2 questions to ask before diving into infrastructure-as-a-service
A migration to IAAS can be daunting, and early planning is key to maximizing returns
- By David Egts
- Feb 10, 2014
We all know that when it comes to embracing virtualization and the cloud, the surf’s definitely up for federal IT managers. And while the first waves may have broken in 2013, the cloud tsunami -- particularly in relation to infrastructure-as-a-service -- is certain to come on strong this year.
There are many reasons for this. First, IT staff no longer must trek to the data center or fuss with hard drives to manage infrastructure – everything can be done from the desktop. Second, properly designed IaaS deployments can actually allow failures to happen without the customer and their users noticing. Third, IaaS allows managers to test new and innovative ideas by the hour in the cloud without capital expenditures, lowering the cost of failure.
Finally, IT professionals are realizing the benefits of not being locked into a single vendor or solution and are increasingly turning to cloud agnostic tools to help them manage their environments.
Still, for federal IT managers seeking to adopt IaaS in 2014, there are a couple questions to ask before diving in headfirst.
How can I maximize efficiencies and cost savings?
The inclusion of IaaS can be tricky, especially for agency IT professionals who are already stretched thin and being asked to learn new solutions while still managing day-to-day tasks. They are finding that managing multiple services can be a challenge -- but it does not have to be. In order to circumvent this, it’s important to focus on solutions that span both public and private clouds, resulting in a single virtual view of a hybrid data center. This saves time in having to learn different facets of each system, and promotes interoperability, workload and data portability, open APIs, and freedom of choice across new and existing heterogeneous infrastructures.
To achieve cost savings, agency IT professionals must ensure applications are written to be resilient and scale. When paying by the hour for cloud services, agencies pay only for what is used, but it is important that they use them only when needed for optimal cost savings. An additional area of cost savings is a cloud-agnostic management tool. This layer of abstraction allows staff to be trained once, instead of on every cloud vendor’s tools; the layer of abstraction also minimizes vendor lock-in and gives the ability for arbitrage, allowing many vendors to bid for business and deliver the highest value for the lowest cost.
How can I get my arms around learning how to manage IaaS solutions?
Like many IT managers, those within federal agencies are being tasked with re-learning some of the tricks of the trade. It’s no longer about managing gigantic servers. Instead, it’s about managing software that allows the entire organization to be more flexible, responsive and dynamic. This is fantastic in theory, but realistically, it is causing IT departments to learn how to effectively use and manage technologies with which they may not be familiar or comfortable, or even have time to oversee.
Mastery of IaaS solutions does take time, but it’s certainly not something agencies should avoid. This mastery is achieved faster by aligning with respected solutions providers that offer technology support services, training and consulting. Importantly, these services and the people who manage them should have deep knowledge of the particular concerns of agency IT staff, including security, uptime, chargeback and other crucial needs.
IaaS waters might be deep, but they do not have to be difficult to navigate. It’s important for agency IT professionals to ask the right questions before wading in, so they can more quickly get to their ultimate destination of simplicity and flexibility.
David Egts is chief technologist, North America Public Sector, Red Hat.