A practitioner's guide to building scalable systems

Building Scalable Systems: Distributed Systems basics to System Design Interviews

We want to give you the tools to compare multiple approaches and pick the one that works for your scenario.

Grokking the Distributed and Scalable Systems

# What is Scalability?
<div style="background-color:#81bd69;color:white;padding: 10px; padding-left:30px;">
Scalability is the ability of your web application to handle more load without a noticeable degradation in user experience and performance.

</div>

When you are designing a web-scale application (imagine millions or billions of users), you need to think about and plan for __Growth__ and __Peak Load__. 

__Growth__: Growth and, more importantly, Exponential Growth is when your service is growing every month, e.g., you are becoming the next Facebook or Snapchat.


__Peak Load__: 

If your application experiences exponential growth or you launch a highly anticipated new product (a good problem to have), 

* More data
* More concurrent users and interactions 
* Need for more compute power
* More transactions
* More network traffic and connections, 
* More failures and more entropy.

Following are a few examples of why you need to scale along the above dimensions as your product becomes more widespread:

1. When Facebook adds a new user, each user might upload GBs of pictures and videos.

2. To enter a new country/region, Uber might have to add millions of new locations and destinations.

3. When Fortnight becomes more popular, more players are simultaneously accessing their servers for matchmaking and playing games online.

4. Apple's Servers experience extremely high load when pre-orders open for a new iPhone.

5. If Amazon grows its inventory and is selling more goods every minute, it needs a more robust way to work with many payment gateways, etc.

## Growth can compound the Scalability needs
Sometimes, your growth can have a compounding effect on the need to scale, e.g., if Kim Kardashian had joined Facebook when Facebook had 1MM users and shared a picture, only 1MM people might have liked that photo. Now with 2 Billion members, 1MM people might like the image in the first minute, and now Facebook must update the newsfeed for 100MM the next second.

# Types of Scaling
We can deploy three different techniques for scaling application servers.

## 1. Vertical Scaling

## 2. Horizontal Scaling

## 3. Functional Scaling

# What is Scalability?
<div>
Scalability is the ability of your web application to handle more load without a noticeable degradation in user experience and performance.

</div>

When you are designing a web-scale application (imagine millions or billions of users), you need to think about and plan for __Growth__ and __Peak Load__. 

__Growth__: Growth and, more importantly, Exponential Growth is when your service is growing every month, e.g., you are becoming the next Facebook or Snapchat.


__Peak Load__: 

If your application experiences exponential growth or you launch a highly anticipated new product (a good problem to have), 

* More data
* More concurrent users and interactions 
* Need for more compute power
* More transactions
* More network traffic and connections, 
* More failures and more entropy.

Following are a few examples of why you need to scale along the above dimensions as your product becomes more widespread:

1. When Facebook adds a new user, each user might upload GBs of pictures and videos.

2. To enter a new country/region, Uber might have to add millions of new locations and destinations.

3. When Fortnight becomes more popular, more players are simultaneously accessing their servers for matchmaking and playing games online.

4. Apple's Servers experience extremely high load when pre-orders open for a new iPhone.

5. If Amazon grows its inventory and is selling more goods every minute, it needs a more robust way to work with many payment gateways, etc.

## Growth can compound the Scalability needs
Sometimes, your growth can have a compounding effect on the need to scale, e.g., if Kim Kardashian had joined Facebook when Facebook had 1MM users and shared a picture, only 1MM people might have liked that photo. Now with 2 Billion members, 1MM people might like the image in the first minute, and now Facebook must update the newsfeed for 100MM the next second.

# Types of Scaling
We can deploy three different techniques for scaling application servers.

## 1. Vertical Scaling

## 2. Horizontal Scaling

## 3. Functional Scaling

What and Why

Node in a Distributed System

Performance and Numbers

Fault Tolerance and Availability

Understanding TIME in Distributed Systems

Data Centers for software engineers

Scaling Distributed Systems

Load Balancers

Structured Storage at Scale

Sharding in Storage Systems

SQL vs NO-SQL

Blob Storage at Scale

Caching in Distributed Systems

Queues and Asynchronous Processing

Data Analytics and Batch Processing

Logging and Monitoring

Searching at Scale

Conclusion

What is Scalability in Distributed Systems?

What is Scalability?