Building the Healthcare Cloud is not for the faint of heart. Prior to becoming a healthcare CIO, I was the president of a Cloud Computing consulting firm from 2008 to late 2011. We helped IT providers build cloud platforms and applications, and provided services to organizations looking to adopt cloud models.
There was one underlying reason that companies adopted cloud models: agility. You could provision new services from any computer with a browser, and you could scale up or down in a matter of minutes. Because the cloud is highly virtualized, you can move resources around the country for greater efficiency, and it typically has a very defined accounting model where you pay for what you use. In comparison, internal IT would take weeks to provision new services, require fixed assets, and put them in an aging data center. When asked how much it cost the business, there would be limited information.
Recently, we worked with a health system that had a portfolio of more than 800 applications. To help better standardize the system, we opted to move the system to a hybrid cloud. Based on that experience, below is a list of 10 things that should be considered when moving to the cloud:
- Security. Moving to the cloud does not allow you to pass responsibility for security and compliance to the cloud provider. It is still incumbent upon the health system to look at data protection, application security, business continuity, identity and access, incident response, network security, and physical security. Getting the BAA signed is only the beginning.
- Staff Development. You need to bring your staff along for the ride. Most will think that puppet is a semi-odd prop for entertainment as opposed to the open-source service platform that allows you to access data center resources programmatically. There is a better way to provision and manage services; get your people trained and let them chart the course.
- Latency is real. Know it and understand it. On the flip side, it’s just a math problem. Too many people use latency as the excuse for why you can’t go to the cloud. They are wrong. Ask questions, do the math.
- Operating Cost. The cloud will probably not save you money. In many cases, you are just moving money from capital to an operating budget. You may save money due to increased agility, but my CFO would never let me talk about soft dollar savings. The main point here is that you will probably increase your operating budget when you move to the cloud.
- Your Data. Make sure you don’t move your data with a one-way ticket to the cloud. You may not use the round-trip ticket, but you want the option. Make sure you know how you will go about getting your data back out of the cloud vendors’ environment.
- There Are Many Clouds. There is a cloud for everything, but the most common I find are Infrastructure, Application, and Storage. They match our common IT stacks. And so the easiest move for us was to Box to modernize our file storage. This was a huge success; however, I wouldn’t say we moved the health system to the cloud once that was implemented. I considered us to be moved to the cloud when more than 60 percent of our technology services had rapid web-based provisioning, we could scale up and down as needed, and we paid for what we used and had a usage based accounting method around our computing resources.
- Identity. Build your identity architecture before you move to the cloud. It’s good to have this baked before you negotiate with vendors.
- Network. A high performing network is a key to any IT operation. The trick in the cloud is to extend your network with the same characteristics that you maintain for your internal network. For example, we hired a company which mapped out our WAN vendor connections to ensure that our carriers didn’t use the same physical paths. It’s not good when you thought you had a redundant path, only to find out they shared the same fiber line.
- Cloud Doesn’t Solve Your Underlying Problems. Nothing reveals the mess like moving houses. Same is true with moving to the cloud. If you don’t know what you have, you will have problems. Nothing reveals the lack of discipline in an IT shop like moving to a new environment. Know your environment before you move, or else you will compound the challenge of troubleshooting not only the move but also the problems you didn’t know about.
- Pilots Are Your Friend. This is true with any new solution or innovation. Break the problem down, conduct small pilots, prove the solution, solidify your learning and adjust your process or thinking based on the results. If I had one do over as a CIO, I would have done more pilots and fewer large implementations.
I hope this helps.