LPM: Putting data in the cloud makes you nervous? You have options

Moving to the cloud doesn’t have to be an all-or-nothing approach. Firms can gradually make the move, says Stuart Chalmers, director of software engineering at LexisNexis Enterprise Solutions.

For many firms, cloud adoption is a given, but there are also several others who are reluctant to transition to the cloud. Fair enough. Reasons can be many – it’s technically daunting, selecting the right hosting platform is confusing, choosing the best option for hosting client data is complicated, or there are security and regulatory compliance concerns, among others.

The benefits of cloud computing are now well-acknowledged – no upfront capital expenditure, no redundant hardware, easy scalability of compute power, enhanced resilience and so forth. All these attributes are valuable, especially as typically most IT departments face cost-cutting challenges in some measure.

The good news is that the path to the cloud doesn’t demand an “all or nothing” approach. You can start small, adopting a hybrid approach as you move up the cloud maturity curve — and even blending the different options — all driven by business needs.

Decision-making considerations
Considerations such as how much control and responsibility must be retained in-house versus delegated to the cloud provider.

Data is another critical item because you will always be responsible for the content of the data, who can access it and from what device. Also, there is the issue of regulatory compliance. Therefore, the question that needs to be answered is “Do we want full control of where the data is hosted?”

It’s feasible to have a strategy that enables you to gradually transition to the cloud. If your firm has a traditional data centre that uses physical servers, you can start by ‘lifting and shifting’ to a virtual machine (VM) environment and adopt an Infrastructure-as-a-Service (IaaS) model, where the firm takes on the responsibility for installing, configuring and maintaining the operating system (OS) and database. Your cloud provider maintains the hardware, network connectivity and physical security. Think of the IaaS model as renting the hardware in a cloud data centre, but you can decide what to do with the hardware.

This arrangement could evolve with the cloud provider also taking on the maintenance of the operating system (OS) and any other tools that make up the cloud solution. In this platform-as-a-service (PaaS) scenario, you don’t have to worry about licensing or patching the OS and databases, you simply stay in control of the applications used. Essentially, the PaaS model will allow you to leverage more of the cloud services – such as app services, containerisation, and database services like Azure SQL and Amazon RDS – via the cloud provider’s managed services offering.

The other option is the software-as-a-service (SaaS) model ‑ where your cloud provider provides a fully developed application, for example, email or financial software. It’s the easiest for you to get up and running on. It also requires minimal technical knowledge on the part of your organisation to deploy.

The key to note is that whilst all these models – IaaS, PaaS and SaaS are available – if you are nervous about putting data in the cloud, you can still choose a hybrid approach with any of these cloud models that you decide to deploy. In doing so, you retain control of data, its security as well as regulatory and legal compliance.

Generative AI on the cards?
If your firm has generative AI adoption in its sights, then cloud adoption does become essential on account of the flexibility it provides. However, you need to have a clear strategy of what you are trying to achieve with AI so that you can design and future-proof your infrastructure to support the technology for the long term. Alongside this, you need to have a clear data strategy. AI is data-intensive, and the cloud will allow you to scale and exploit compute power to accelerate model training, but having the correct kind of data is crucial to the initiative’s success. All this combined will enable you to accelerate the velocity of your AI initiatives.

However you approach the cloud, you need a clear, business-driven and forward-looking IT strategy. Compute power, storage, resilience, flexibility and agility all come with associated costs. Verify and validate the various cloud services based on SLAs, your firm’s risk appetite and aspirations for modern and emerging technologies. This approach to decision-making will ensure a future-proof IT infrastructure.

Read the article in LPM today.

LPM: Putting data in the cloud makes you nervous? You have options preview