AWS Lambda
- AWS Lambda is a fully managed service from AWS that offers Functions as a Service (FaaS).
- Function is a piece of code that we want to execute to achieve functionality which is often triggered by an event.
- Function can be written in many different languages
- Node js
- Python
- Ruby
- Java
- Go
- .NET Core
- Developers are responsible for writing, testing and deploying the code to the AWS lambda
- When deploying aws lambda function we get options on how much memory should be allocated to the function, CPU share.
- The minimum amount of memory you can allocate is 128 MB, this goes all the way up 3008 MB
- Once we deploy the function, functions are invoked by something called as events
- Lets try to create our first AWS lambda function.
- We can create lambda functions and then to trigger them we need to configure triggers
Pricing model
- With AWS Lambda we pay only for the amount of time the code is executing.
- Each AWS Lambda function is charged per request based on the combination of following
- How long function executes (rounded to the nearest 100 ms)
- The allocated memory that function was configured with.
- AWS calls this gigabyte-seconds or GB-sec.
- Refer Here for the offical pricing and examples from AWS
- With free tier account we can do 1 million invocations per account per mont (or the equivelent of 400000 GB-Secs) for free
Lambda Foundational concepts
- Functions: This is a resource to run our code in AWS Lambda
- Triggers: A trigger is a resource or configuration that invokes AWS Lambda function. Triggers can include AWS services that you can configure to invoke a function.
- Event: An Even is a JSON Formatted document that contains data for Lambda function to process.
- Execution Environment: An execution environment provides isolated runtime environment for lambda function.
- Deployment package: We can deploy lambda function function code using a deployment package. Lambda supports two types of deployment packages
- A .zip file that contains the function code & its dependencies
- Container image
- Runtime: The runtime provides language specific environment that runs in an execution environment
Ephemeral state
- What happens when the code finishes executing in Lambda?
- The function runtime is terminated
- This means access to the memory state and any temporary files that were created on the filesystem etc or in short everything is gon
- So if we need to maintain or keep tarck of state, then it should be saved external to the function such as AWS DynamoDB
- Functions have finited execution, Any function has a hardlimit of 15 minutes i.e when Lambda function execution takes more than 15 minutes it will be terminated
Service Limits
- For concurrency, there is a soft limit of 1000 executions at a time (This can be raised by creating support request)
- Size of environmental variables can’t be greater than 4 KB
- Size of invocation payload can be no greater that 6 MB for synchronous requests and 256 KB for asynchronous requests.
Use Cases
- AWS Lambda enables event-driven architectures
- Some of the popular use cases of lambda are
- Backend-compute:
- Data Processing
- Chatbots
- Alexa Skills
- Operations and automation
Execution Policies
- Lambda functions executing on AWS might require to access other AWS services like ec2, s3 or others. For this to work we need to give permissions
- IN AWS giving a permission to an AWS service to access other AWS service is called as IAM Role.
- To create an execution role follow the steps as mentioned below
- Navigate to IAM select roles and click on create role
- Navigate to IAM select roles and click on create role
