in X-Ray xray