Leon Adato is Head Geek™ at SolarWinds.
Cloud computing’s mounting importance and the shift to hybrid IT is propelling organizations of all shapes and sizes to hurriedly migrate systems infrastructure up and out of the physical data center—whether they’re truly ready to or not. As the results of the recent SolarWinds IT Trends Report 2017 show, organizations have migrated applications, storage, and databases to the cloud more than any other area of IT in the past 12 months.
This probably comes as no surprise: cloud computing is a compelling and exciting alternative to traditional IT and you can expect organizations to drive even further in the cloud in the years ahead. Still, patterns of implementations and challenges associated with the cloud and hybrid IT are beginning to emerge that can help you and your business better manage your infrastructure in the cloud. On that note, here are three emerging trends in systems migration and how they can impact your cloud strategy.
To the Cloud… and Back?
According to the report, 95 percent of IT professionals reported migrating some part of their infrastructure to the cloud in the past 12 months. Despite this amazing race to the cloud, 35 percent of those same respondents said they had also ultimately moved workloads out of the cloud and back on-premises in the past year, either due to security/compliance or performance concerns.
This begs the question: why were those applications migrated in the first place? Most organizations have implemented a virtualization strategy in some shape or form by now, which makes a lift-and-shift into the cloud much more achievable. But what’s much more likely is that these applications and other pieces of systems infrastructure are simply being caught up in the sheer enthusiasm of cloud migration. The old on-premises frustration of needing to scale but experiencing cost, resource, and management restrictions no longer applies when you move to an elastic-scale platform like the cloud, and for many, it’s too tempting to ignore.
Still, while it’s easy to be dazzled by the cloud’s benefits, not every workload is a great candidate for the cloud. Ultimately, this trend illustrates that pre-testing and workload performance and security considerations prior to migration—which should be a cornerstone of any migration strategy—are coming second to speed of deployment. To better work with business leadership and avoid realizing too late that a workload does not perform better in the cloud, you should look to participate in cloud conversations early (and often).
To start, your organization should be prepared to properly test any workload, application, or piece of infrastructure prior to migration to accurately gauge how it will perform in the cloud and what support you will require from the cloud service provider (CSP). A comprehensive monitoring tool that provides visibility into not only your on-premises systems, but also those in the cloud, should be implemented to help establish baseline performance metrics that make it easier to identify whether a workload belongs “on the ground” or in the cloud.
No Visibility or Control, No Service
In order to be successful as an IT professional, you need three things: responsibility, accountability, and authority. The first two are a natural part of your job, but in the era of hybrid IT, authority is what you’re constantly fighting for. In fact, over half of IT professionals reported that a lack of control over the performance of cloud-based workloads was a top challenge and is still a considerable barrier to migration.
When you think of control and authority over your workloads, the scenario you’re most likely familiar with is when a workload in the cloud begins to experience performance degradation, and although your owned systems and alerts indicate everything is roses, the cloud provider insists the problem does not stem from their services, either. Who’s to blame, and how do you validate your suspicions that it’s actually the CSP?
What’s almost as good as authority is visibility. If you’re able to see the problem in question and communicate it to the provider down to the most minute detail, it’s a much faster route to resolution, which in turn alleviates some of the stress associated with migrating a workload to the cloud. To that end, “trust but verify” should be the IT professional’s mantra in the year ahead, as organizations work to identify how best to maintain an element of control and visibility into workloads and applications that are hosted in the cloud. It will be critical to leverage comprehensive hybrid IT monitoring, beyond what is typically offered by cloud service providers, to ensure you have enough data and visibility to truly understand how workloads are performing in the cloud and the reasons for that performance.
You should also work with business leadership to make the case for multi-region or multiple cloud strategies to avoid catastrophic downtime in the cloud due to a single point of failure. Remember: this is the same lesson we learned from internet service providers and our off-site data center vendors before the advent of cloud. It’s better to be safe than sorry, so make sure you have a management and monitoring strategy that protects your data and delivers a strong end-user experience.
You’re Never Done Learning
One of the primary challenges we face today is staying abreast of new technology—how it works, how it interacts with other systems, and most importantly, how to manage it all. Complexity remains both a barrier to cloud adoption and a key challenge for those who have already migrated, and—again, according to the 2017 report—many administrators admit that they feel IT professionals entering the workforce don’t have the skills to successfully do their current job.
But that doesn’t mean these skills can’t be taught, learned, and implemented to better equip you to manage hybrid IT environments and oversee systems migration in the cloud. Think back to how you skilled up when you first became an IT professional: trying to build a single instance of something and seeing how it went, then tearing it down and building it again. Back when we were all at the start of our bright IT careers, practice made—and still makes—perfect, after all. You should also look for opportunities to use new features included in vendor tools you already own, such as automated functions, or try your hand at something like migrating an on-premises test server to the cloud to better understand how lift-and-shift techniques work.
The rate of technology abstraction isn’t slowing down. It’s important for businesses and you, as the IT professional, to “fail fast” (which does not mean actually fail, but rather, discover points of failure as quickly as possible). This involves constantly testing and implementing new solutions, because in today’s on-demand environment, availability, durability, and an acceptable response time from the end-user perspective are expected no matter where an application service is hosted or delivered from. This requires a comprehensive understanding of the technology you’re tasked with managing and migrating. You should work step by step and explore as many technology avenues as you can—containers, microservices, serverless computing—to gain meaningful, usable skills that will drive a more proactive, efficient, and effective cloud strategy.
We are in a new era of work as organizations of all sizes are implementing cloud computing to better meet the demands of a modernized workforce. In looking to the year of IT transformation ahead, the rate of technology abstraction promises to increase systems migration and hybrid IT complexity, requiring you and your business to be prepared for the shift in management and monitoring requirements. The above trends in systems migration to the cloud not only help paint a portrait of the modern hybrid IT organization and the us, the IT professionals who manage them, but provide key considerations for your company to evaluate and leverage when crafting your cloud strategy.
Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Penton.