Project setup:
- Create and initialize S3 Bucket to store the results
- Create SQS FIFO queue to send the metrics from the producer to the consumer 
- Create consumer Lambda function
- Configure AWS CLI to use an account with permission to write to the queue

S3 Setup:
- Create the bucket with default settings and name of your choice
- Initialize the bucket with two (ex. faulty-metrics, memory-consumption-metrics) empty files,
    which are going to store the results

SQS Setup:
- Create an SQS queue with the name of your choice;
    In the "Details" section select FIFO instead of Standard queue type;
    In the "Configuration" section modify "Message retention period" to 1 minute;
    In the "Access Policy" section select "Advanced" and delete the content to specify empty policy

Lambda setup:
- Create a new Lambda function with the name of your choice;
    Select "Node.js 18.x" runtime (or other of your choice, which supports the consumer's features)
- Edit the newly created function's execution role by switching to "Configuration" tab,
    selecting "Permissions" from the menu on the left,
    clicking the role's name at the top of the "Execution Role" section,
    expanding the policy in the "Permission policies" section of the newly opened page,
    clicking "Edit" of the newly expanded section,
    adding following data to the policy's statement:
    {
        "Effect": "Allow",
        "Action": [
            "sqs:ReceiveMessage",
            "sqs:DeleteMessage",
            "sqs:GetQueueAttributes"
        ],
        "Resource": [
            "<your queue's ARN>"
        ]
    },
    {
        "Effect": "Allow",
        "Action": [
            "s3:PutObject",
            "s3:PutObjectAcl",
            "s3:GetObject",
            "s3:GetObjectAcl",
            "s3:AbortMultipartUpload"
        ],
        "Resource": [
            "<your bucket's ARN>",
            "<your bucket's ARN>/*"
        ]
    }
- Add an SQS trigger, by pressing "Add trigger" button and specifying your SQS queue
- Add a lambda layer containing AWS SDK, matching previously specified Node.js runtime
- Edit the function's code by pasting the "consumer.js" content, or uploading the file
- Edit the Environmental Variables in "Configuration" > "Environmental variables" section,
    specifying following variables, according to your S3 Bucket setup:
    BUCKET_NAME	<your bucket's name>
    FAULTY_METRICS_OBJECT_KEY <name of the object, to store faulty results>
    MEMORY_CONSUMPTION_METRICS_OBJECT_KEY <name of the object, to store the metrics>

CLI Setup
- Install AWS CLI and configure it ("aws configure") by specifying the credentials
    of the account used by the producer
- Grant the account permissions to send messages to the queue,
    for examply by editing its IAM policy. Ex.:
    {
        "Version": "2012-10-17",
        "Statement": [
            {
                "Sid": "sid1",
                "Effect": "Allow",
                "Action": "sqs:SendMessage",
                "Resource": <your queue's ARN>
            }
        ]
    }


Once the project is set up, run the code by executing "node producer.js".
The bucket's objects will get updated with the results

Improvement ideas:
-Locking the S3 file with DynamoDB, to prevent concurrent access
-More precise data validation with more descriptive errors
-Currently every record is being stored in the file;
    These could be aggregated to provide smaller granularity and reduce S3 space consumption