Everything you need to know about Serverless

Serverless is a cost-efficient way to host your APIs and it forms the crux of systems like Chatbots and Online Judge.

Serverless does not mean that your code will not run on the server; it means that you do not manage, maintain, access, or scale the server your code is running on.

The traditional way to host APIs is by spinning up a server with some RAM, and CPU. Say the resources make your server handle 1000 RPS, but you are getting 1000 RPS only 1% of the time which means for the other 99% you are overprovisioned.

So, what if there was an Infrastructure that

• scales up and down as per the traffic
• is billed per execution
• is self-managed maintained and fault-tolerant

These requirements gave rise to Serverless Computing.

Real-world applications

Chatbot

Say, we build a Slack chatbot that responds with the Holiday list when someone messages holidays . The traffic for this utility is going to be insignificant, and keeping a server running the whole time is a waste. This is best modeled on Serverless which is invoked on receiving a message.

Online Judge

Every submission can be evaluated on a serverless function and results can be updated in a database. Serverless gives you isolation out of the box and keeps the cost to a bare minimum. It would also seamlessly handle the surge in submissions.

Vending Machine

Upon purchase, the Vending machine would need to update the main database, and the APIs for that could be hosted on Serverless. Given the traffic is low and bursty, Serverless would help us keep the cost down.

Scheduled DB Backups

Schedule daily DB backups on the Serverless function instead of running a separate crontab server just to trigger the backup.

Batch and Stream Processing

Use serverless and invoke the function every time a message is pushed on the broker making the system reactive instead of poll-based.

Advantages

• No need to manage and scale the infra
• The cost is 0 when you do not get any traffic
• Scale is out of the box; so no capacity planning is needed

Disadvantages

• Takes time to serve the first request as the underlying infra might boot up
• The execution has a max timeout, so your job should complete within the limit
• Debugging is a challenge
• You are locked in on the vendor you chose

When NOT to use Serverless

• Load, usage, and traffic pattern is consistent
• Execution will go beyond the max timeout
• You need multi-tenancy

When to use Serverless

• Quick build, prototype, and deploy the changes
• Usecase is lightweight
• Traffic is bursty

Here's the video of my explaining this in-depth ?? do check it out

Serverless Computing is one of the hottest topics of discussion today, but the term "serverless" is slightly misleading and it does not mean that your code will not need a server to run. We have to be extremely cautious while deciding on adopting serverless for our use case, but it is not something that fits all the use cases.

In this video, we talk about what serverless computing is, see why it was built in the first place, learn about 5 real-world use-cases that become super-efficient with serverless architecture, understand the advantages and more importantly, the disadvantages of adopting it, and conclude with acknowledging when to use and when not to use this computation pattern.

Outline:

• 00:00 Agenda
• 03:01 Need and the idea of the Serverless Computing
• 11:16 Usecase 1: Chatbots
• 13:57 Usecase 2: Online Judge
• 16:27 Usecase 3: Vending Machines
• 18:00 Usecase 4: CRON Jobs
• 19:54 Usecase 5: Batch and Stream Processing
• 22:16 Advantages of Serverless Computing and Architecture
• 26:21 Disadvantages of Serverless Computing and Architecture
• 31:37 When NOT to use Serverless
• 34:00 When to use Serverless

You can also

• Subscribe to the YT Channel Asli Engineering
• Download the notes
• Listen to this on the go on Spotify

Thank you so much for reading ?? If you found this helpful, do spread the word about it on social media; it would mean the world to me.

You can also follow me on your favourite social media LinkedIn , and Twitter .

Yours truly,

Arpit

arpitbhayani.me

Until next time, stay awesome :)

I teach a course on System Design where you'll learn how to intuitively design scalable systems. The course will help you

• become a better engineer
• ace your technical discussions
• get you acquainted with a massive spectrum of topics ranging from Storage Engines, High-throughput systems, to super-clever algorithms behind them.

I have compressed my ~10 years of work experience into this course, and aim to accelerate your engineering growth 100x. To date, the course is trusted by 600+ engineers from 10 different countries and here you can find what they say about the course.

Together, we will build some of the most amazing systems and dissect them to understand the intricate details. You can find the week-by-week curriculum and topics, benefits, testimonials, and other information here https://arpitbhayani.me/masterclass .