Jumping into the cloud

By Laura Haight
Originally published as the Digital Maven in Upstate Business Journal, Oct. 27

Every business has a toe in the cloud with internet-based software and online services ranging from email to accounting as the most common starting points. But for many it is hard to let go of the comfort of knowing you've got your data, in your office, under your control.

Kind of the way you feel about the stock market: 'Great, I made some money now I want to take it and stuff it in my mattress!'

But we are on a seemingly inexorable path toward businesses migrating all their data, files, and functions into some form of cloud. A Cisco Forecast Report estimates that by 2020 the amount of business data living in the cloud  will have grown more than 21 percent over just five years. Cloud data centers are killing it over their traditional counterparts with a five-year growth trend of nearly 30 percent.

IBM announced two new services just last week designed to make it easy for business clients to migrate to the cloud.

Mountaineer George Mallory is best known for three immortal words spoken in 1923 in answer to a reporter’s question about his quest to reach the peak of Mt. Everest: “Because it’s there.”

Many decisions - invigorating, inspiring, and disastrous -- have referenced this idea. While it makes a great sound bite, it’s generally not a great justification for a business decision.

2018 is expected to be a big year for cloud migrations. That may be great for you. But before you jump, consider:

What’s the business reason? There are a lot of business buzzwords association with moving to the cloud: agility, flexibility and, my favorite, digital transformation. But those are slogans, not strategies. Game this out the same way you would any other business decision: cost vs. benefits, risk vs rewards, soft benefits vs. hard choices. If you currently have your own internal servers, cost may be an issue when you factor in hardware replacements on a 5-to-7-year basis (servers aren’t cheap), as well as other network infrastructure expenses. Also consider the human factor: how well will the staff adapt, and how will our procedures and processes have to change.

Ask good questions, get better answers. Once you know what you expect to get out of a cloud migration, figure out how you are going to monitor and evaluate it. Setting those expectations will depend on getting good information from cloud providers and that will depend on asking good questions. Here are a couple:

• Ask for uptime reports and some historical tracking (are they getting better - or falling off). Uptime may be presented in 9s. A company promising 99.9 percent is estimating that your services will be down for 8 hours and 45 minutes a year. Sounds pretty good. But with 4 9s uptime, you get services all but 52 minutes a year. A company promising 6 9s, is expecting downtime to be 31.6 seconds a year. There’s a calculator for this. 
• How quickly can you restore your data. Many companies move to the cloud to increase security and improve disaster recovery prospects. But not all cloud providers are the same. You need to know how often restore points are created; the more frequent the restore point, the less data loss you’ll have in the event of a ransomware or business continuity situation. Ask how many recovery points are maintained. In the event of ransomware, you don’t necessarily want the most recent data restored, you will want the data from BEFORE the hack. You’ll also want to know how long it will take to complete a recovery.
• Since you will likely not start but moving your entire business to the cloud, you need to plan for growth. So ask cloud providers you are considering how easy it will be to add to your environment. How much control will you have?
• While you’re on that topic, ask your provider how they plan for growth. Disk and infrastructure are expensive, so a good provider who can grow with you effectively has to be well capitalized.

Start with a pilot. Yes, your company probably is in the cloud either with email or any number of online services. But a larger migration has many different issues and challenges. Don’t try to eat this elephant in one byte. Take a department or even a particular product or project to start with. Monitor and document this process so you can replicate it or learn to identify the potholes and avoid them as you move forward.

Evaluate staffing. A lot of cloud migration projects are based on a cost-saving strategy. That usually involves IT staff. Before you start counting those chickens, make sure you have evaluated not just the number of positions you might save, but what functions those staffers perform. Systems in the cloud still need support, user interaction, training and a whole lot more.

Security. Worried about the security of the data in your business? You probably should be. We all should be. But moving to the cloud will not make everything inherently safe. Depending on how much of a hardware or software investment you’ve made, you may get a significant improvement. But the path of least resistance for hackers and crackers goes straight through your staff. True security comes from developing stronger humanware. And that’s not something the cloud can do for you.

More articles on cloud computing