Simulation: Getting Started with

AWS Lambda
Simulation overview
In this simulation, you create and configure an AWS Lambda function to be initiated
when an image is uploaded to an Amazon Simple Storage Service (Amazon S3)
bucket. You review related logs to determine how the function is performing. You
modify the function configuration to optimize for performance.

Objectives
After completing this simulation, you should be able to do the following:

• Create a Lambda function.

• Configure an Amazon S3 trigger on a Lambda function.
• Observe logs and metrics for a Lambda function to determine performance.
• Size and configure a Lambda function for optimal performance.

Duration
This simulation requires approximately 45 minutes to complete.

Task 1: Creating a Lambda function

1. On the AWS Management Console, in the Search box, enter Lambda, and
then press Enter on your keyboard.
2. From the search results, choose Lambda.
3. Choose Create function.
4. For the Function name, enter resize_image, and then press Enter on your
keyboard.
5. From the Runtime dropdown list, choose Python 3.9.
6. Expand the Change default execution role section.
7. Choose Use an existing role.
8. From the Existing role dropdown list, choose ResizeImageLambdaRole.
9. Choose Create function.
10. Select the scroll bar to scroll to the Layers section.
11. Choose Add a layer.
12. Choose Custom layers.
13. From the Custom layers dropdown list, choose PillowPythonLambdaLayer.
14. From the Version dropdown list, choose the available option.
15. Choose Add.
16. Select the scroll bar to scroll to the Code source section.
17. Press Shift+S on your keyboard to switch to a virtual version of these
simulation instructions.

18. Press Shift+C to select the following code in the simulation instructions.

import boto3
import os
import sys
import uuid
from urllib.parse import unquote_plus
from PIL import Image
import PIL.Image

s3_client = boto3.client('s3')

def resize_image(image_path, resized_path):

with Image.open(image_path) as image:
image.thumbnail(tuple(x / 2 for x in image.size))
image.save(resized_path)

def lambda_handler(event, context):

print('Begin resizing image')
for record in event['Records']:
bucket = record['s3']['bucket']['name']
key = unquote_plus(record['s3']['object']['key'])
print(bucket)
print(key)
tmpkey = key.replace('/', '')
download_path = '/tmp/{}{}'.format(uuid.uuid4(), tmpkey)
upload_path = '/tmp/resized-{}'.format(tmpkey)
s3_client.download_file(bucket, key, download_path)
resize_image(download_path, upload_path)
s3_client.upload_file(upload_path, os.environ['RESIZED_BUCKET'], key)
print('Resizing image complete')

19. Choose (right-click) the selected text to open the context menu, and copy the
code.
20. Press Shift+S on your keyboard to switch back to the AWS Management
Console.
21. Press Shift+C on your keyboard to copy the text in the Code source section.
 22. Choose (right-click) the selected text to open the context menu to paste and
replace the default source code in the lambda_function.py file.
23. To deploy your Lambda function, choose Deploy.
24. Choose the Configuration tab.
25. Choose Edit.
26. For Memory, enter 512 MB, and then press Enter on your keyboard.
27. Choose Save.
28. In the Search box, enter S3, and then press Enter on your keyboard.
29. From the search results, choose S3.
30. From the list of buckets, open the bucket with resizedbucket as part of the
title.
31. Press Shift+C on your keyboard to select the name of the bucket, and then
open (right-click) the context menu to copy its name.

You will use it as the value later.

32. In the Search box, enter Lambda, and then press Enter on your keyboard.
33. From the search results, choose Lambda.
34. Choose the resize_image function.
35. On the Configuration tab, choose Environment variables.
36. In Environment variables section, choose Edit.
37. Choose Add environment variable.
38. For Key, enter RESIZED_BUCKET, and press Enter on your keyboard.
39. For Value, open (right-click) the context menu, and paste in the value you
retrieved.
40. Choose Save.

You have created a Lambda function.

41. Choose Continue to move on to task 2.

Task 2: Configuring an Amazon S3 trigger to

invoke a Lambda function
In this task, you configure an Amazon S3 trigger on an existing S3 bucket and your
Lambda function. The Lambda function resizes images and places them in another
bucket.

42. In the Function overview section of the Lambda console, choose Add trigger.
43. In the Trigger configuration section, choose S3 from the dropdown list.
 44. For Bucket, choose the bucket with original in the name.
45. For Event types, choose All object create events if it is not already selected.
46. For Recursive invocation, select the check box to acknowledge the statement
47. Choose Add.
48. Select the scroll bar to scroll to the bottom of the page.
49. In the Triggers section, for the S3 trigger that was just created, expand Details.
50. Verify that the Bucket arn has originalbucket in the name and that Event
types contains s3-ObjectCreated.

You have configured your Lambda function to be initiated when a new object is
uploaded to the S3 bucket.

51. Choose Continue to move on to task 3.

Task 3: Uploading an image to the Amazon S3

bucket
In this task, you upload an image file to your bucket.

52. Choose (right-click) the link to open the context menu, choose Save Link As…,
and then choose Save to download the file to your computer:

• large-image.jpg

53. Press Shift+S on your keyboard to return to the AWS Management Console.
54. On the AWS Management Console, in the Search box, enter S3, and then press
Enter on your keyboard.
55. From the search results, choose S3.
56. Choose the link for the bucket that has original in the name.
57. Choose Upload.
58. Choose Add files.
59. Choose the file that you downloaded, and choose Open.
60. Choose Upload.

