Push Spring Boot Docker images on ECR

On a previous blog we integrated a spring boot application with EC2.
It is one of the most raw forms of deployment that you can have on Amazon Web Services.

On this tutorial we will create a docker image with our application which will be stored to the Amazon EC2 container registry.

You need to have the aws cli tool installed.

We will get as simple as we can with our spring application therefore we will use an example from the official spring source page. The only changes applied will be on the packaging and the application name.

Our application shall be named ecs-deployment

rootProject.name = 'ecs-deployment'

Then we build and run our application

gradle build
gradle bootRun

Now let’s dockerize our application.
First we shall create a Dockerfile that will reside on src/main/docker.

FROM frolvlad/alpine-oraclejdk8
VOLUME /tmp
ADD ecs-deployment-1.0-SNAPSHOT.jar app.jar
RUN sh -c 'touch /app.jar'
ENV JAVA_OPTS=""
ENTRYPOINT [ "sh", "-c", "java $JAVA_OPTS -Djava.security.egd=file:/dev/./urandom -jar /app.jar" ]

Then we should edit our gradle file in order to add the docker dependency, the docker plugin and an extra gradle task in order to create our docker image.

buildscript {
    ...
    dependencies {
        ...
        classpath('se.transmode.gradle:gradle-docker:1.2')
    }
}

...
apply plugin: 'docker'


task buildDocker(type: Docker, dependsOn: build) {
    push = false
    applicationName = jar.baseName
    dockerfile = file('src/main/docker/Dockerfile')
}

And we are ready to build our docker image.

./gradlew build buildDocker

You can also run your docker application from the newly created image.

docker run -p 8080:8080 -t com.gkatzioura.deployment/ecs-deployment:1.0-SNAPSHOT

First step is too create our ecr repository

aws ecr create-repository  --repository-name ecs-deployment

Then let us proceed with our docker registry authentication.

aws ecr get-login

Then run the command given in the output. The login attempt will succeed and your are ready to proceed to push your image.

First tag the image in order to specify the repository that we previously created and then do a docker push.

docker tag {imageid} {aws account id}.dkr.ecr.{aws region}.amazonaws.com/ecs-deployment:1.0-SNAPSHOT
docker push {aws account id}.dkr.ecr.{aws region}.amazonaws.com/ecs-deployment:1.0-SNAPSHOT

And we are done! Our spring boot docker image is deployed on the Amazon EC2 container registry.

You can find the source code on github.

Integrate Spring Boot and EC2 using Cloudformation

On a previous blog we integrated a spring boot application with elastic beanstalk.
The application was a servlet based application responding to requests.

On this tutorial we are going to deploy a spring boot application, which executes some scheduled tasks on an ec2 instance.
The application will be pretty much the same application taken from the official spring guide with some minor differences on packages.

The name of our application will be ec2-deployment

rootProject.name = 'ec2-deployment'

Then we will schedule a task to our spring boot application.

package com.gkatzioura.deployment.task;

import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.scheduling.annotation.Scheduled;
import org.springframework.stereotype.Component;

/**
 * Created by gkatzioura on 12/16/16.
 */
@Component
public class SimpleTask {

    private static final Logger LOGGER = LoggerFactory.getLogger(SimpleTask.class);

    @Scheduled(fixedRate = 5000)
    public void reportCurrentTime() {
        LOGGER.info("This is a simple task on ec2");
    }

}


Next step is to build the application and deploy it to our s3 bucket.

gradle build
aws s3 cp build/libs/ec2-deployment-1.0-SNAPSHOT.jar s3://{your bucket name}/ec2-deployment-1.0-SNAPSHOT.jar 

What comes next is a bootstrapping script in order to run our application once the server is up and running.

#!/usr/bin/env bash
aws s3 cp s3://{bucket with code}/ec2-deployment-1.0-SNAPSHOT.jar /home/ec2-user/ec2-deployment-1.0-SNAPSHOT.jar
sudo yum -y install java-1.8.0
sudo yum -y remove java-1.7.0-openjdk
cd /home/ec2-user/
sudo nohup java -jar ec2-deployment-1.0-SNAPSHOT.jar > ec2dep.log

This script is pretty much self explanatory. We download the application from the bucket we uploaded it previously, we install the java version needed and then we run the application (this script serves us for example purposes, there are certainly many ways to set up you java application running on linux).

