The Perils of 3- (or More) Tier Infrastructure
The Perils of 3- (or More) Tier Infrastructure
If your infrastructure is still built on foundations created in 1999, here are some reasons that your infrastructure is costing you more than it's worth.
Join the DZone community and get the full member experience.Join For Free
Learn how to migrate and modernize stateless applications and run them in a Kubernetes cluster.
Once upon a time, a long time ago (in 1999), 3-tier infrastructure was the go-to for the needs of applications…from 1999.
I don’t mean to be glib — 3-tier was transformational during the web app explosion of that era, but that’s also the point: any technology that has to be amended with “from that era” is likely due for a good, long look, and especially if you’re planning to run today’s cloud-ready, shiny new apps and the dynamic workloads that have evolved from those apps of yesteryear.
But why not stay in the comfort zone of your 3-tier world? What are the dangers?
It Can Stunt Your Growth
Separate server vendors, storage vendors, networking vendors, virtualization vendors — with this setup, companies can pick best of breed technologies and carefully integrate them to build their data centers. However, this approach is not well-suited to scale out as business needs put pressure on IT systems, or as new application initiatives are spun up. Simply trying to expand infrastructure-based 3-tier infrastructure becomes complexity feeding complexity. And speaking of complexity …
It’s More Complex than A Rubix Cube, and Just as Time-Consuming
Keeping the lights on. Managing multiple vendors. Using multiple management interfaces. Relying on specialists. And the lost weekends, Friday nights, and holidays? Let’s not even go there. The complexity and associated management overhead that comes with 3-tier is often what sends IT looking for a new, better way to operate. With a system in place that ensures smooth operations and gives universal insight and control from one interface, you’ll find, (as Cardinal Innovations Healthcare did as an example), that you legitimately can get your Sundays (psst … and more!) back. Freeing up IT calendars means those people have the opportunity to build new things, grow their skills, and keep innovating—all good for the business.
It Doesn’t Have the Most Stable Long-Term Financial Future
First, we’ll be clear: hyper-converged infrastructure is not always the least expensive option upfront. But if you zoom out to the long-term infrastructure and operations picture — costs related to upgrades, rack space, power, cooling, administrative costs, services, other software, etc. over a period of 5 years or so — it makes much more sense to invest now so you (literally) won’t pay later. Opportunity cost is key here, too — whether it’s revenue when you have infrastructure that allows you to go to market faster, or not burning your budget on hardware that you only need 3 years from now. (If you’re curious what the numbers would look like for your unique business requirements, run them for yourself with this quick TCO Calculator).
It Doesn’t Give You a Firm Foundation for DevOps
This is major. If you’re not moving towards DevOps, you’re moving backwards. It’s a matter of static vs. dynamic. Organizational and operational fragmentation is public enemy number one here: if you’re working on an infrastructure with multiple platforms and management tools, it will be hard to automate, monitor, troubleshoot, and protect. You’ll also need to call for reinforcements, enlisting IT specialists (cue the “ca-ching!” sound effect) to bridge the gaps. To top it all off, provisioning in this type of environment is too time consuming and expensive to to make scaling a tenable option.
The silos, complexity, and inflexibility of 3-tier disappear when hyperconverged infrastructure enters the picture. It simplifies IT operations by converging servers, storage, networks and applications across distributed environments into a single management platform. It also enables staff to get greater control across the stack, automate processes, and scale on need. They truly become cloud engineers that build and operate an Enterprise Cloud.
So what’s holding companies back from moving off of 3-tier and onto HCI?
The old culprit – change. It’s hard. When it comes to implementing HCI-based solutions you know the strategic advantages, but it’s difficult to shake up a department where everything is running just fine.
Your team doesn’t have to suffer from whiplash; the change towards HCI can be gradual. But, the impact is transformational to the business and IT’s operational agility. Start small and scale. Pick one new application initiative or VDI initiative — something being sponsored within IT. Once you’ve succeed with your first project, you can roll out others over time and standardize upon your HCI deployment.
Is thinking beyond 3-tier on your to-do list? This guide can help you get started.
Published at DZone with permission of Jordan McMahon . See the original article here.
Opinions expressed by DZone contributors are their own.