AWS Lambda Use Case
In this Career Genese project we have used AWS Lambda to run our own code on Amazon’s servers, so we don’t need to host the code by ourselves. The best part of using it as we don’t have to pay a dime when our Lambda function isn’t being triggered and only need to pay for computation time.
AWS Lambda Use Cases in this project:
- Security Updates
- Automated Backups
- Serverless Website
- File Conversion
- Face Recognition
Attendance System Architecture Diagram Lambda
Description of type of serverless microservice
As there are different microservice available in AWS, we have used AWS Lambda and API gateway as a serverless microservices. The API of a serverless microservice is the central entry point for all client requests. The application logic hides behind a set of programmatic interfaces, typically at REST web services API. This API accepts and processes calls from clients and might implement functionality such as traffic management, request filtering, routing, caching, and authentication and authorization.
AWS Lambda let us run code without provisioning or managing servers. Lambda is highly integrated with API Gateway. The possibility of making a synchronous call from API Gateway to AWS Lambda enables the creation of fully serverless applications and it is described in our documentation.
Total estimated number of events per month:
AWS Lambda triggers are merely actions caused by specific events that will further trigger the lambda function. Any work that triggers the lambda function is an event. And our Lambda functions have approximately 10 thousand events that generate per month. For example, daily user login, Leave requested by the users, daily attendance in entry and exit time, etc.
Description of any challenge or customer objection overcome to drive a lambda adoption:
When we introduce AWS Lambda to our customer they were quite impressed by the service but they told us that they will not be using this service because Lambda is quite complex for the developers who are not familiar with AWS.
The number of Lambda functions deployed.
In this task, we have a sum of 7 lambda capacities, which continue with various activities.
Following are the various list of our function that are used to create this application :
Deployment Patterns: At least 5 functions, Use of best practice deployment application model (SAM)
We had to manage diverse kind of AWS administrations which is recorded beneath:
Lambda function, DynamoDb,API Gateway, Rekognition, Cognito Service, Simple Storage Service, Simple Email Service.Additionally Serverless. Moreover, AWS Serverless Application Model (AWS SAM) is utilized which is an open-source system we have to make utilization of it to manufacture and convey AWS serverless applications. Code Commit, Code Build, Code pipeline are utilized to mechanize programming.
Workload 1: Connection with at least 3 other AWS Services
AWS Lambda functions act as a backend to the application. This application utilizes other AWS services and resources that are organized and manipulated by lambda functions. An application uses DynamoDB tables for storing data and records. Lambda is responsible for reads and writes of data from and to the database.
For storing static files, AWS storage service S3 is used and all the GET and PUT operation is done by Lambda functions. Amazon Rekognition is implemented in Lambda function codes to do image processing tasks. Cognito is used for signup and sign in whereas Amazon SES is used to notify the administrator if an event is triggered in Lambda function.
Workload 2: Use of servers or containers only in supporting roles
Since this application is absolutely serverless, all the application rationale live inside lambda improve the performance, minimize the complexity and avoid of using the recursive code of our function. There is no requirement for any servers or holders in any sort of jobs or assignments in the application.
Workload 3: Serverless workload should use AWS Lambda to handle and process the core data flow for the system
Every single backend rationale that the application depends on is facilitated in AWS Lambda. All the logic behind DynamoDB table read and writes, S3 GET and PUT, Image processing tasks with Rekognition, Cognito Admin User creation and so on resides inside lambda functions we have used.
Workload 4: Automating Infrastructures provisioning and managing is not serverless
AWS Lambda serves client side with appropriate data queries, user registration, and many other application tasks. Automation of Infrastructure provisioning and management is not applicable in this use case.
Solution Characteristics 1: Load/Performance Testing
We utilized a bundle of the hub (npm) named load test to test the API ask for and Lambda Invocations.