Next step would be to proceed to our cloudformation script. Since we will download our application from s3 it is essential to have an IAM policy that will allow us to download items from the s3 bucket we used previously. Therefore we will create a role with the policy needed

"RootRole": {
      "Type": "AWS::IAM::Role",
      "Properties": {
        "AssumeRolePolicyDocument": {
          "Version" : "2012-10-17",
          "Statement": [ {
            "Effect": "Allow",
            "Principal": {
              "Service": [ "ec2.amazonaws.com" ]
            },
            "Action": [ "sts:AssumeRole" ]
          } ]
        },
        "Path": "/",
        "Policies": [ {
          "PolicyName": "root",
          "PolicyDocument": {
            "Version" : "2012-10-17",
            "Statement": [ {
              "Effect": "Allow",
              "Action": [
                "s3:Get*",
                "s3:List*"
              ],
              "Resource": {"Fn::Join" : [ "", [ "arn:aws:s3:::", {"Ref":"SourceCodeBucket"},"/*"] ] }
            } ]
          }
        } ]
      }
    }

Next step is to encode our bootstrapping script to Base64 in order to be able to pass it as user data.
Once the ec2 instance is up and running it will run the shell commands previously specified.

Last step is to create our instance profile and specify the ec2 instance to be launched

    "RootInstanceProfile": {
      "Type": "AWS::IAM::InstanceProfile",
      "Properties": {
        "Path": "/",
        "Roles": [ {
          "Ref": "RootRole"
        } ]
      }
    },
    "Ec2Instance":{
      "Type":"AWS::EC2::Instance",
      "Properties":{
        "ImageId":"ami-9398d3e0",
        "InstanceType":"t2.nano",
        "KeyName":"TestKey",
        "IamInstanceProfile": {"Ref":"RootInstanceProfile"},
"UserData":"IyEvdXNyL2Jpbi9lbnYgYmFzaA0KYXdzIHMzIGNwIHMzOi8ve2J1Y2tldCB3aXRoIGNvZGV9L2VjMi1kZXBsb3ltZW50LTEuMC1TTkFQU0hPVC5qYXIgL2hvbWUvZWMyLXVzZXIvZWMyLWRlcGxveW1lbnQtMS4wLVNOQVBTSE9ULmphcg0Kc3VkbyB5dW0gLXkgaW5zdGFsbCBqYXZhLTEuOC4wDQpzdWRvIHl1bSAteSByZW1vdmUgamF2YS0xLjcuMC1vcGVuamRrDQpjZCAvaG9tZS9lYzItdXNlci8NCnN1ZG8gbm9odXAgamF2YSAtamFyIGVjMi1kZXBsb3ltZW50LTEuMC1TTkFQU0hPVC5qYXIgPiBlYzJkZXAubG9n"
      }
    }

KeyName stands for the ssh key name, in case you want to login to the ec2 instance.

So we are good to go and create our cloudformation stack. You have to add the CAPABILITY_IAM flag.

aws s3 cp ec2spring.template s3://{bucket with templates}/ec2spring.template
aws cloudformation create-stack --stack-name SpringEc2 --parameters ParameterKey=SourceCodeBucket,ParameterValue={bucket with code} --template-url https://s3.amazonaws.com/{bucket with templates}/ec2spring.template --capabilities CAPABILITY_IAM

That’s it. Now you have your spring application up and running on top of an ec2 instance.
You can download the source code from GitHub.

Integrate Spring boot and Elastic Beanstalk using Cloudformation

AWS beanstalk is an amazon web service that does most of the configuration for you and creates an infrastructure suitable for a horizontally scalable application. Instead of Beanstalk the other approach would be to configure load balancers and auto scalling groups, which requires a bit of AWS expertise and time.

On this tutorial we are going to upload a spring boot jar application using amazon elastic beanstalk and a cloud formation bundle.

Less is more therefore we are going to use pretty much the same spring boot application taken from the official Spring guide as a template.

The only change would be to alter the rootProject.name to beanstalk-deployment and some changes on the package structure. Downloading the project from github is sufficient.

Then we can build and run the project

gradlew build
java -jar build/libs/beanstalk-deployment-1.0-SNAPSHOT.jar 

Next step is to upload the application to s3.

aws s3 cp build/libs/beanstalk-deployment-1.0-SNAPSHOT.jar s3://{you bucket name}/beanstalk-deployment-1.0-SNAPSHOT.jar

You need to install the elastic beanstalk client since it helps a lot with most beanstalk operations.

