Serverless

# Introduction

:::tip Serverless architecture is a way to build, run applications and services without manage infrastructure. Your application still run on servers, but all server management (hardware and software, bandwidth) is done by service Provider. :::

# Why use Serverless

# The Advantage and Disadvantage of Serverless

✏️ Reduced time to develop and quicker software release.

✏️ A smaller cost to scale. Developer don’t need to implement code to scale or admin do not need to upgrade existing servers or add additional ones.

✏️ Allow developers to focus on code and to deliver faster.

✏️ Fits with microservices, which can be implemented as functions.

✏️ Charge based on usage => lower prices.

✏️ No need to run at 24/7 (run when have request only).

✏️ Lower operational and development costs.

# The disadvantage of serverless:

✏️ Difficult for debug.

✏️ Performance may be affected.

✏️ Dependency the cloud provider (vender lock-in).

✏️ Limited about time, memories, …

✏️ Security issues.

# Serverless

Serverless divide into two type: BaaS (Backend as a Service) and FaaS (Function as a Service). Serverless is also known as FaaS if not specific.

# BaaS (Back end as a Service)

::: warning BaaS (Back end as a Service): the service that code logic is handles as the client side, the server-side can be use APIs from the third-party. :::

BaaS advantage:

  • Development speed

  • Development price

  • Outsource cloud infrastructure management

BaaS disadvantage:

  • Less flexibility in comparison to custom coding

  • Less secure

👉 BaaS use cases: Data management, multimedia, file storage, push notifications, …

# FaaS (Function as a Service)

::: warning Faas (Function as a service): the service that developer deploy code as a function, which can be called by RestAPI. :::

# FaaS advantages:

  • Do not need to manage server infrastructure

  • Developer focus more on code logic

  • Never pay for idle resource

  • Built in availability

# FaaS disadvantages:

  • Difficult for debug

  • Auto-scaling of function calls often means auto-scaling of cost. This can make it tough to gauge your business expenses.

  • Waiting time for handle higher

👉 FaaS use cases: Web Apps, Backends, Data/stream processing, Chatbots, Scheduled tasks, …

🔸 Azure Functions (Microsoft)

🔸 AWS Lambda Functions (Amazon)

🔸 Google Cloud Functions (Google)