As Cloud Expo approaches, the ongoing, evolving open-stacks scrum emerges as one of the most interesting stories in the business.
The high-level view finds three unique strategic ploys:
OpenStack launches its foundation, with IBM, HP, RedHat, and now VMware (primarily because of Ncira, we hope) as key members.
CloudStack pins its hopes on being an Apache Software Foundation project
Eucalyptus buddies up with Amazon
OpenNebula, led by OpenStack pioneer Christopher C. Kemp, is also in any open-source discussion, as it continues its strategy as a cloud-management toolkit.
Analysts and pundits enjoy viewing the world in term of absolute winners and losers, and indeed, some enterprise shops will make a sole winner out of one of the open-source cloud stacks. The reality is more complex, as one will no doubt find any or all of these products within many enterprises within a few years.
VMware is often said to be “the enemy” of open-source, although in many instances (so to speak) its foundational product is not said to be cloud at all, but merely a nice way to achieve server consolidation.
Meanwhile, the true enemy is the idea of a “cloud in a box,” offered by a company that has its core database technology in 90% of enterprise shops in the US, and likely, the world.
I don’t have a problem with the premise that cloud can be on-premise, but it must be decoupled into services, virtualized, metered, and highly elastic to be so. The same should be said for off-premise cloud as well – can’t live on hosted services alone.
Otherwise, we’re just fooling ourselves. Cloud computing’s promise is to let us use our IT resources wisely – one doesn’t need ideology or religion to be wise, but we can’t just sing the same old song either.