At CloudBeat, a panel of experts were in consensus about one thing and one thing only. Open cloud is a marketing-driven concept, and it’s not easy to distill into a simple sentence or two. “I don’t think there is a clear distinction about what it is,” said Chris Pinkham, cofounder and CEO of Nimbula, a company that delivers cloud computing software for building private, public or hybrid clouds.
“Open cloud means with relative ease you can replace any given components [whether it’s] hardware or software,” explained Randy Bias, co-Founder and CTO of Cloudscaling.
In minutes, the panelists got riled up about the purpose of vendor lock-in, the benefits of open source tools, and the ways that companies can move their workloads between different infrastructure providers.
Tension arose around the subject of “locking-in customers” year-on-year — while Bias argued that vendors should build solutions that their customers love and aren’t stuck with year-on-year, Pinkham made the case that a certain degree of lock-in is needed to maintain a steady rate of innovation.
“You can’t do anything new and interesting without locking yourself in [to a vendor] in some way,” Pinkham explained.
“My customers don’t want to be milked like a cow,” hit back Bias, who claimed that customers are too often paying $100 million in licensing fees to a behemoth like Oracle. “You could buy a company for that…or a 50 person noSQL team,” he said. Bias said it’s important that companies own their own code so they can kick their cloud providers “to the curb” if expectations aren’t met.
Fellow panelist Boris Renski, Co-Founder & EVP of Mirantis made the case that most customers will have a degree of lock-in, and will typically need to pay hefty fees to switch vendors.
What are your thoughts? Should customers be able to walk away at anytime? Is there a balance between vendor lock-in and control? And finally, how would you define the term open cloud?