24 Comments

Summary:

Buyer beware when it comes to adopting OpenStack — or any open-source cloud — according to a newly released Gartner report. Long story short: evaluate OpenStack just like you’d assess a commercial offering, cautions analyst Lydia Leong.

Just weeks before for the big  OpenStack Summit, a new report cautions prospective users not to buy into the hype around this (or any) open-source cloud computing platform.

Gartner Research VP Lydia Leong suggests that prospective users proceed with eyes wide open and recognize that OpenStack, while promoted as an end to vendor lock-in can represent its own sort of “ecosystem lock-in.”  She recommends the use of a third-party cloud management tool or API  library that will work with multiple clouds.

OpenStack hype vs. reality

In the report (entitled “Don’t let OpenStack hype distort your selection of a cloud management platform in 2012″) Leong wrote that while people have been led to believe that because OpenStack is open source it is an “open and widely adopted standard.” The reality is different. She wrote:

“OpenStack is dominated by commercial interests, as it is a business strategy for the vendors involved, not the effort of a community of altruistic individual contributors. Some of the participants, notably Rackspace and other service providers are afraid of the growing dominance of AWS in the cloud IaaS market and do not believe that they have the ability to muster, on their own, the engineering resources necessary to successfully compete with [Amazon Web Services] at scale, nor do they want to pay an ongoing license fee for a commercial [cloud management platform]  like VMware’s vCloud stack.”

And, perhaps of more concern to the OpenStack faithful, Leong added: “Do not plan the future of your data center with the assumption that OpenStack will be at its core.” Ouch.

On the other hand, OpenStack is fine for companies that consider themselves early adopters and are willing to do the blocking and tackling needed to field early code, according to the Gartner report, which is dated September 14, but was released broadly on Friday.

OpenStack: Linux or Unix?

One nagging concern about OpenStack is that it comprises the work of a lot of industry giants — initially Rackspace and NASA, but with IBM, HP, Cisco, Dell, SUSE, and now VMware jumping on board – that have their own competitive agendas. All of them are concerned, at least privately, with Amazon’s huge cloud clout, but all of them also compete with each other. The question then becomes whether OpenStack will really become the Linux of cloud — widely adopted by competitors but also broadly interoperable — or the Unix of cloud, with companies churning out slightly different and semi-interoperable versions.  In short: stuff happens.

The formation of the OpenStack Foundation – now pretty much complete — alleviated some concerns that Rackspace held too much influence over the effort. And the latest  “Folsom” release out  this week shows less Rackspace influence according to some observers. Still worry persists.

Some other common sense advice from Leong: Evaluate OpenStack or CloudStack (backed by Citrix, once a big OpenStack backer)  or Eucalyptus or OpenNebula or whatever open-source cloud infrastructure just as you’d check out any vendor’s offering.

  1. Speaking of being “dominated by commercial interests,” I wonder who might have commissioned the Gartner report?

    Share
    1. Lol. :D

      Share
      1. any bets?

        Share
      2. I double the LOL! One might bet it was Overlord Landlord Bezos, but I would not put it past Stevie Ballmer and company to have commissioned the Gartner report (am not sure Bezos boy cares so much about what Gartner says but Ballmer and company probably care quite a bit)

        Share
    2. When I first read the report I wondered the same thing. Can’t say I question the results. It has a bias but can’t say that I found anything false. I’ve had a similar perspective on OpenStack. It’s great for providers and OEMs but I’m not too excited about it in the enterprise.

      Share
  2. Gartner states the obvious by telling companies to evaluate software they plan to use.

    Share
  3. when is the last time IBM and DELL worked together? or HP and IBM for that matter? competing agenda’s? thats an understatement.

    Share
  4. Awesome, how this carefully avoids the truly open and much more proven alternative OpenNebula.

    Share
    1. i should have listed OpenNebula as an option. In fact i will tweak. thanks.

      Share
  5. This article is hype itself! Talk about yellow journalism. There is no hype with Open Stack what so ever. You have a choice to use it which comes with the freedom of access to the source code. According to the original Gartner article the Linux kernel itself shouldn’t be used at the core of your data center. It’s exactly the same…the most contributors to the Linux kernel are commercial developers with an agenda as well. Free and Open Source software allows them to leverage the hard work of others while contributing and not merely siphoning off the community.

    Barb they are just blowing smoke

    Share
    1. No hype? sorry can’t agree with that. But hype comes with marketing — it’s to be expected.

      Share
  6. Wow they should read a bit more openstack has a new member of the foundation VMWare hmmm seems that if Amazon applied they would be accepted too. So the fear seems to be misplace in this uninformed opinion article

    Share
  7. This could be the most worthless article about OpenStack ever written. Congrats GigaOM. Regurgitation of useless analysis from a Gartner “analyst” who has little to no clue about the OpenStack ecosystem is just not worth publishing.

    Share
  8. I think Lydia Leong’s comments about not placing OpenStack at the core of the datacenter are probably over exaggerated, considering that Hadoop is most certainly dominated by certain key players like Cloudera, IBM, MapR, etc. but that doesn’t necessarily mean that you shouldn’t let Hadoop be at the core of your Big Data infrastructure.

    Regardless, having an infrastructure orchestration / cloud orchestration layer is important, and OpenStack alone doesn’t provide that — that’s where things like Puppet, Chef, or Ironfan come in. Particularly because, if coded with proper best practices, it can be relatively straightforward to migrate your infrastructure from datacenter or cloud to another.

    (My company, Infochimps, helps develop Ironfan, which couples Chef with Fog, service discovery, integrated monitoring, system diagram -level orchestration, as well as cloud abstraction — the idea is to bake those best practices into the software framework itself so that you can achieve that “third-party cloud management tool or API library that will work with multiple clouds” Lydia refers to).

    Share
  9. I think, being cloud agnostic is a common strategy adopted when you want to implement a PaaS and want it to work under a federated cloud model. There are options like jcloud, deltacloud, the apache libcloud library. So its a big ado about nothing. We are seeing a co-innovation paradigm in the open innovation model with several enterprises sharing the development efforts. This has happened with projects such as Apache web server for example, and that has not deterred enterprises from widely adopting the technology. How can there be a lock-in around an ecosystem that is so diverse an open? Also Openstack continues to support both API’s, native Openstack and Euca2ools.

    Share
  10. I say beware.. don’t use a computer because it will become an eco-system lockin.

    Share

Comments have been disabled for this post