pallishreeb / aws-lamda-practical

What is AWS Lamda?

• It is serverless compute service

• Run code without managing servers (AWS will take care of your server , where you will deploy , which machine what configuration and all)

-And the main part is run the code whenever you need only

-It allows running your code without provisioning or managing servers

-You pay for what you use.

What are some of its advantages over other cloud services like EC2, ECS and Fargate ?

• You don't have to worry about scaling up/down instances as it scales automatically based on demands
• No maintenance required
• Cost effective
• Highly scalable

How does Lambda work underneath?

Lambda runs your code in response to triggers such as HTTP requests via Amazon API Gateway, modifications to objects in Amazon S3 buckets, database operations

• The lambda function runs in a container that gets created when we create our first lamdba function
• This container contains all the dependencies needed by this particular function
• When we invoke the lambda function, an event object will be passed which can contain any data related to the invocation request (e.g., API )
• Whenever an event triggers, it executes the relevant code within the function
• When the execution completes successfully then the response is sent back to the caller
• If there is any error during the process, retries can be done for certain types of errors upto three times
• After maximum attempts, the invocation fails and no further attempt will be made
• There may also be cases where the timeout period expires before the function finishes executing
• In such case, the request will fail with a time out exception
• We can configure how long should the lambda wait until timing out using configuration settings
• For example if we set the value to 10 seconds,then after 10 seconds the lambda will stop waiting and return a failure message

When should you deploy to lamda

• event based triggers
• Batch processing jobs (for example: data transformation)
• Real time file processing (like real time image uploads being processed)
• Need one function to quickly develop and deploy
• For backend utilities which do not require constant CPU usage
• To reduce costs, if the application includes resources that are only used infrequently or are expensive to run.

Step 1 - Create a new lambda function Step 2- Choose runtime environment Step 3- Write the code Step 4- Configure the trigger Step 5- Deploy