As we’ve become involved in the OpenStack community and the Trove project in particular, the enthusiasm that we’ve seen has been tremendous. We’ve also seen a lot of interest in the ability to take advantage of Trove to run database as a service in private clouds.
To try to quantify this enthusiasm a bit and understand some of the reasons behind it, we decided to conduct a survey along with some friends in the open source community. We polled over 500 users, mostly in medium to large enterprises about their current usage of public and private clouds and, in particular, their use of databases in these environments.
Many of the findings were expected; (Amazon is still the most popular public cloud, by far; Relational databases are still dominant but with NoSQL growing fast; The number one reason for using a cloud is to save money.) A few of the results of the survey were a bit more surprising, for example:
- OpenStack is catching up to VMware as the preferred private cloud platform even though it has only been around for a few years. Of the organizations polled that are already using a private cloud, more than one-third now use OpenStack.
- More than half of the developers we polled would be likely to use a DBaaS in their company’s private cloud, but only about a fifth of the organization have implemented it or are planning to in the next 2 years.
- Given the popularity of Amazon’s cloud, I thought AWS compatibility would be important to people, but only 9% of our respondents said it was important to them
- The most popular workload on the private cloud is “database” and it is the second most popular on public clouds.