Sign in to confirm you’re not a bot
This helps protect our community. Learn more

Optimizing AWS Lambda Performance and Cost for Your Serverless Applications - AWS Online Tech Talks

AWS Developers
1.74M subscribers
7.1K views 3 years ago #AWS
Developers using AWS Lambda have a range of options for performance-tuning their functions. Understanding how to get the most from Lambda can improve your application throughput, reduce latency, and lower overall cost. In this tech talk, we cover common misconceptions about cold starts, and the different impact of cold starts between sync, async, and streaming invocations. Learn about some of the newest features that can optimize Lambda performance and reduce cost and see how making good architectural decisions can help improve performance. Learn why memory allocation is so important and how to profile your Lambda functions to find the optimal configuration. We’ll also talk about best practices for Lambda to help you get the most from your serverless applications. Learning Objectives:  …...more
...more
These chapters are auto-generated

Intro

0:00

Anatomy of an AWS Lambda function

1:23

Function lifecycle - worker host

6:41

Measuring with AWS X-Ray

7:38

X-Ray Trace Example

8:40

Three areas of performance

9:27

Function lifecycle - a warm start

10:08

Function lifecycle - a full cold start

10:37

Cold starts - Execution environment

11:02

Lambda + VPC - Major performance improvement

14:36

Cold starts - Static initialization

16:14

Provisioned Concurrency on AWS Lambda

18:41

Function lifecycle: Provisioned Concurrency start

19:29

Function lifecycle: Provisioned Concurrency invoke

20:19

Provisioned Concurrency: Things to know

20:36

Provisioned Concurrency: Configuration

23:18

Provisioned Concurrency: Application Auto Scaling

23:55

Resources allocation

24:53

CPU-bound example

25:43

Meet AWS Lambda Power Tuning

26:56

AWS Lambda Power Tuning (input)

28:13

AWS Lambda Power Tuning (visualization)

31:09

Comparing x86 to Arm/Graviton2

31:46

CPU-bound (numpy: inverting 1500x1500 matrix)

33:29

CPU-bound (prime numbers)

34:07

Network-bound (third-party API call)

34:59

Cost/performance patterns

36:47

Optimized dependency usage: (Node.js SDK & X-Ray)

38:11

Lazy initialization example: (Python & boto3)

38:52

Amazon RDS Proxy

39:40

Optimization best practices (performance/cost)

40:40

Reusing connections with Keep-Alive

42:05

The Lambda invocation model Synchronous Asynchronous Event source mapping request/response

43:28

Comparing sync and async

44:29

Lambda Performance - Summary

45:36
Optimizing AWS Lambda Performance and Cost for Your Serverless Applications - AWS Online Tech Talks
152Likes
7,123Views
2021Oct 25
Developers using AWS Lambda have a range of options for performance-tuning their functions. Understanding how to get the most from Lambda can improve your application throughput, reduce latency, and lower overall cost. In this tech talk, we cover common misconceptions about cold starts, and the different impact of cold starts between sync, async, and streaming invocations. Learn about some of the newest features that can optimize Lambda performance and reduce cost and see how making good architectural decisions can help improve performance. Learn why memory allocation is so important and how to profile your Lambda functions to find the optimal configuration. We’ll also talk about best practices for Lambda to help you get the most from your serverless applications. Learning Objectives:
  • Understand how to allocate memory to optimize Lambda performance.
  • Correct misconceptions about cold starts.
  • Learn how to leverage new features to improve performance and reduce cost.
***To learn more about the services featured in this talk, please visit: https://aws.amazon.com Subscribe to AWS Online Tech Talks On AWS: https://www.youtube.com/@AWSOnlineTec... Follow Amazon Web Services: Official Website: https://aws.amazon.com/what-is-aws Twitch:   / aws   Twitter:   / awsdevelopers   Facebook:   / amazonwebservices   Instagram:   / amazonwebservices   ☁️ AWS Online Tech Talks cover a wide range of topics and expertise levels through technical deep dives, demos, customer examples, and live Q&A with AWS experts. Builders can choose from bite-sized 15-minute sessions, insightful fireside chats, immersive virtual workshops, interactive office hours, or watch on-demand tech talks at your own pace. Join us to fuel your learning journey with AWS. #AWS

Follow along using the transcript.

AWS Developers

1.74M subscribers