We explore serverless functions, how they stack up against traditional architecture, and how to determine if they are the right fit for your business.
A serverless function is a single-purpose, programmatic function hosted on managed infrastructure which is maintained by a cloud computing company. These functions are used for an individual purpose and are externally triggered to operate. Once their action has been executed, they then ‘turn off’. This is economically beneficial for many businesses as they are only charged for when the function is actually ‘on’, down to the millisecond.
An example of when it would be beneficial to use a serverless function would be if a part of your app runs a weekly report. You don’t need to use a whole server for this function and it only needs to run once per week, therefore you can set it up as a serverless function and trigger it weekly, only being billed for the time that the report was being generated.
So, what are the main benefits of using Serverless Functions vs Traditional Architecture?
- Cost: With serverless functions, after your initial project set-up, you only pay for what you use – down to the millisecond! This means you no longer have to pay for a whole server running 24/7.
- Minimized risk of downtime: Serverless functions break down various components of your app infrastructure into individual functions, meaning if anything were to fail, it would be limited to that individual function rather than bringing down your whole app.
- Built-in scaling: If traffic suddenly increases to your application then it can easily scale up to meet demand.
- Built-in functionality: All the main cloud providers have serverless function solutions and these come with lots of built-in functionality such as authentication, authorization, logging and server configuration.
Who would be an ideal candidate for a project including serverless functions?
Serverless functions can be utilized across a wide range of projects from eCommerce to reporting. If you are creating an app that has many API integrations, you could individualize each integration as a serverless function. This means that they each have a smaller code base and can be easily updated and deployed independently from the rest of the system.
On an eCommerce website there are many stages a user goes through, from searching for a product, to ordering, paying and receiving confirmation details. All the steps to making a purchase can be separated into serverless functions. Each play their own small role in the process and don’t need to be running all at once; they can be triggered to function when needed.
If your team require a daily report for a meeting each morning, you could use a serverless function to run at a certain time to process the data and put it in a readable format in an email for your staff in the morning.
The use cases for serverless functions are so broad, they can be used in many instances that you’d like a process to be automated.
Are there any disadvantages to using serverless functions?
If you have a program with a lot of traffic or that needs to be running constantly, then serverless functions aren’t for you. In that case it would be a more costly solution. Furthermore, if your app is too tightly coupled, meaning that lots of processes are dependent upon each other, then serverless functions would also not be suitable. This is because the app would be too difficult to separate into individual functions.
At Chateaux we have been working with serverless functions for years and have implemented them as part of many projects. This experience coupled with our dedicated team of experts mean that we are well positioned to execute any serverless function project. If you’d like to know more about microservices in general, take a look at our service page. Or if you’d like to get in touch please contact us at [email protected].