- Install-Package Locally with a command: npm install -g loadtest
- Run load test without a header:
Command: ‘loadtest -c 10 –rps 200 https://ginyod5w77.execute-api.ap-south-1.amazonaws.com/prod/openapi?addNewIntern=SendCredentials’
- Run loadtest with HTTP header:
Command: ‘loadtest -c 10 -H token:eyJraWQiOiI4dEY4a1o3Y01pMjZCenRTenY1dGo5b2RtdWZFSU1RcTZXRml0U251cFRNPSIsImFsZyI6IlJTMjU2In0.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.N6n_Le7MYTP7dF1MEcvbHHLE5R1cNvNxVcaG7mJph2CKog43LgLzMSXsOLbip1m64kbO52ovaQM7ATxDzCnb0GqeYdPxZQ09RumoYCqjka3q2sP3Xh4BqMw_zU2AT-Jj7giblklVmQsaXAETmGV3HjyehK2ssEEIurSaklf1EcLgHD0MHoC6Sp41XqFBvS3u4ANOblVjOiZ6eoGs76Aphm0Db4uQ7oXGdSbB6Xj-BCj6mCZ4dIwGBjw62R4JAMGl3pmLTbyARqCn57dn49s6bV_wSfXcjYbKjJSzeJWoeq39ClKYNUV44VkVc6Wkie_bj2N9Af0ESjNLZ6wqBIRZZg –rps 200 https://ginyod5w77.execute-api.ap-south-1.amazonaws.com/prod/companyrequesthandler?param1=Monthly¶m2=October 2018¶m3=ashlesha’
Solution Characteristics 2: IAM Policy Definition:
In this policy, we have utilized diverse IAM approaches to connect it with the IAM Roles to execute necessary tasks in Lambda functions. Using the permissions policy associated with this role, we can grant our Lambda function the permissions that it needs. Different Policies Used are:
- AttendanceCognitoPolicy(Inline Policy)
- AttendaceDynamodbPolicy(Inline Policy)
- ses-full-access(Inline Policy)
- AWSS3DFullAccess(AWS Managed Policy)
- AttendaceRecognitionPolicy(Inline Policy)
- AttendaceS3Policy(Inline Policy)
- AttendaceSESPolicy(Inline Policy)
Solution Characteristics 3: Managing Failed Executions:
A Lambda capacity can fall flat for any of the accompanying reasons:
- The capacity times out while endeavoring to achieve an endpoint.
- The function fails to successfully parse input data.
We have used python boto3 library in our code. How the exemption is taken care of relies on how the Lambda work was invoked. In this task, some of these event sources are set up to invoke a Lambda function synchronously and others invoke it asynchronously. In specific cases, possibly the info parameters might not be right and code rationale may not consent to the information. In such case, lambda distinguishes the special case and returns a fitting message to the customer side.
Solution Characteristics 4: Management of Multiple Microservice functions
- The API of a microservice is the central entry point for all client requests. The application logic hides behind a set of programmatic interfaces, typically a RESTful web services API.
- Since clients of a microservice are served from the closest edge location and get responses either from a cache or a proxy server with optimized connections to the origin, latencies can be significantly reduced. However, microservices running close to each other don’t benefit from a CDN. In some cases, this approach might even add more latency. It is a best practice to implement other caching mechanisms to minimize the latencies.
- Amazon DynamoDB is used to create a database table that can store and retrieve any amount of data and serve any level of request traffic.
- The Server Side Logic is written in Lambda functions, which are independent microservice functions of AWS.
Solution Characteristics 5: VPC vs No-VPC and concurrency
Lambda capacities are not put in any VPC’s as the application doesn’t have to interface to any servers that require VPC Secure Environment. Regarding concurrency limits, the default limit of 1000 concurrent executions is enough for a current use case. We might need a limit increase in the future for Higher Performing Microservices.
Solution Characteristics 6: Monitoring of Alarms and Metrics:
- For observing certain measurements for the majority of our Lambda capacities consequently. We have checked diverse CloudWatch measurements of Lambda by the Name of Function. For every serverless service we run, we care about Errors and Throttles we also want to know if our code is failing for any reason, whether errors in our code or too many concurrent Lambda invocations in our account. We can handle both of these problems with CloudWatch custom metrics. Using the AWS SDK, you can make a put_metric_data() call with a CloudWatch client.
- Name of Different Metrics that we monitored for our Lambda includes:
- Invocation per lambda function
- Duration of execution of lambda function
- Error in a lambda function
- Throttles to manage the flow of execution