Course Introduction

What You'll Learn in This Course

Golang Useful Resources and Tools

The authors

The history of Go

Installing Go

Golang: An Overview of Fundamental Concepts

Introduction

A Quick Primer on Key-Value Servers and Hash Tables

How to Create a Basic hello world Webserver with Go

Build a Basic hello world JSON API Server with Go + Chi

How to Create an In-Memory Hash Table API with Go + Chi

Conclusion

A key-value server

Testing a key-value server

What is End-to-End Testing, Integration Testing, and Unit Testing?

Writing Unit Tests for JSON Functions in Go

Parallel Testing in Golang: How to Use t.Parallel()

A Brief Overview of Golang Benchmarks

Integration testing

Further testing types

Distributed key-value server

What is Distributed Consensus and How Does it Work?

An Introduction to Using Raft for Distributed Consensus

Create a Distributed Webserver using Golang and Raft

Connecting Golang Servers With a Handler and Middleware

Building a Store Package to Sync Data in Go with Raft's FSM

Testing Distributed Golang Webservers in Docker Containers

Distributed Key Value Server

Monitoring

White Box Monitoring vs Black Box Monitoring

A Introduction to Golang Logging

Defining and Implementing Distributed System Tracing with Go

Alerting

Understanding the Basics of Canary Deployments

Wait, what's a load balancer?

What aren't we doing?

So what will we be doing?

How to Create a Basic Random-per-request Load Balancer

What is Weighted Round Robin Load Balancing?

Streamlining Configuration Updates for Canary Deployments

Other considerations

Three Open-Source Load Balancers Written in Go

A canary deployment server

Chat server

Chatting about chat servers

What features don't we want (yet)?

What features do we want?

Using a SQLite database in Golang

How to Create an API for a Golang Chat Server

Creating a Frontend Interface in Go for a Chat Server

How to Deploy a Golang Webserver With Docker

Future improvements

Future thinking

Project ideas

Areas of more research

In this book we will show you the some of the concerns of backend web development through the lens of Go.

It is taught by Nat Welch (formerly an SRE at Google) and Steve McCarthy (Etsy), two Go developers who have been writing software for more than 25 years combined.

In the course, we'll be covering the creation and concerns of a number of fundamental backend building blocks or larger applications:

- A key value server, including the use of distributed consensus algorithms
- A canary deployment server
- A real time chat service
- Testing and monitoring

Regardless of your background, Go can make you a _better backend developer_, but learning Go is challenging because the language looks simple on the outside, but you won't get its full benefit unless you know what it can _really_ do. 

**By the end of the first chapter, you'll already _have built a key-value server reminiscent of similar systems like Redis and memcached._** The subsequent chapters will dive into more hands-on projects and Go concepts including:

- How to test your Go systems with unit tests and integration tests
- **Project**: create a **distributed version of the key-value server** in order to solve the typical problems of server systems such as _node failure_
- How to solve the **problem of "consensus"** in distributed Go systems
- How to write code to **keep data in sync** when running a system for servers located around the globe
- Implementing **black box and white box monitoring** systems for Go servers
- **Project**: building a **load balancer server** from scratch
- Why and how to implement a **canary deployment server** for bug-free gradual deployments
- How to build a **configuration server** to allow zero-downtime deploys 
- Understanding the various tradeoffs in the **design decisions** for Go server systems
- **Project**: build a Go chat server system from the ground up 
- **Database modeling** for backend development as part of your chat server project
- **and a lot more**...

## A Key-Value Server Like Redis And Memcached

n the early chapters of the book we learn how to build a Golang version of a key-value server that works in a manner similar to Redis and Memcached. You'll implement each of these features in Go:

- set and retrieve values by key
- adding persistent storage
- using data encodings such as `JSON` and `base64`
- testing the server with unit tests and integration tests

In these chapters, we also learn how to build a more advanced solution for the problem: **a distributed version of the key-value server**. You will learn how to use **consensus algorithms** in Go to solve the problem of keeping distributed servers in sync. The solid grounding you get in programming Go in these first chapters readies you for more complex Go projects.

## A Real-Time Chat Service

A real-time chat server implemented in Go gives you the opportunity to bring together systems thinking as well as concerns common for web apps:

- relational data modeling with SQL
- authentication
- implementing RESTful resources

Working through this comprehensive project will give you **a high degree of confidence about using Go to implement a huge project from scratch**. You will feel ready to use Go with relational databases such as PostgreSQL or MySQL, and databases in general.

## A Canary Deployment Server And Load Balancer

The work you do in the chapters covering building a canary deployment server and load balancer will open your eyes to the full power of Go. Gradual deployments are a commonly used strategy for code deployments, and the projects you build in these chapters get you familiar with building this kind of **infrastructure software**. 

