At Acolad we definitely notice a public cloud wave. Both for digital experience and enterprise content solutions, the public cloud has become the default option for about 70% of our clients. This also implies that development is more and more taking place on cloud platforms such as Microsoft Azure and Amazon AWS. And that’s where we became excited about the serverless capabilities that cloud platforms offer.
What is serverless? Serverless can mean many things and be used in many contexts, but it boils down to this: you use a service without having to hassle with infrastructure. For example, you can run a database in the cloud without worrying about license issues, backups, upgrades and even scaling (common sense is still required here, of course). Software and web developers can concentrate on what they do best, without having to think on which (and how many) servers their work will run on.
The three big cloud providers (Microsoft Azure, Amazon AWS and Google Cloud Platform) all provide many different Serverless services, including but not limited to: storage, artificial intelligence (AI), databases and even compute power. This last one is commonly known as FaaS (Function-as-a-Service) which enables running business logic in the cloud without running servers.
With the adoption of public cloud hosting at its peak, we look at the four reasons why we believe serverless is the future.
It’s economically attractive
Cloud is not free, but a fierce competition between cloud providers and economies of scale have had a positive influence on prices going down. Comparing it to on-premises solutions, a lot of the traditional IT operations tasks such as setup, backup/restore, monitoring are simplified in serverless solutions, resulting in a decreased operational cost.
With pure pay-as-you-go pricing models, serverless also means “fair” pricing. You can start small with a limited cost and only upgrade as business picks up. Or make sure you have just the right resources to provide a good service, even at peak times, without having to over scale server or database capacity.
The speed we are talking about here is the time it takes to get your service up and running and the effort required to automate your deployments. Serverless can accelerate implementations by offering various levels of abstraction over the back-end infrastructure and automation tools for capacity provisioning and scaling, for example.
Regarding the response times of your website or application running in the (serverless) cloud, this will be determined by finding the perfect balance of your needs and budget. Again, applying the fairness principle: while some options might cost more, the performance increases might be worth it.
While AWS’s first real public cloud service - Simple Storage Service or S3 - has now been around for more than 10 years, in the last few years the rate of innovation in public cloud has increased dramatically. Cloud providers are gearing up with more and more comprehensive artificial intelligence (AI) and machine learning (ML) embedded services, more players have entered the space and the market is increasing every year. The figure below proves the success in public cloud adoption.
From the nearly 1000 organizations enquired for The State of the Cloud 2018 about their adoption of cloud infrastructure, a large majority (over 68%) reported they were already running business applications on public cloud providers.
We see the same rise in interest amongst our own clients as they advance in their digital transformation strategies, including in some of the more conservative industries such as banking. So cloud is not only accepted, but even becoming the preferred option of CIOs looking for more IT agility. And serverless is then a logical cost-effective and efficient option.
It’s offering opportunities
The serverless model eliminates of a lot of the more mundane tasks in server management, such as server setup, backup/restore, scaling and it simplifies other tasks such as monitoring. For Acolad as digital solution provider, this allows our teams to focus on the “things we’re good at”, such as understanding the needs of our clients and implementing business logic to accommodate these requests. A sound architecture and implementation are still a must, and that’s why we partner with leading-edge public cloud providers that know all the ins-and-outs and handle the latest best-practices.
One of the areas which traditionally requires a lot of specialized knowledge to set up and maintain is Artificial Intelligence. While some public cloud providers offer custom AI platforms for organizations to build and train their own machine learning models, most already provide general pre-built algorithms which can be used in your application. One good example here is Azure Cognitive Services, which provides a whole set of general AI services such as image recognition, text interpretation, speech recognition, etc. Within our Office 365 collaboration solutions this functionality is already used productively, for example to automatically transcribe video files, thus generating automated captions and textual search possibilities on video content (see below).
An intranet portal built on Valo. While the highlighted video news on the top is playing, automated captions are displayed to better capture the attention of the user (in case sound is muted) or ease comprehension (if the user is not a fluent speaker).
So, is the public cloud right for your organization?
At Acolad we’re strong believers in public cloud and the benefits flexible and secure cloud environments bring to our clients. As a strategic tool for digital transformation, going serverless is just a logical next step for CIOs and IT leaders in this ‘efficiency’ evolution.
But the adoption of a public cloud platform is a significant shift in strategy with opportunities and challenges at both a technology and a wider organizational level. This change can impact a wide range of teams, so getting it right is critical.
Having the right resources on your side ensures a well-developed cloud migration strategy. Following a structured approach to determine your migration requirements maximizes your chances of a successful cloud migration. So, make sure you understand the advantages and disadvantages of each option before deciding on the solution that best suits your business.