State as a Service

(still serverless...)

Melbourne Serverless Meetup - Feb 28 2017

Created by Shiva Narayanaswamy / @shivamaan

Evolution

Serverless Reference Architecture

Lambda Explosion

Apps not Functions

Sequence Functions
Data based decision trees
Retry and Error Handling
Try-Catch-Finally
Workflows that run for hours

We need Coordination

Enter State Machines

State Machines in two minutes

AWS Step Functions

Hello World


{  
	”Comment”: “Hello World Example",  
	"StartAt” : "HelloWorld”,  
	"States” : {    
	"HelloWorld” : {      
		"Type” : "Task",      
		"Resource” : "arn:aws:lambda:REGION:ACCOUNT_ID:function:FUNCTION_NAME”,      
		"End” : true    
		}  
	}
}

Seven state types

Task	A single unit of work (Activity or Lambda)
Choice	Branching logic
Parallel	Fork and join data across tasks
Wait	Delay
Fail	Stops execution as a failure
Suceed	Stops execution as a success
Pass	Input to output pass through

Poll or Push

Task State


{  
	”Comment”: “Hello World Example",  
	"StartAt” : "HelloWorld”,  
	"States” : {    
	"HelloWorld” : {      
		"Type” : "Task",      
		"Resource” : "arn:aws:lambda:REGION:ACCOUNT_ID:function:FUNCTION_NAME”,      
		"End” : true    
		}  
	}
}

Task State - Retry Failures


"HelloWorld": {      
	"Type": "Task",      
	…
	"Retry": [        
	{          			
		"ErrorEquals": ["HandledError"],          	
		"IntervalSeconds": 1,          	
		"MaxAttempts": 2,          	
		"BackoffRate": 2.0        
	},
	… ]

Task State - Catch Failures


"HelloWorld": { 
	"Type": "Task",      
	…
	"Catch": [        
	{          
	"ErrorEquals": ["HandledError"],          	
	"Next": "CustomErrorFallback"        
	},        
	{          
	"ErrorEquals": ["States.TaskFailed"],          	
	"Next": "ReservedTypeFallback"        
	… ]       
	{

Task State - Catch Failures

Complex error handling and retries


"X": {
  "Type": "Task",
  "Resource": "arn:aws:states:us-states-1:123456789012:task:X",
  "Next": "Y",
  "Retry": [
    {
      "ErrorEquals": [ "ErrorA", "ErrorB" ],
      "IntervalSeconds": 1,
      "BackoffRate": 2,
      "MaxAttempts": 2
    },
    {
      "ErrorEquals": [ "ErrorC" ],
      "IntervalSeconds": 5
    }
  ],
  "Catch": [
    {
      "ErrorEquals": [ "States.ALL" ],
      "Next": "Z"
    }
  ]
}

Choice State


"ChoiceState": {      
	"Type" : "Choice",      
	"Choices": [        
		{          
			"Variable": "$.foo",          
			"NumericEquals": 1,          
			"Next": "FirstMatchState"        
		},        
		{          
			"Variable": "$.foo",          
			"NumericEquals": 2,          
	… ]

Choice State

Parallel State


"LookupCustomerInfo": {
  "Type": "Parallel",
  "Branches": [
    {
      "StartAt": "LookupAddress",
      "States": {
        "LookupAddress": {
          "Type": "Task",
          "Resource":
            "arn:aws:lambda:us-east-1:123456789012:function:AddressFinder",
          "End": true
        }
      }
    },
    {
      "StartAt": "LookupPhone",
      "States": {
        "LookupPhone": {
          "Type": "Task",
          "Resource":
            "arn:aws:lambda:us-east-1:123456789012:function:PhoneFinder",
          "End": true
        }
      }
    }
  ],
  "Next": "NextState"
},

Wait State


"wait_using_seconds": {      
	"Type": "Wait",      
	"Seconds": 10,      
	"Next": "wait_using_timestamp"    
	},    
"wait_using_timestamp": {      
	"Type": "Wait",      
	"Timestamp": "2015-09-04T01:59:00Z",     
	"Next": "wait_using_timestamp_path"    
	},

Amazon States Language Spec

https://states-language.net/spec.html

https://github.com/awslabs/statelint

Serverless Reference Architecture: Image Recognition and Processing Backend

https://github.com/awslabs/lambda-refarch-imagerecognition

An image is uploaded to the PhotoRepo S3 bucket under the "Incoming/" prefix
The S3 upload event triggers the ImageProcStartExecution Lambda function, which kicks off an execution of the ImageProc state machine in AWS Step Functions, passing in the S3 bucket and object key as input parameters.
The ImageProc state machine has the following sub-steps:

Read the file from S3 and extract image metadata (format, EXIF data, size, etc.)
Based on output from previous step, validate if the file uploaded is a supported file format (png or jpg). If not, throw NotSupportedImageType error and end execution.
Store the extracted metadata in the ImageMetadata DynamoDB table
In parallel, kick off two processes simultaneously:

Call Amazon Rekognition to detect objects in the image file. If detected, store the tags in the ImageMetadata DynamoDB table
Generate a thumbnail and store it under the "Thumbnails/" prefix in the PhotoRepo S3 bucket

Manual Approvals

https://aws.amazon.com/blogs/compute/implementing-serverless-manual-approval-steps-in-aws-step-functions-and-amazon-api-gateway/

Use Cases

Asynchronous data processing (Also think AWS Batch)
Business process workflow (FoodPanda)
Intelligent application delivery pipelines
Event driven workflows
Infrastructure automation (Shell scripts)
Building a business from a great idea - Simon Wardley

Gotchas

(Feature Requests ?!)

AWS Limits (StartExecution API has a limit of 100 and gets throttled to 2 per second)
Cannot edit State Machines
Task Activity is polling only (Hint : Needs SNS or HTTP endpoint to be truly asynchronous)
Activity ARN is potentially guessable
Event Driven State Machines
State Machine based development, testing, operations and tooling
Available in N.Virginia, Ohio, Oregon, Tokyo, Ireland

T H A N K Y O U

Presentation available at https://shivaman.github.io/

Shiva Narayanaswamy / @shivamaan