Since we will use Java 8 I would get a list with elastic beanstalk environments in order to retrieve the correct SolutionStackName.

aws elasticbeanstalk list-available-solution-stacks |grep Java 

Based on the results I will use the “64bit Amazon Linux 2016.09 v2.3.0 running Java 8” stackname.

Now we are ready to proceed to our cloudformation script.

We will specify a parameter and this will be the bucket containing the application code

  "Parameters" : {
    "SourceCodeBucket" : {
      "Type" : "String"
    }
  }

Then we will specify the name of the application

    "SpringBootApplication": {
      "Type": "AWS::ElasticBeanstalk::Application",
      "Properties": {
        "Description":"Spring boot and elastic beanstalk"
      }
    }

Next step will be to specify the application version

    "SpringBootApplicationVersion": {
      "Type": "AWS::ElasticBeanstalk::ApplicationVersion",
      "Properties": {
        "ApplicationName":{"Ref":"SpringBootApplication"},
        "SourceBundle": {
                  "S3Bucket": {"Ref":"SourceCodeBucket"},
                  "S3Key": "beanstalk-deployment-1.0-SNAPSHOT.jar"
        }
      }
    }

And then we specify our configuration template.

    "SpringBootBeanStalkConfigurationTemplate": {
      "Type": "AWS::ElasticBeanstalk::ConfigurationTemplate",
      "Properties": {
        "ApplicationName": {"Ref":"SpringBootApplication"},
        "Description":"A display of speed boot application",
        "OptionSettings": [
          {
            "Namespace": "aws:autoscaling:asg",
            "OptionName": "MinSize",
            "Value": "2"
          },
          {
            "Namespace": "aws:autoscaling:asg",
            "OptionName": "MaxSize",
            "Value": "2"
          },
          {
            "Namespace": "aws:elasticbeanstalk:environment",
            "OptionName": "EnvironmentType",
            "Value": "LoadBalanced"
          }
        ],
        "SolutionStackName": "64bit Amazon Linux 2016.09 v2.3.0 running Java 8"
      }
    }

The last step would be to glue the above properties by defining an environment

    "SpringBootBeanstalkEnvironment": {
      "Type": "AWS::ElasticBeanstalk::Environment",
      "Properties": {
        "ApplicationName": {"Ref":"SpringBootApplication"},
        "EnvironmentName":"JavaBeanstalkEnvironment",
        "TemplateName": {"Ref":"SpringBootBeanStalkConfigurationTemplate"},
        "VersionLabel": {"Ref": "SpringBootApplicationVersion"}
      }
    }

Now you are ready to upload your cloudformation template and deploy your beanstalk application

aws s3 cp beanstalkspring.template s3://{bucket with templates}/beanstalkspring.template
aws cloudformation create-stack --stack-name SpringBeanStalk --parameters ParameterKey=SourceCodeBucket,ParameterValue={bucket with code} --template-url https://s3.amazonaws.com/{bucket with templates}/beanstalkspring.template

You can download the full sourcecode and the cloudformation template from Github.

Java on the AWS cloud using Lambda, Api Gateway and CloudFormation

On a previous post we implemented a java based aws lambda function and deployed it using CloudFormation.

Since we have our lambda function set up we will integrate it with a http endpoint using AWS API Gateway.

Amazon API Gateway is a fully managed service that makes it easy for developers to create, publish, maintain, monitor, and secure APIs at any scale. With a few clicks in the AWS Management Console, you can create an API that acts as a “front door” for applications to access data, business logic, or functionality from your back-end services, such as workloads running on Amazon Elastic Compute Cloud (Amazon EC2), code running on AWS Lambda, or any Web application

For this example imagine API gateway as if it is an HTTP Connector.

