The modern CIO doesn’t really need an introduction to Amazon Web Services. To say that AWS is already well-known is an understatement.
IT leaders can still use a helping hand, however, in navigating the massive scope and scale of AWS while making sure their cloud strategy is tightly focused and aligned with business goals.
[ Want to accelerate application development while reducing cost and complexity? Get the eBook: Modernize your IT with managed cloud services. ]
That’s a taller task than it might first appear. AWS and other public clouds are called "hyperscale" clouds for a reason. Keeping up with the vast number of technologies and services available can become overwhelming. Moreover, adopting and using said technologies and services just because they exist doesn’t promise any tangible results other than a bill.
5 tips for IT leaders considering AWS cloud services
With that in mind, we asked several cloud experts for their best advice on maximizing ROI while not getting bogged down with things that don’t deliver impact in your business. Here are their top five tips.
1. Deriving value from the platform requires clear business cases.
The primary question worth asking, now and over time: What’s the business case for doing this?
There’s a long list of good answers to that question, and they can be quite specific to the organizations – or even at a departmental or team level.
There’s one answer that should give you pause – and that requires direct attention: I don’t know.
"Harnessing the power of AWS in a secure, cost-efficient, performant, and reliable way requires CIOs to not only look outward at the capabilities and services available on the platform itself, but to look inward at their business," says Jonathan LaCour, CTO at Mission Cloud Services.
This is a fundamental truth of IT in general – if you’re implementing new tools or services without clear reasons for doing so, something’s usually amiss. The expansiveness of Amazon’s public cloud makes it all the more important here.
"AWS provides an incredible number of APIs that empower a multitude of choices and pathways, so the most important consideration is how you apply these to your own business and what really matters the most," LaCour says.
[ Get the checklist: Top security and compliance considerations for IT modernization. ]
2. Clear workload-objective cases will help sharpen your focus.
Michael Allen, software architect at Laserfiche, reframes the high-level business strategy question for the more specific task of workload selection: What’s our objective in migrating workloads to a public cloud?
"Is it reducing labor costs in operating systems that are known to be very difficult to run, like Hadoop? Is it to achieve disaster recovery targets? Does it allow developers to iterate faster through using self-service?" Allen says, describing some common goals teams seek to achieve by moving select workloads to a public cloud.
Connecting objectives to workloads (or vice versa) is a must – otherwise, you’re just floating around in a massive public cloud ecosystem without any real direction – which in turn means results will remain elusive.
"Workload sele†ction is going to be driven by business objectives and workload selection drives requirements and risk management activities," Allen says.
Developing clear workload-objective connections is a means of staying focused when faced with a massive menu of choices on AWS.
LaCour shares some more example questions to ask, ideally before you embark on a migration or modernization project, though better late than never:
Does your workload and business case prioritize performance over cost?
Are you in a highly regulated industry with specific compliance requirements?
Do you have parts of your workload that, while critical, aren’t time-sensitive?
"These questions all inform the decisions you will need to make to be successful on AWS," LaCour says. "Don’t simply look at the breadth and depth of the AWS platform. Think carefully about your own business case and needs, both current and forward-looking."
Investing in a workload-specific strategy will not only increase the odds of success for a specific project, according to LaCour, but also pay long-term dividends.
"It’ll rapidly accelerate your strategic planning by focusing and narrowing the scope of consideration to what matters the most," LaCour says.
3. Pay attention to the emerging "supercloud" concept.
E.G. Nadhan, global chief architect leader, Red Hat, advises IT leaders to keep tabs on the "supercloud," an idea that began to emerge at AWS Re:invent 2021. Nadhan points to a SiliconANGLE analysis of the event – and its supercloud definition:
"Supercloud describes an architecture that taps the underlying services and primitives [Kubernetes objects] of hyperscale clouds to deliver additional value above and beyond what’s available from public cloud providers," writes author Dave Vellante. "A supercloud delivers capabilities through software, consumed as services, and can run on a single hyperscale cloud or span multiple clouds."
In this manner, the supercloud is essentially a reflection of the reality that cloud is no longer simply an infrastructure choice or an IT operating model, but a transformative operating model for entire businesses. Nadhan thinks CIOs will want to ask questions and think about how the supercloud idea might be applicable in their specific organizations and industries – as well as those of key partners.
[ Also read: Hybrid cloud: 4 trends to watch in 2022 ]
"Amazon’s roadmap for superclouds should be of great interest to CIOs – especially for business domains most relevant for their enterprises," Nadhan says. "CIOs must also question the viability of superclouds not only for their own industry but the peripheral industries they intersect with across their ecosystem of supply chain partners."
4. Who’s going to manage this?
It’s a question some seasoned IT pros might be careful about asking, because the answer is often: "You." But IT leaders do need to identify who on their team will handle the operational work involved in migrating and managing workloads on AWS.
Allen from Laserfiche says leaders should be asking questions like: What additional training do we need to provide our staff? Are the existing IT staff certified or trained to use cloud tooling? And: What about our developers?
In some cases, the necessary capabilities may already exist in-house. In others, you’ll need to hire them or ensure ample professional development via training and/or certification.
Not every organization will want or be able to invest in the necessary skills, in which case a fully managed option like Red Hat OpenShift on AWS (ROSA) is worth considering.
5. Be realistic – and holistic – about costs.
As we reported recently, IT leaders are best served by a holistic view of costs with cloud service providers in general.
"Cloud service providers can get expensive," Gordon Haff, technology evangelist at Red Hat, told us. "That doesn’t mean you shouldn’t use them. But you need to understand where they do provide good value for your organization and where you should consider taking workloads on-prem."
The basic question to start with here, according to Allen: "How are we going to budget for projects and control costs in the cloud?" That leads to other cost-related choices, such as: "Do we want to pre-pay to reduce total spend and limit unexpected costs or lean more on pay-as-you-go and operational expenditures?"
Assumptions like "this will probably be cheaper than my own datacenter" can get wrecked pretty fast. Replace assumptions with a plan, ideally one that is not too myopic about costs. Yes, the budget matters – a lot. But as LaCour told us recently, a successful cloud strategy depends on a holistic, long-term view.
That should include "a platform’s ability to enable constant innovation, accelerate their time-to-market, and, yes, reduce the distraction of costly operational overhead the company would otherwise be on the hook for," LaCour says. "The more you get that right, the faster and more easily you can focus on your customers and growth, not on babysitting your infrastructure."
[ What should you know about AI/ML workloads and the cloud? Get the eBook: Top considerations for building a production-ready AI/ML environment. ]