Your load balancer will implement _random-per-request_ as well as _round-robin_ load balancing. These projects will give you a thorough immersion in systems problem solving with Go. You'll be able to show off your skills on bold projects of your own, in job interviews, and on the job when collaborating with colleagues.

## This book is the _fastest_ way to learn modern backend Go development

With _The newline Guide to Reliable Webservers with Go_, you won't waste time reading shallow blog posts that can't go beyond showing you the simplest Go examples or Todo apps. Instead, **within the first chapter** you'll be well on your way to creating a **Redis-like key-value server**, and the complexity of the systems you're able to build only accelerates from there.

This book opens the door to **more effective backend development** and utilizing the full power of Golang for **microservices**. As Go transforms the systems development landscape, you can use this book and the skills it teaches you to contribute to a faster, more effective systems development experience for your team.

With _The newline Guide to Reliable Webservers with Go_ you can **go from novice to proficiency in Go in a predictable way, without wasting time, and without picking up bad Go coding habits**.

So grab a copy now to advance your journey with Go.

In this book we show important concepts in backend web development through the lens of Go.

Reliable Webservers with Go

The importance of testing your web servers

How to apply distributed consensus algorithms to your key value server

Round robin is a style of load balancing where instead of a server being chosen randomly, the load balancer iterates through the servers based on a predetermined order. In 

 round robin load balancing, the number of consecutive uses of the same server will be based on the 

 is a number you choose that increases the percentage of requests that will go to an indivudal server. Servers with higher weights tend to be more powerful servers with higher CPU counts and RAM capacity.

You can think of weight like a counter. As we receive requests, we send them to one server while keeping count. Once the count is greater than the weight set for that server, we switch to the other one, and restart our count. Again once we have counted more than the number assigned to the second server, we will switch to another (or in our case back to our original server), and begin counting again.

In this section, we're going to be implementing a relatively simple weighted round robin. But first, why would we want round robin, and in particular why a weighted round robin?

If our two servers were to be relatively similar to each other, it shouldn't matter which one a request is sent to, so round robin would be a "fair" load balancer, evenly distributing load to each server. The addition of weighting could help us in cases where our servers aren't well-matched, sending an appropriate ratio of requests to each server.

Regardless, it is a relatively common load balancing strategy, so we're going to add it as an option for our canary deployment server.

Because we now have two methods of load balancing, let's first create a 

 for our load balancing style, as well as constants for those two possibilities. In the 

// LBStyle is an enum for what type of LB we have.
type LBStyle int

const (
	// RandomPerRequest sends you to a random backend every time.
	RandomPerRequest LBStyle = iota

	// RoundRobin sends you to the next server in the list.
	RoundRobin
)

---
title: "What is Weighted Round Robin Load Balancing?"
# description: desc
# privateVideoUrl: url
---

# Round robin load balancer

Round robin is a style of load balancing where instead of a server being chosen randomly, the load balancer iterates through the servers based on a predetermined order. In _weighted_ round robin load balancing, the number of consecutive uses of the same server will be based on the _weight_ each server has. In this case, _weight_ is a number you choose that increases the percentage of requests that will go to an indivudal server. Servers with higher weights tend to be more powerful servers with higher CPU counts and RAM capacity.

You can think of weight like a counter. As we receive requests, we send them to one server while keeping count. Once the count is greater than the weight set for that server, we switch to the other one, and restart our count. Again once we have counted more than the number assigned to the second server, we will switch to another (or in our case back to our original server), and begin counting again.

In this section, we're going to be implementing a relatively simple weighted round robin. But first, why would we want round robin, and in particular why a weighted round robin?

If our two servers were to be relatively similar to each other, it shouldn't matter which one a request is sent to, so round robin would be a "fair" load balancer, evenly distributing load to each server. The addition of weighting could help us in cases where our servers aren't well-matched, sending an appropriate ratio of requests to each server.

Regardless, it is a relatively common load balancing strategy, so we're going to add it as an option for our canary deployment server.

Because we now have two methods of load balancing, let's first create a `type` for our load balancing style, as well as constants for those two possibilities. In the `load_balancer` file:

```go
// LBStyle is an enum for what type of LB we have.
type LBStyle int

const (
	// RandomPerRequest sends you to a random backend every time.
	RandomPerRequest LBStyle = iota

	// RoundRobin sends you to the next server in the list.
	RoundRobin
)
```

Here we've set up our two types of `LBStyle`, using Go's [`iota`](https://github.com/golang/go/wiki/Iota) keyword to give our two types values. Iota works by starting with the value `0` and increasing by 1 for every new line that isn't a comment or all empty space. It combines well with `const`s implicit repitition, where any line without the `=` expression will repeat the most recent line that included it. So here our `RandomPerRequest` is set to `0`, and `RoundRobin` is implicitly set to `1`. Both of them have also have the `LBStyle` type, instead of directly being an `int`.