We will change our original function in order to implement a division.

package com.gkatzioura.deployment.lambda;

import com.amazonaws.services.lambda.runtime.Context;
import com.amazonaws.services.lambda.runtime.RequestHandler;

import java.math.BigDecimal;
import java.util.Map;
import java.util.logging.Logger;

/**
 * Created by gkatzioura on 9/10/2016.
 */
public class RequestFunctionHandler implements RequestHandler<Map<String,String>,String> {

    private static final String NUMERATOR_KEY = "numerator";
    private static final String DENOMINATOR_KEY = "denominator";

    private static final Logger LOGGER = Logger.getLogger(RequestFunctionHandler.class.getName());

    public String handleRequest(Map <String,String> values, Context context) {

        LOGGER.info("Handling request");

        if(!values.containsKey(NUMERATOR_KEY)||!values.containsKey(DENOMINATOR_KEY)) {
            return "You need both numberator and denominator";
        }

        try {
            BigDecimal numerator = new BigDecimal(values.get(NUMERATOR_KEY));
            BigDecimal denominator= new BigDecimal(values.get(DENOMINATOR_KEY));
            return  numerator.divide(denominator).toString();
        } catch (Exception e) {
            return "Please provide valid values";
        }
    }

}

Then we will change our lambda code and update it on s3.

aws s3 cp build/distributions/JavaLambdaDeployment.zip s3://lambda-functions/JavaLambdaDeployment.zip

Next step is to update our CloudFormation template and add the api gateway forwarding requests to our lambda function.

First we have to declare our Rest api

    "AGRA16PAA": {
      "Type": "AWS::ApiGateway::RestApi",
      "Properties": {"Name": "CalculationApi"}
    }

Then we need to add a rest resource. Inside the DependsOn element we can see the id of our rest api. Therefore cloudwatch will create the resource after the rest api has been created.

"AGR2JDQ8": {
      "Type": "AWS::ApiGateway::Resource",
      "Properties": {
        "RestApiId": {"Ref": "AGRA16PAA"},
        "ParentId": {
          "Fn::GetAtt": ["AGRA16PAA","RootResourceId"]
        },
        "PathPart": "divide"
      },
      "DependsOn": [
        "AGRA16PAA"
      ]
    }

Another crucial part is to add a permission in order to be able to invoke our lambda function.

    "LPI6K5": {
      "Type": "AWS::Lambda::Permission",
      "Properties": {
        "Action": "lambda:invokeFunction",
        "FunctionName": {"Fn::GetAtt": ["LF9MBL", "Arn"]},
        "Principal": "apigateway.amazonaws.com",
        "SourceArn": {"Fn::Join": ["",
          ["arn:aws:execute-api:", {"Ref": "AWS::Region"}, ":", {"Ref": "AWS::AccountId"}, ":", {"Ref": "AGRA16PAA"}, "/*"]
        ]}
      }
    }

Last step would be to add the api gateway method in order to be able to invoke our lambda function from the api gateway. Furthermore we will add an api gateway deployment instruction.

"Deployment": {
      "Type": "AWS::ApiGateway::Deployment",
      "Properties": {
        "RestApiId": { "Ref": "AGRA16PAA" },
        "Description": "First Deployment",
        "StageName": "StagingStage"
      },
      "DependsOn" : ["AGM25KFD"]
    },
    "AGM25KFD": {
      "Type": "AWS::ApiGateway::Method",
      "Properties": {
        "AuthorizationType": "NONE",
        "HttpMethod": "POST",
        "ResourceId": {"Ref": "AGR2JDQ8"},
        "RestApiId": {"Ref": "AGRA16PAA"},
        "Integration": {
          "Type": "AWS",
          "IntegrationHttpMethod": "POST",
          "IntegrationResponses": [{"StatusCode": 200}],
          "Uri": {
            "Fn::Join": [
              "",
              [
                "arn:aws:apigateway:",
                {"Ref": "AWS::Region"},
                ":lambda:path/2015-03-31/functions/",
                {"Fn::GetAtt": ["LF9MBL", "Arn"]},
                "/invocations"
              ]
            ]
          }
        },
        "MethodResponses": [{
          "StatusCode": 200
        }]
      }

So we ended up with our new cloudwatch configuration.

