These days, more often than not, typical migration projects mean moving applications or functions from an on-premises system to the cloud in some form or another. Migration has always been tough enough, requiring lots of pre-cutover planning and weekend work, coordinating a bunch of moving parts, as well as a lot of hand-holding for affected employees and executives. (Especially executives!)
Now, the direction of movement in many migrations is in the direction of the cloud, and despite all the talk of how simple and easy cloud makes things, it really doesn’t make things any easier for the people overseeing the migration.
If anything, migration failures have seen a dramatic rise over the past few years, in line with the growing shift to cloud applications and services. A survey of 1,598 IT professionals, recently released by Vision Solutions, finds the incidence of “migration failure” rose 42% in the two most recent years the survey was conducted — rising from from 36% of IT managers reporting failures in 2014, to 44% in 2015 and 51% in 2016.
So what gives? The cloud — which gives everyone and anyone license to make their own IT messes — has made things much more complicated, the survey’s authors surmise. “Technology professionals are shaping and tiering the data center and want to make deliberate decisions about what software to move to the cloud,” they state. “At the same time, business units maneuver around IT to gain more agile cloud-based applications, leaving companies vulnerable. But IT has to figure out the best way to inventory and manage these apps, rather than trying to root them out.”
About 25% of professionals seem to be aware that business users run cloud applications outside the control of IT, while 33% admit they just don’t know who uses what. Two-thirds of survey respondents now use cloud in one form or another, but managing these environments is still an inexact science. For example, IT professionals lack consensus about who is responsible for protecting data and applications in a public cloud, the survey finds. About 43% believe cloud providers are ultimately responsible, while 39% believe internal IT departments should be in charge.
At the same time, migrations — cloud or no cloud — have never, ever been easy. “Migrations often involve different types of hardware and software assets, planning, testing, staffing, and scheduling, so it’s no surprise that they can fail,” the survey’s authors state.
The survey also finds largest companies (1,000 employees or more) were more likely to have experienced a migration failure (60%) versus 44% for all others. “No doubt, large organizations have more complex systems and are migrating many servers and databases, as well as applications,” the survey’s authors explain.
The biggest issues encountered with migrations include 44% reporting that their staffs had to work overtime (no surprise there!), coupled with system downtime (42%). How much downtime are we talking about here? The survey finds 83% of IT managers report having some degree of downtime due to a migration, and 58% reported migration downtime of an hour or more,.
Migrations often don’t happen as planned, either. Two-thirds of the IT managers surveyed report they have had to postpone migrations, mainly due to concerns about downtime. The prospect of working overtime (read: weekends) also did not excite staff members for some strange reason. In fact, the majority of IT professionals worked an extra 25 hours or more during migration.
Of course, there are accompanying pains for the business, especially those still on outdated hardware and software: “performance degradation, operational inefficiencies, data loss, equipment failures or added costs as leases overlap,” the survey’s authors add.
The pain points cited in the survey include an inability to start applications on the new server in the required timeframe (60%), and a lack of testing resulted in late discovery of issues (39%).
“These findings indicate that the root causes of failed migration are likely poor, unrealistic planning and goal setting, and faulty testing procedures,” the report’s authors conclude. “While inadequate tools might account for some of these migration failures, it’s clear that the human factor – including training and planning – plays a vital role.” Organizations successful in their migration efforts “plans better, tests earlier, and has access to a migration tool that enables continuous uptime during migration.”
And along with this advice, there’s plain common sense: value everyone’s feedback at all stages of the process, keep everyone in the loop and informed about what to expect, and commuincate how the new platform is going to improve their lives.