If you attended one or both of the IT industry’s key technology conferences in Las Vegas recently, you likely came away with two general messages:
- The cloud, specifically AWS, is out to consume nearly every IT workload it can.
- Complex challenges persist around IT governance and hybrid IT deployment.
While neither of these challenges has a one-size-fits-all solution, it was clear that IT professionals should resist the urge to move too far and too fast into the world of public cloud services. Read on for more on this, plus my top five lessons learned at Gartner IOCS and AWS re:Invent 2018.
1. Overspending on public cloud services continues to be an issue
Attendees at Gartner IODC shared stories of customers overspending on public cloud services, a trend that currently affects about 40% of cloud consumers. Fueled by a lack of understanding among IT administrators about which cloud IT services are being deployed, how services are being managed (governed) and how to budget for their company’s average resource consumption, administrators often choose to “over-purchase” services rather than risk a shortfall.
2. When it comes to data center migration strategies, remember “The Five Rs”
As IT consumers consider effective data center migration strategies for cloud services, Gartner provides the following five strategies, aka “The Five Rs”—
Rehost: This refers to lifting and shifting applications, e.g., shifting to on-premise colocation services on the same server.
Refactor: This is a slightly modified version of rehosting, typically on the same platform.
Rearchitect/Revise: Modify the application and replace the infrastructure platform.
Rebuild: Completely rebuild or rewrite the code for an existing application, often with more cloud-native services or a new infrastructure.
Replace: Replace the application with SaaS or a combination of applications and services.
Of course, factors of time and cost (value) are key. Decision-makers should weigh the time it takes to accomplish the migration against their associated business goals, as well as the long-term value and potential cost of the application.
3. The data center is dead
This idea is, in my opinion, subjective. In my view, the current state of colocation, cloud and edge depend upon: (1) your perspective; and (2) where specific applications are in their respective lifecycles. More importantly, nearly across the board everyone agrees that building and managing your own data center makes little sense. Companies like Flexential that specialize in this are more efficient and more likely to provide better uptime, availability, and overall total cost of ownership.
Data continues to be the key focus as new options develop and evolve, but IT consumers should practice restraint. Ever heard the adage, “Just because you can doesn’t necessarily mean you should”? While there may be relative value to a particular application, IT consumers should practice ongoing optimization to keep everything in check.
4. Edge-enabled IT services abound
Both conferences saw the introduction of products and concepts for edge-enabled IT services.
Of note was the unveiling by Amazon of Outpost and other processor options that can optimize workload deployments and provide more cost-effective options. Outpost allows for broader deployment options to edge, colocation, private cloud and on-premise locations. This enables native AWS services to be provisioned and consumed locally to the end-user or platform needed. Processor sets will continue to diversify to provide best-in-class computer services.
Gartner called edge a 3D tapestry or a “rich topology” of network services that will potentially enable services to scale, allow storage and computer services to work in conjunction with machine learning, and improve artificial intelligence functions to help focus and monetize data across the network.
5. Partnerships are key
At Gartner IOCS, Flexential and Hitachi Vantara together presented a joint object storage technology platform that enables edge use cases within Flexential data centers, providing de-centralized storage well beyond the hyper-scale cloud deployments like AWS. Initial use cases include real-time analytics for video, big data and smart city applications, including campus safety.
Overall, the two conferences presented numerous new offerings in the marketplace that have exciting new capabilities and potential. The digital transformation is shifting into high gear, and that calls for solutions for local deployments, large-scale data and networks that support an ever-evolving set of IT services and requirements. As 2019 approaches, we will see a continued push for diversity in workload locations tied together by networks, data centers and, more importantly, trusted partners to help make sense of it all.