{
  "AWSTemplateFormatVersion": "2010-09-09",
  "Resources": {
    "LF9MBL": {
      "Type": "AWS::Lambda::Function",
      "Properties": {
        "Code": {
          "S3Bucket": "lambda-functions",
          "S3Key": "JavaLambdaDeployment.zip"
        },
        "FunctionName": "SimpleRequest",
        "Handler": "com.gkatzioura.deployment.lambda.RequestFunctionHandler",
        "MemorySize": 128,
        "Role": "arn:aws:iam::274402012893:role/lambda_basic_execution",
        "Runtime": "java8"
      }
    },
    "Deployment": {
      "Type": "AWS::ApiGateway::Deployment",
      "Properties": {
        "RestApiId": { "Ref": "AGRA16PAA" },
        "Description": "First Deployment",
        "StageName": "StagingStage"
      },
      "DependsOn" : ["AGM25KFD"]
    },
    "AGM25KFD": {
      "Type": "AWS::ApiGateway::Method",
      "Properties": {
        "AuthorizationType": "NONE",
        "HttpMethod": "POST",
        "ResourceId": {"Ref": "AGR2JDQ8"},
        "RestApiId": {"Ref": "AGRA16PAA"},
        "Integration": {
          "Type": "AWS",
          "IntegrationHttpMethod": "POST",
          "IntegrationResponses": [{"StatusCode": 200}],
          "Uri": {
            "Fn::Join": [
              "",
              [
                "arn:aws:apigateway:",
                {"Ref": "AWS::Region"},
                ":lambda:path/2015-03-31/functions/",
                {"Fn::GetAtt": ["LF9MBL","Arn"]},
                "/invocations"
              ]
            ]
          }
        },
        "MethodResponses": [{"StatusCode": 200}]
      },
      "DependsOn": ["LF9MBL","AGR2JDQ8","LPI6K5"]
    },
    "AGR2JDQ8": {
      "Type": "AWS::ApiGateway::Resource",
      "Properties": {
        "RestApiId": {"Ref": "AGRA16PAA"},
        "ParentId": {
          "Fn::GetAtt": ["AGRA16PAA","RootResourceId"]
        },
        "PathPart": "divide"
      },
      "DependsOn": ["AGRA16PAA"]
    },
    "AGRA16PAA": {
      "Type": "AWS::ApiGateway::RestApi",
      "Properties": {
        "Name": "CalculationApi"
      }
    },
    "LPI6K5": {
      "Type": "AWS::Lambda::Permission",
      "Properties": {
        "Action": "lambda:invokeFunction",
        "FunctionName": {"Fn::GetAtt": ["LF9MBL", "Arn"]},
        "Principal": "apigateway.amazonaws.com",
        "SourceArn": {"Fn::Join": ["",
          ["arn:aws:execute-api:", {"Ref": "AWS::Region"}, ":", {"Ref": "AWS::AccountId"}, ":", {"Ref": "AGRA16PAA"}, "/*"]
        ]}
      }
    }
 }
}

Last but not least, we have to update our previous cloudformation stack.

So we uploaded our latest template

aws s3 cp cloudformationjavalambda2.template s3://cloudformation-templates/cloudformationjavalambda2.template

And all we have to do is to update our stack.

aws cloudformation update-stack --stack-name JavaLambdaStack --template-url https://s3.amazonaws.com/cloudformation-templates/cloudformationjavalambda2.template

Our stack has just been updated.
We can got to our api gateway endpoint and try to issue a post.

curl -H "Content-Type: application/json" -X POST -d '{"numerator":1,"denominator":"2"}' https://{you api gateway endpoint}/StagingStage/divide
"0.5"

You can find the sourcecode on github.

Java on the AWS cloud using Lambda

Amazon Web Services gets more popular by the day. Java is a first class citizen on AWS and it is pretty easy to get started.
Deploying your application is a bit different, but still easy and convenient.

AWS Lambda is a compute service where you can upload your code to AWS Lambda and the service can run the code on your behalf using AWS infrastructure. After you upload your code and create what we call a Lambda function, AWS Lambda takes care of provisioning and managing the servers that you use to run the code.

Actually think of lambda as running a task that needs up to five minutes to finish. In case of simple actions or jobs that are not time consuming, and don’t require a huge framework, AWS lambda is the way to go. Also AWS lambda is great for horizontal scaling.

The most stripped down example would be to create a lambda function that responds to a request.

We shall implement the RequestHandler interface.

package com.gkatzioura.deployment.lambda;

import com.amazonaws.services.lambda.runtime.Context;
import com.amazonaws.services.lambda.runtime.RequestHandler;

import java.util.Map;
import java.util.logging.Logger;

/**
 * Created by gkatzioura on 9/10/2016.
 */
public class RequestFunctionHandler implements RequestHandler<Map<String,String>,String> {

    private static final Logger LOGGER = Logger.getLogger(RequestFunctionHandler.class.getName());

    public String handleRequest(Map <String,String> values, Context context) {

        LOGGER.info("Handling request");

        return "You invoked a lambda function";
    }

}

Somehow RequestHandler is like a controller.

