Go serverless and transform your business

Most will have heard the term ‘serverless’ and may appreciate that it does not literally mean that software operates without a server. However, many are unsure of what serverless does actually mean.

In simple terms, serverless refers to software running on servers that you neither rent nor own and are billed for on a pay-per-use model, rather than conventional subscription or rental models.
Serverless is billed by the millisecond for the duration of operations that are carried out and, in most cases, this will result in considerable cost savings compared to the cost of renting computing power by the month regardless of whether it is used or not.

The cost benefits for most use cases can be likened to comparing the total cost of car ownership over a given period, to the cost of using Uber for the most part and occasionally renting a car. As well as being cheaper (when you factor in depreciation, insurances, service and repairs, fuel, parking, taxation etc) using Uber can be far more convenient – you never have to waste time or effort in finding parking, cleaning the car, taking it to be serviced and having to make alternative arrangements when its not available, risking speeding tickets and so on.

There are, of course, scenarios where serverless might not be cost effective or might not be as convenient and this is where an inclusive or hybrid approach will pay dividends.

Where serverless really excels, is its ability to add features or functionality to your existing services, without having to be part of your software application.

Another fantastic benefit of serverless is that it doesn’t have to be physically located in one place or in the same place as your existing software. It can exist and run at any or all global ‘edge’ locations at no additional cost.

A simple but effective example of this, would be a function that runs at edge, identifying the country and screen width of all incoming viewer requests, adding a cookie and redirecting them to locale and device specific URLs, all without having to trouble your already overworked server.

Because the operations are carried out at edge, there are no performance or latency penalties and the cost for 5 million requests in a month would be under $1.

Serverless also mitigates the ‘single point of failure’ risk to a large degree. That isn’t to say that it is impossible to cause downtime from a serverless function, but that their decoupled nature would make it less likely.

Several hundred popular serverless functions, such as image processing, authentication handling and much more, are freely available to use through online repositories, saving development and dev-ops time.

The serverless stack also includes ways to incorporate logic to handle nested or conditional functions, error handling and retries and other logic usually associated with applications.

An exciting recent addition to the serverless family is serverless database at edge locations, offering many cost and performance advantages over traditional database deployments.

There are environmental benefits associated with serverless. Computers consume a lot of energy, even when they are left idle. Optimising energy expenditure by cutting out energy wasted on idling servers, does make a difference when applied at a global scale.

In conclusion, serverless is an absolute must for inclusion into the technology stack of every business. Serveless can enable businesses to do more at a lower cost and with a reduced risk, while helping to save the planet.

For help with serverless and other digital transformation technologies, please contact us to talk to one of our expert consultants.

GET IN TOUCH