back to article Dedicated (Local) Cloud Infrastructure-as-a-Service to grow almost 1000 per cent in five years

Analyst firm IDC has a new abbreviation for your cloudy dictionary: DCIaaS, which stands for Dedicated (Local) Cloud Infrastructure-as-a-Service. “This model is essentially a dedicated version of a publicly available cloud offering, modified to run on-premises or in a specially certified colocation environment, including …

  1. Lunatic Looking For Asylum

    Great News Suckers - On Prem Really Is Better

    Providing it's our kit that's on prem - not yours.

    The cloud providers must be wetting themselves with laughter over this. The tin vendors as well. It's a win win for everybody apart from the poor customer who's decimated their data centre, culled their staff and outsourced everything.

    Cue the architects going to the board telling them this is the future and polishing their CV for the next job.

    I *&^%ing hate this industry at times.

    1. Roger Kynaston Silver badge

      Re: Great News Suckers - On Prem Really Is Better

      It was always like this. I well remember sweating blood to learn how to do Solaris containers - it looked like a good idea only for Oracle to effectively kill off any wish to own Sparc kit. Then docker and the other container technology came along that was all shiny and new.

      And, what is the cloud if not a mainframe writ large?

    2. M-Singh

      Re: Great News Suckers - On Prem Really Is Better

      Having lead the charge with Inframod programs initially with solutions like VxBlock from VCE to SoftwareDefined stacks from Dell with commodity servers - Local-Cloud-aaS seems to be the natural evolution of this trend. The goal being to make the infra as invisible as possible to the business.

      If we agree thats the goal then why is this not a good idea?

      Obviously these solutions command a premium, but what your paying for as you put it, is outsourcing those traditional day2 tasks that traditional IT teams used to do as a feature of the platform backed by an SLA. I’m cognizant of the impact on real people, however from my analysis of customers with 300VM environments to 30K VM environments there’s some general trends for the next 5-10yrs that we can say with some confidence:

      - 40-60% of customer workloads can/should run in the Public Cloud and from an economics, operation, strategy, dont belong in the DC

      - of all these large populations of apps/VM’s, 70-90% fit this managed infra model

      - There’s still 10-20% of workloads that won’t be a good fit for HCI or public cloud without serious rearchitecture that will take a lot of time and might not even be worth it given the value some of those apps bring to the business. Therefore will still be powered by ‘3Tier’ traditional building of IT will still be around, skills required, just greatly diminished.

  2. jake Silver badge


    They've invented the server!

    What geniuses!

    1. IGotOut Silver badge

      Re: OMG!

      No, not a server, a cluster.

  3. Millwright

    Maths, dear chaps. 138M to 14B is a touch over 100x, which is 10,000%

    1. Lunatic Looking For Asylum

      Yes but but but it's :-

      "That’s a compound annual growth rate (CAGR) of 151.8 per cent."

      compound - all the bean counters and shareholders love that.

      They're still wrong though - 138 million to 14 billion is ~ 251% compounded...

      Happy days for the bean counters and shareholders.

  4. elsergiovolador Silver badge

    New but the same old

    The glorified server in the utility room.

    Who had to go to the office at night because the cleaners pulled the wrong plug to connect the hoover?

    1. Chris G

      Re: New but the same old

      History constantly repeats itself but each repeat has subtle differences.

      In this case, instead of Xerox renting copiers and making the client pay extra for toner, paper and a bloke to polish the drum, it's server rental and maintenance.

      Still with the lock in and inflated hidden costs tied up by a contract written on stone, I assume.

      1. Tom 38 Silver badge

        Re: New but the same old

        Beancounters love this one weird trick to reduce CapEx

  5. karlkarl Silver badge

    We still have all our own servers.

    Does that make us... modern, hip and with the times?

  6. Anonymous Coward
    Anonymous Coward

    Please define "cloud" for me....

    Can someone define this new version of "cloud" for me please?

    If it's on prem and its your hardware its not cloud by my understanding of the term "cloud".

    1. Felsesser

      Re: Please define "cloud" for me....

      Its still the same Cloud definition but managed locally, by the vendor. This mainly offers low latency benefits, residency requirements, and true hybrid- cloud capability. Having VMware, Xen servers, or hyper-V in your own data center, even if connected to a public Cloud vendor is not hybrid cloud, but a hybrid environment.

      (private cloud on prem + public cloud) = hybrid cloud

      Same Cloud computing definition in both private and public implementations.

      1. jake Silver badge

        Re: Please define "cloud" for me....

        IOW, selling refrigerators to the Eskimo/Inuit.

    2. Felsesser

      Re: Please define "cloud" for me....

      On-demand self-service. A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human

      interaction with each service provider.

      Broad network access. Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations).

      Resource pooling. The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of

      abstraction (e.g., country, state, or datacenter). Examples of resources include storage,

      processing, memory, and network bandwidth.

      Rapid elasticity. Capabilities can be elastically provisioned and released, in some cases

      automatically, to scale rapidly outward and inward commensurate with demand. To the

      consumer, the capabilities available for provisioning often appear to be unlimited and can be appropriated in any quantity at any time.

      Measured service. Cloud systems automatically control and optimize resource use by leveraging a metering capability1 at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.

      1. jake Silver badge

        Re: Please define "cloud" for me....

        So basically, increasing the size and scope of potential attack vectors in pretty much every aspect of computing, and laughing in the face of basic security practices, then.

        Doesn't sound like something I'd trust with my company's business.

  7. roselan

    10 to 20x cheaper

    Our baldyHairy sys guru made kubernetes work on our nutanix cluster (karbon). Turns out it's 10 to 20 times cheaper to run apps and services on premise rather than on azure offering. And that's before data extraction/bandwith costs...

    Plus hardware is managed by the vendor itself, so no 3am call in theory. It really looks like the best of both worlds for now.

    It's coming home!

    1. Robert Grant Silver badge

      Re: 10 to 20x cheaper

      > Turns out it's 10 to 20 times cheaper to run apps and services on premise rather than on azure offering

      That's pretty good news. Care to share some figures?

    2. Sin2x

      Re: 10 to 20x cheaper

      That's all nice and dandy while he's there to support that zoo he built. And when he's not -- well, you'd better hope he documented the monstrosity rigorously. From business point of view having a well-established system with availability of specialists on the market and presence of support -- like, say, Azure Stack, makes much more sense than the pay-less-now-pay-google-times-more-later DIY approach you seem to champion here.

  8. RowlyB

    Seen it before

    Didn't Sun Microsystems already try this under the banner.. Capacity on Demand

  9. Kev99 Silver badge

    Is this just a revised, not as secure, version of the old "dedicated lines" used by industry for decades before DARPA released DARPAnet into the wild?

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like