To proceed we will have to create a jar file with the dependencies needed, therefore we will create a custom gradle task

apply plugin: 'java'

repositories {
    mavenCentral()
}

dependencies {
    compile (
            'com.amazonaws:aws-lambda-java-core:1.1.0',
            'com.amazonaws:aws-lambda-java-events:1.1.0'
    )
}

task buildZip(type: Zip) {
    from compileJava
    from processResources
    into('lib') {
        from configurations.runtime
    }
}

build.dependsOn buildZip

Then we should build

gradle build

Now we have to upload our code to our lambda function.

I have a s3 bucket on amazon for lambda functions only. Supposing that our bucket is called lambda-functions (I am pretty sure it is already reserved).
We will use aws cli wherever possible.

aws s3 cp build/distributions/JavaLambdaDeployment.zip s3://lambda-functions/JavaLambdaDeployment.zip

Now instead of creating a lambda function the manual way we are going to do so by creating a cloud formation template.

{
  "AWSTemplateFormatVersion": "2010-09-09",
  "Resources": {
    "LF9MBL": {
      "Type": "AWS::Lambda::Function",
      "Properties": {
        "Code": {
          "S3Bucket": "lambda-functions",
          "S3Key" : "JavaLambdaDeployment.zip",
        },
        "FunctionName": "SimpleRequest",
        "Handler": "com.gkatzioura.deployment.lambda.RequestFunctionHandler",
        "MemorySize": 128,
        "Role":"arn:aws:iam::274402012893:role/lambda_basic_execution",
        "Runtime":"java8"
      },
      "Metadata": {
        "AWS::CloudFormation::Designer": {
          "id": "66b2b325-f19a-4d7d-a7a9-943dd8cd4a5c"
        }
      }
    }
  }
}

Next step is to upload our cloudformation template to an s3 bucket. Personally I use a separate bucket for my templates. Supposing that our bucket is called cloudformation-templates

aws s3 cp cloudformationjavalambda.template s3://cloudformation-templates/cloudformationjavalambda.template

Next step is to create our cloudformation stack using the template specified

aws cloudformation create-stack --stack-name JavaLambdaStack --template-url https://s3.amazonaws.com/cloudformation-templates/cloudformationjavalambda.template

In order to check we shall invoke the lambda function through the amazon cli

aws lambda invoke --invocation-type RequestResponse --function-name SimpleRequest --region eu-west-1 --log-type Tail --payload '{}' outputfile.txt

And the result is the expected

"You invoked a lambda function"

You can find the source code on github.

Scan DynamoDB Items with DynamoDBMapper

Previously we covered how to query a DynamoDB database either using DynamoDBMapper or the low level java api.

Apart from issuing queries, DynamoDB also offers Scan functionality.
What scan does, is fetching all the Items you might have on your DynamoDB Table.
Therefore scan does not require any rules based on our partition key or your global/local secondary indexes.
What scan offers is filtering based on the items already fetched and return specific attributes from the items fetched.

The snippet below issues a scan on the Logins table by filtering items with a lower date.

    public List<Login> scanLogins(Long date) {

        Map<String, String> attributeNames = new HashMap<String, String>();
        attributeNames.put("#timestamp", "timestamp");

        Map<String, AttributeValue> attributeValues = new HashMap<String, AttributeValue>();
        attributeValues.put(":from", new AttributeValue().withN(date.toString()));

        DynamoDBScanExpression dynamoDBScanExpression = new DynamoDBScanExpression()
                .withFilterExpression("#timestamp < :from")
                .withExpressionAttributeNames(attributeNames)
                .withExpressionAttributeValues(attributeValues);

        List<Login> logins = dynamoDBMapper.scan(Login.class, dynamoDBScanExpression);

        return logins;
    }

Another great feature of DynamoDBMapper is parallel scan. Parallel scan divides the scan task among multiple workers, one for each logical segment. The workers process the data in parallel and return the results.
Generally the performance of a scan request depends largely on the number of items stored in a DynamoDB table. Therefore parallel scan might lift some of the performance issues of a scan request, since you have to deal with large amounts of data.

    public List<Login> scanLogins(Long date,Integer workers) {

        Map<String, String> attributeNames = new HashMap<String, String>();
        attributeNames.put("#timestamp", "timestamp");

