BACKGROUND IMAGE: iSTOCK/GETTY IMAGES
Unless you've been stuck working on nothing but Cobol programs for the last five years, you have to have heard...
of the cloud and all of its promises. The cloud lets you deploy faster, reduce costs and spend more time on IT core competencies, just to name a few benefits. It's definitely a game changer, and like many people in IT, I'm particularly a fan of what cloud databases can provide me and my company.
That being said, as a long-time database administrator (DBA), I tend to be protective of my company's data. We have to make the appropriate data available to those that need it to do their work or buy our products and services. At the same time, we have to secure the data to keep it away from those that shouldn't have it. Data breaches are detrimental and costly to any company that experiences one. There are a few things to think about before you rush to turn your on-premises databases into cloud databases.
I try to be very thoughtful in my approach before I start pushing data to the cloud. The cloud makes it easy to move your data there -- that's part of its promise, after all. With just a few clicks and a swipe of a corporate credit card, you can have a full-fledged Oracle cloud database up and running in no time. Sometimes, though, that ease of use can be a big downside as much as a benefit. In the cloud, it can be too easy to spin up a new database environment without thinking. An organization deploying to the cloud needs to be mindful of the costs to keep them from spiraling out of control.
Pay as you go cloud models can be a great way to reduce Oracle database costs, especially for startups and projects just starting to take flight. If you're only using a few CPU cycles daily, you can save money by leveraging the cloud. But it doesn't take long for successful startups and initiatives to increase their resource utilization. This can also quickly increase the cost of cloud database services.
And just like on-premises databases, cloud databases can be resource pigs. It only takes one bad SQL statement to drive CPU utilization sharply upward in a cloud environment, bringing cost increases with it.
Keep cloud databases legally grounded
Regulatory issues may also complicate decisions to move databases to the cloud. Different countries have different regulations for managing personal data -- and complying with them can be a challenge, depending on where the cloud database systems you're using are located.
In addition, the regulatory landscape is in flux. For example, the European Court of Justice invalidated the Safe Harbor agreement for moving data between the European Union and the U.S. in October 2015 in the wake of Edward Snowden's revelations about the surveillance programs of U.S. intelligence agencies. A replacement agreement called Privacy Shield took effect this month, but it could also face a court challenge in Europe. When you set up cloud databases, make sure you aren't putting your company in a bad legal position.
You don't want to lock your company into a particular cloud platform, either. We'd all like to believe that our cloud provider will be with us for life, or at least until the end of our career. However, we've likely all worked with our fair share of vendors that we were very happy to sever ties with. You need to prepare for the possibility that you may switch cloud providers sometime in the future and ensure that you'll be able to get your data out of that provider's cloud.
Stay secure while your cloud database flies high
DBAs need to be vigilant about security as well. Before you move a database to the cloud, make sure you know the basics of how security will work there. Find out if your cloud provider will be providing regular security updates or if that is going to be your responsibility. Not understanding your role in cloud security can lead to serious problems if you aren't careful.
Every DBA knows the importance of backups to provide further protection for data. Talk with your cloud provider in advance and find out how to back up your cloud databases. It's also important to keep an eye on how much extra the backups may cost you and whether backup copies are going to be stored in the same physical location as the primary database.
Another big issue with the cloud is latency. Companies that run applications internally may start experiencing latency issues when they move to the cloud. This is because the data store will sometimes be moved to a location geographically distant from the local data center. Make sure the potential added latency doesn't negatively affect the ability of a cloud-based database to support the business.
Cloud providers tell sometimes tall tales of how your life will be better in the cloud. It may turn out to be, but do your homework before you make the leap with your Oracle databases.
OAUG president Patricia Dues discusses Oracle's cloud strategy
Cloud adoption goes up with third-party vendor support
Oracle IaaS challenges AWS with elastic compute cloud service