+Comment OpenStack is the IT industry, sans Amazon, Google and Microsoft, coming together to craft an open-source cloud OS alternative to ... wait for it ... Amazon, Google and Microsoft.
It is because the terrible trio's public clouds threaten to eat the enterprise data centre – supplying the IT industry's lunch, and dinner, and tea and breakfast.
When keynote speakers congratulate the audience, you just know they're having to rally the troops because not enough progress is being made.
Here are snippets from the keynote presentation by Jim Zemlin, exec director of the Linux Foundation:
Open Source is redefining tech industry ... There's too much software to be written for any one organisation to write on its own. ... 80 per cent of code in Sony smartphones is open source ... Open source is the dominant form of software development ... Are you ready for the next open source blockbuster? ... OpenStack without question is a blockbuster ... Every market Linux has gone into it's conquered.
See what I mean? Self-congratulatory blather, but then this is the six-monthly OpenStack papal congress. The audience are the believers and just love their popes.
This is a reprise of the Unix industry versus Microsoft writ large. Unix variants more or less collapsed into Linux so Linux lovers in the broad band of IT suppliers love OpenStack and its antipathy towards the proprietary Death Star clouds of Amazon, Microsoft and Google.
X-Open collapsed shortly after it amazingly certified Windows NT as an open operating system; it always was a bizarre organisation. OpenStackers with long memories will recall that and resist providing S3 compatibility in OpenStack releases. The Juno release came out last with with a Kilo release scheduled for April next year.
Lots of startups are jumping on the OpenStack bandwagon, hoping it will drive them towards fame, glory, an acquisition or an IPO. More or less all enterprise IT vendors support it, and why not? There's no downside and they all pretty much hate and fear the hyperscale public cloud suppliers. So pump a few hundred thousands of marketing dollars in, have marketing pour out warm words, get product engineering to support OpenStack components in some way, and, if it comes good, reap the rewards.
But the philosophy of OpenStack is the philosophy of open source and that is free software. We have the weirdness of proprietary software and HW/SW providers supporting open source OpenStack. Oh yes, some storage marketeer says, our proprietary array software supports OpenStack Swift and so everything changes and everything stays the same.
Let VMs in Linux-running CPUs call up storage and networking resources through OpenStack, but those resources will most probably be delivered by proprietary systems. All OpenStack does is layer what could become an industry-standard open-source abstraction layer over them, which is worthwhile in itself but of limited utility.
OpenStack is a worthwhile effort and if it succeeds in cramping the growth of Amazon, Google and Microsoft's public clouds that will be a very good outcome indeed we we'll have 4-way competition entrenched in the public and private cloud-supplying marketplace.
2010: A Space Odyssey
But it is not all roses. NASA was one of the two original founders of OpenStack with RackSpace. The space agency contributed its Nebula private cloud compute and storage software to Open Stack along with RackSpace's object storage technology in 2010.
It has evaluated private cloud provision and its OpenStack focus since then and withdrew from active OpenStack development (PDF) in July last year, saying it would focus on - heresy alert - public clouds. In its audit report NASA said:
While the private cloud alternative enables agencies to manage their critical IT services and control access to sensitive data directly, these benefits come at the high cost of owning and operating data centres. Conversely, the public cloud alternative frees organisations from the expense of data centre ownership but requires that they effectively manage contractor performance to ensure key business and IT security requirements are met.
The NASA report said: "After investing $19.7m, NASA suspended funding for Nebula in 2012 and shifted its cloud strategy to the purchase of cloud services from public providers."
Come back, NASA, all is forgiven
The big, big thing is that private cloud requires private data centres and the public cloud does not. Were NASA to return to the OpenStack fold that would send a resounding message throughout the IT cloud supply industry. So far it has not.
It could well be that NASA was simply incompetent at managing its own private cloud initiatives and we shouldn't draw any negative conclusions about OpenStack from its withdrawal. It happened a couple of years or so ago and the OpenStack software has moved on since then.
But we must also remember suppliers like VMware would far rather customers used its overall data centre vSphere-centric resource-managing software rather than OpenStack. (Apart from VCloud Director which we're told is directly equivalent to what OpenStack provides.) What proprietary IT supplier does not prefer its customers to inhabit a walled garden with a price for admission and presence?
We hear that business IT might obey a CIO-level use-OpenStack directive and then use OpenStack links/support from installed proprietary suppliers like VMware to carry on using their proprietary kit because it meets operational business needs far better than a pure OpenStack, KVM-only, Linux-only route would do.
Perhaps the two main proprietary IT supplier sponsors of OpenStack – HP and IBM – will do for it what IBM did for Linux several years ago, and legitimise it.
The attendees at the OpenStack summit here in Paris would be greatly encouraged if the two monarchs of the IT industry, IBM's Ginni Rometty and HP's Meg Whitman, carry on dipping their hands into their wallets and showering billions of development, support and marketing dollars to encourage OpenStack adoption.
HP's Helion OpenStack cloud and IBM's SoftLayer OpenStack clouds are giant-ish steps in the right direction.
If they do this and get mass adoption by enterprises of their clouds in preference to those of Amazon, Google and Microsoft, then OpenStack will have justifiable optimism about its future prospects. If they don't do this, if there is only OpenStack lip service, then it's effectively going to be "Off with their heads" – with the executioner's axe wielded by Messrs Bezos, Page and Nadella. ®