        Map<String, AttributeValue> attributeValues = new HashMap<String, AttributeValue>();
        attributeValues.put(":from", new AttributeValue().withN(date.toString()));

        DynamoDBScanExpression dynamoDBScanExpression = new DynamoDBScanExpression()
                .withFilterExpression("#timestamp < :from")
                .withExpressionAttributeNames(attributeNames)
                .withExpressionAttributeValues(attributeValues);

        List<Login> logins = dynamoDBMapper.parallelScan(Login.class, dynamoDBScanExpression,workers);

        return logins;
    }

Before using scan to our application we have to take into consideration that scan fetches all table items. Therefore It has a high cost both on charges and performance. Also it might consume your provision capacity.
Generally it is better to stick to queries and avoid scans.

You can find full source code with unit tests on github.

Query DynamoDB Items with DynamoDBMapper

On a previous post we issued queries on a DynamoDB database using the low level java api.

Querying using the DynamoDBMapper is pretty easy.

Issue a query using a hash key is as simple as it gets. The best candidate for a query like this would be the Users table by searching using the email hash key.

    public User getUser(String email) {

        User user = dynamoDBMapper.load(User.class,email);
        return user;
    }

Since we use only hashkey for the Users table, our result would be limited to one.

The load function can also be used for composite keys. Therefore querying for a Logins Table Item would require a hash key and a range key.

    public Login getLogin(String email,Long date) {

        Login login =  dynamoDBMapper.load(Login.class,email,date);
        return login;
    }

Next step is to issue more complex queries using conditions. We will issue a query that will fetch the login attempts between two dates.


 public List<Login> queryLoginsBetween(String email, Long from, Long to) {

        Map<String,String> expressionAttributesNames = new HashMap<>();
        expressionAttributesNames.put("#email","email");
        expressionAttributesNames.put("#timestamp","timestamp");

        Map<String,AttributeValue> expressionAttributeValues = new HashMap<>();
        expressionAttributeValues.put(":emailValue",new AttributeValue().withS(email));
        expressionAttributeValues.put(":from",new AttributeValue().withN(Long.toString(from)));
        expressionAttributeValues.put(":to",new AttributeValue().withN(Long.toString(to)));

        DynamoDBQueryExpression<Login> queryExpression = new DynamoDBQueryExpression<Login>()
                .withKeyConditionExpression("#email = :emailValue and #timestamp BETWEEN :from AND :to ")
                .withExpressionAttributeNames(expressionAttributesNames)
                .withExpressionAttributeValues(expressionAttributeValues);

        return dynamoDBMapper.query(Login.class,queryExpression);
    }

We use DynamoDBQueryExpression, in the same manner that we used it in the low level api.
The main difference is that we do not have to handle the paging at all. DynamoDBMapper will map the DynamoDB items to objects but also it will return a “lazy-loaded” collection. It initially returns only one page of results, and then makes a service call for the next page if needed.

Last but not least querying on indexes is one of the basic actions. It is the same routine either for local or global secondary indexes.
Keep in mind that the results fetched, depend on the projection type we specified once creating the Table. In our case the projection type is for all fields.

   public Supervisor getSupervisor(String company,String factory) {

        Map<String,String> expressionAttributesNames = new HashMap<>();
        expressionAttributesNames.put("#company","company");
        expressionAttributesNames.put("#factory","factory");

        Map<String,AttributeValue> expressionAttributeValues = new HashMap<>();
        expressionAttributeValues.put(":company",new AttributeValue().withS(company));
        expressionAttributeValues.put(":factory",new AttributeValue().withS(factory));

        DynamoDBQueryExpression<Supervisor> dynamoDBQueryExpression = new DynamoDBQueryExpression<Supervisor>()
                .withIndexName("FactoryIndex")
                .withKeyConditionExpression("#company = :company and #factory = :factory ")
                .withExpressionAttributeNames(expressionAttributesNames)
                .withExpressionAttributeValues(expressionAttributeValues)
                .withConsistentRead(false);

        List<Supervisor> supervisor = dynamoDBMapper.query(Supervisor.class,dynamoDBQueryExpression);

        if(supervisor.size()>0) {
            return supervisor.get(0);
        } else {
            return null;
        }
    }

Pay extra attention to the fact that consistent read is set to false. DynamoDBQueryExpression uses by defaut consistent reads. When using a global secondary index you cannot issue a consistent read.

You can find full source code with unit tests on github.