Your file is uploaded to the bucket. Notice the file size. The size of the file is around
4.9 MB.

61. Choose Close.

When your Lambda function runs correctly, the image file that you uploaded is
reduced in size and placed in the S3 bucket that you specified when you set the
environment variable for RESIZED_BUCKET.

62. To return to the Buckets section on the Amazon S3 console, choose Buckets.
63. Choose the link for the bucket that has resized in the name.

Notice the file size. It's significantly reduced from the original size of 4.9 MB.

When you uploaded the image file, Amazon S3 initiated the resize_image Lambda
function. Review the logs to see how your function performed.

64. In the Search box, enter Lambda, and then press Enter on your keyboard.
65. From the search results, choose Lambda.
66. Choose the link for the resize_image function.
67. Choose the Monitor tab.
68. Select the scroll bar to scroll to CloudWatch Logs.
69. In the Recent invocations table, choose the most recent row (row 1) to expand
the details.

Notice the metrics that are recorded for each function invocation. In the actual AWS
Management Console, you might have to wait for up to 1 minute for the data to be
updated.

The DurationInMS column tells you how long your function ran for this invocation.

The first time that your Lambda function is invoked, the Lambda execution
environment has to download your code and start a new execution environment. This
process is called a cold start. The @initDuration metric in the Recent invocations
details signifies the cold start time.

Note the MemorySetInMB column. The amount of memory that's available to your
Lambda function can be adjusted to affect the performance of your Lambda function.

The amount of memory also determines the amount of virtual CPU available to a
function. Adding more memory proportionally increases the amount of CPU, which
increases the overall computational power available. If a function is CPU-, network-,
or memory-bound, then changing the memory setting can dramatically improve its
performance.

Notice how long it took your function to run. It could be faster. Adjust your Lambda
function so that it runs faster.

70. Choose Continue to move on to task 4.

Task 4: Optimizing Lambda function memory
for performance
In this task, you first adjust the memory to 1024 MB.

71. Choose the Configuration tab.

72. Choose General configuration.
73. Choose Edit.
74. For Memory, enter 1024 MB, and then press Enter on your keyboard.
75. Choose Save.
76. In the Search box, enter S3, and then press Enter on your keyboard.
77. From the search results, choose S3.
78. Choose the link for the bucket that has original in the name.
79. Choose Upload.
80. Choose Add files.
81. Choose the file that you downloaded, and chose Open.
82. Choose Upload.

Your file is uploaded to the bucket.

83. Choose Close.

84. In the Search box, enter Lambda, and then press Enter on your keyboard.
85. From the search results, choose Lambda.
86. Choose the link for the resize_image function.
87. Choose the Monitor tab.
88. Select the scroll bar to scroll to CloudWatch Logs.
89. In the Recent invocations table, choose the most recent row (row 1) to expand
the details.

Note how long it took your function to run.

90. Choose Continue to try the next memory adjustment.

Next, you adjust the memory to 2048 MB.

91. Choose the Configuration tab.

92. Choose Edit.
93. For Memory, enter 2048 MB, and then press Enter on your keyboard.
94. Choose Save.
95. In the Search box, enter S3, and then press Enter on your keyboard.
96. From the search results, choose S3.
 97. Choose the link for the bucket that has original in the name.
98. Choose Upload.
99. Choose Add files.
100. Choose the file that you downloaded, and choose Open.
101. Choose Upload.
102. Choose Close.
103. In the Search box, enter Lambda, and then press Enter on your keyboard.
104. From the search results, choose Lambda.
105. Choose the link for the resize_image function.
106. Choose the Monitor tab.
107. Select the scroll bar to scroll to CloudWatch Logs.
108. In the Recent invocations table, choose the most recent row (row 1) to expand
the details.

Note how long it took your function to run.

109. Choose Continue to try the next memory adjustment.

Next, you adjust the memory to 3008 MB.

110. Choose the Configuration tab.

111. Choose Edit.
112. For Memory, enter 3008 MB, and then press Enter on your keyboard.
113. Choose Save.
114. In the Search box, enter S3, and then press Enter on your keyboard.
115. From the search results, choose S3.
116. Choose the link for the bucket that has original in the name.
117. Choose Upload.
118. Choose Add files.
119. Choose the file that you downloaded, and choose Open.
120. Choose Upload.
121. Choose Close.
122. In the Search box, enter Lambda, and then press Enter on your keyboard.
123. From the search results, choose Lambda.
124. Choose the link for the resize_image function.
125. Choose the Monitor tab.
126. Select the scroll bar to scroll to CloudWatch Logs.
127. In the Recent invocations table, choose the most recent row (row 1) to expand
the details.

Notice how long it took your function to run. Your function now runs in
approximately 500 milliseconds with 3008 MB of memory and an image that is 5 MB.
Summary
In this simulation, you created a Lambda function and configured an Amazon S3
trigger on that Lambda function. You uploaded an image to an S3 bucket, which
initiated the Lambda function to resize the image and place it in another bucket. After
reviewing the logs to see performance metrics, you optimized the Lambda function
by increasing the available memory for the function.

Excellent work!

Simulation complete
Congratulations! You have completed the simulation.

Your feedback is welcome and appreciated.

If you would like to share any suggestions or corrections, provide the details in the
AWS Training and Certification Contact Form.

©2024 Amazon Web Services, Inc. or its affiliates. All rights reserved. This work may not
be reproduced or redistributed, in whole or in part, without prior written permission
from Amazon Web Services, Inc. Commercial copying, lending, or selling is prohibited.