Serverless
Microservices
Microservices
- An architecture that is loosely coupled
- An architectural and organizational approach to software development
- Software is composed of small independent services
- Communicate over well-defined APIs
- Owned by small, self-contained teams
Microservicce: SW System을 작은 독립적인 서비스들로 나누는 아키텍처 스타일로 여러 모듈로 구성된 작은 서비스들이 독립적으로 배포, 확장, 관리될 수 있도록 설계한다.
Tightly vs Loosely Coupled
- Tightly Coupled: monolithic architecture
- Loosely Coupled: microservices architecture
- Microservices is better approach: more durable, reliable, and manageable
Downside of Monoliths
- All processes are tightly coupled and run as a single service.
- Adding or improving a monolithic application's features becomes more complex as the code base grows.
- Dependent and tightly coupled processes increase the impact of a single process failure.
Benefits of Microservices
- Use public APIs: Microservices communicate over APIs -> Hide your data and document of your APIs
- Use the right tool for the job: built for explicit functionality
- Secure your services: provide security at every step
- Be a good citizen: 각 microservice가 책임을 가지고 서로 협력해야 함
- Company transformation: shift requires transformation in the organization and communication mechanism btw teams
- Automate everything
Serverless
Serverless
- With serverless computing, you can build and run applications and services without thinking about servers.
- Developers can focus on their code product not worrying about managing and operating servers or runtimes
- Serverless technologies can complement traditional application models.
- You can build nearly any type of application or backend service
Microservices and Serverless Architecture
- Microservice: method of building your architecture
- Serverless: service that's used to build your architcture
마이크로서비스는 소프트웨어 아키텍처의 한 방식이며, 서버리스 아키텍처는 그 방법 중 하나를 실현하기 위해 사용되는 특정한 서비스나 플랫폼
Benefits of Serverless
- No server management
- Pay-for-value services: you never pay for over-provisioning
- Continuous scaling: most serverless services can be configured to scale automataically
- High availability
Event-Driven Architecture
Event-Driven Architecture
- Uses events to initiate actions and communication btw decoupled services.
- When event occurs, the information is published for other services to consume it.
- In event-driven architecture, events are the primary mechanism for sharing information across services
Benefits of Event-Driven Architecture
- Scale and fail independently
- by decoupling your services, they are only aware of the event router, not each other
- if one service has a failure, the rest will keep running
- Audit with ease
- event router acts as a centralized location to audit
- Develop with agility
- event router will automatically filter and push events to customers
- Cut costs
- on-demand -> not paying for continuous polling to check for an event
AWS Lambda
AWS Lambda
AWS Lambda
- serverless, event-driven compute service
- can run code
- Challenge: you need coding experience and expertise with:
- Linux OS commands and concepts
- Familiarity with AWS IAM
- Cloud concepts and IP networking
- Distributed computing concepts
Design Best Practices
- Separate business logic
- Modulara functions
- Treat functions as stateless
- Only include what you need
AWS Lambda Function
Three Ways to Build an AWS Lambda Function
- Author from scratch
- Hello World examples include just enough code for your choosen language to get started
- Use a blueprint
- bluprints are language-specific samples that include function code and settings for a runtime and trigger combination
- Container image
- you can select container image to deploy for your function when creating a function
- containers are a method of OS virtualization taht allow you to run an application and its dependencies in resource-isolated processes
Function Basic Information
- Basic components of an AWS Lambda function include the function name, runtime, and permissions.
Handler Method
- Lambda function handler is the method in your function code that processes events
- Handler Method takes two objects
- Event object: all of the data and metadata Lambda function needs to drive its logic
- Context object: allow your function code to interact with the Lambda execution environment
Configuring Options
Configuring Memory and Timeout
- Lambda allocates CPU powers in proportion to the amount of memory configured
- Memory is the amount of memory available to your Lambda function at runtime
- Lambda runs your code for a set amount of time before timing out
- Timeout is the maximum amount of time in seconds that a Lambda function can run
Configuring Concurrency and Scaling
- Concurrency: number of requests your function can handle at the same time
- for each concurrent request, Lambda provisions a separate instance of your execution environment
- Reserved Concurrency: guarantees the maximum number of cconcurrent instances
- Provisioned Concurrency: initializes a requested number of execution environments instances so that they are prepared to respond imediately to your function's invocations
- Lambda automatically handles scaling the number of execution environments
Choosing a Trigger
- You can configure other AWS services to trigger your function each time a specified event occurs
Destination
- Destination can send records of asynchronous invocations to other services
- Can configure separete destinations for failed events and successed events
- Can configure destinations on a function, a version, or an alias
- With destinations, you can address errors and successes without needing to write more code
Types of Monitoring Graphs
- Invocations: number of times that your function code run
- Duration: amount of time that your function code spends processing an event
- Errors: number of invocations that result in a error
- Throttles: number of times that a process failed because of concurrency limits
- IteratorAge: the age of the last record in the event
- Concurrent executions: number of function instances that are processing events
Amazon CloudWatch Lambda Insights
Amazon CloudWatch Lambda Insights
- monitoring and troubleshooting solution for serverless applications running on Lambda
- Lambda Insights collect, aggregate, summarize system-level metrics and diagonostic information such as cold starts and Lambda worker shutdowns
Lambda Insights Dashboard
- Multi-function
- aggregates the runtime metrics for the Lambda functions in the concurrent AWS account and Region
- identify over and under utilized Lambda functions
- Single-function
- shows the available runtime metrics for a single Lambda function
- troubleshoot invidiual requests
'CS > 클라우드컴퓨팅' 카테고리의 다른 글
| 12, 13. Virtualization (0) | 2023.12.07 |
|---|---|
| 11. Serverless II (0) | 2023.12.05 |
| 9. Security (0) | 2023.12.03 |
| Lec 08: Cloud Operations (0) | 2023.12.02 |
| Lec 07: Database (0) | 2023.10.22 |
