Cloud lock-in: What firms can be doing now to avoid the traps

Cloud lock-in: What firms can be doing now to avoid the traps

Summary: A lack of standards to facilitate moves between cloud providers doesn't mean companies are helpless in the face of vendor lock-in.


Even though a fear of lock-in is hampering uptake, cloud customers could still be doing more themselves to minimise the pain of moving providers.

By applying certain measures when, say, picking cloud services or designing cloud software, firms can go some way towards compensating for a lack of standards, according to Dr Chris Harding, interoperability director at vendor-neutral industry consortium The Open Group.

"Some of these things apply at the point when you're choosing a cloud service, and there's not a lot you can do after that once you've made your choice," Harding said.

"Other things apply to the way you design applications that use the cloud. Obviously, right through your application design and further modifications you can bear those things in mind."

Detailed advice and recommendations are contained in a free guide to avoiding cloud vendor lock-in just published by The Open Group.

Harding said the cloud consists of so many elements, including public, private and hybrid arrangements and SaaS, PaaS and IaaS, the specific approaches individual firms can adopt to improve portability will depend on their particular usage.

"For example, if you're designing applications that use the cloud or cloud services, make them loose-coupled. Don't go for ACID transactionality unless you absolutely have to," he said.

"If you're working at the platform level, think about — and this is something that has been under discussion in the technical community for a while — whether you're going to use SOAP and the web services protocols or are you going to use raw HTTP and JSON, and there are some considerations as to which might be your better approach.

"If you are going to be managing the cloud, look to see whether the service supports any management standards — that sort of thing."

Only picking cloud providers that offer vendor-independent programming interfaces is something that only really applies when an organisation is choosing a cloud platform, according to Harding.

However, producing clear, human-readable descriptions of services could be carried out even by those organisations that have already bought into a provider's product.

"That's actually something that comes right at the other extreme because you can even do that after development although it's obviously better if you do that at the time," he said.

Dr Jeff Jaffe, chief executive of the World Wide Web consortium, said recently the growth of the cloud industry is being held back by a lack of standardisation, which is reducing interoperability between competing cloud services and slowing adoption due to fears about lock-in.

However, The Open Group's Harding believes that at some point the cloud will be subject to a more extensive range of standards, which will help ensure greater portability and allay user concerns.

"I think it will happen. It's part of the natural progression of the way standardisation works that you get a period of innovation — which we've certainly had with cloud — and during that period of innovation you don't really want standardisation because it's a block on progress," Harding said.

"But then there comes a point at which differentiation for the sake of it is a bad thing and I think we're reaching that point now."

He said there are signs that take-up of cloud is beginning to be constrained by people's concerns over lack of portability and interoperability.

"At that point you start to see standardisation happening. Obviously it can go at a greater or lesser pace and go further or less far. What we're trying to do with this guide is accelerate its pace and make it go to as greater extent as possible."

Topics: Cloud, Enterprise Software

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • There's no free lunch ...

    Innovation means doing something different. Commoditization means doing everything exactly the same. When cloud gets commoditized then we will rapidly go down to 2-3 suppliers who compete on decimal points. Everybody else will move on to some other business offering, and portability will be moot because there is no real competition to choose between.
    terry flores
    • Seconded

      Then the price fixing will kick in and private clouds will be the next move.
      • Thirded.

        Vendor lock in is inevitable, because vendors will eat each other, until you have Microsoft, Amazon, and Google.
  • Does he know what those words mean?

    "'For example, if you're designing applications that use the cloud or cloud services, make them loose-coupled. Don't go for ACID transactionality unless you absolutely have to,' Harding said."

    . . . does Harding know what either of those words mean?

    (Hint: They're unrelated. ACID is for transactions, loose coupling is for components.)
  • To all the smiley happy people on thread...

    It may not be as bad as you predict.

    There is a potentially a better chance to avoid vendor lock in with cloud products than there has been for many other infrastructure based products. There are two reasons I can think of and both revolve around understanding the level of abstraction in the architecture. And that means actually understanding it and getting people involved to set it up properly not just paying for the latest platform/model/product and expecting it to work perfectly for any given scenario.

    The two reasons:

    1) Abstraction from the cloud infrastructre vendor - both OpenStack and full blown vCloud are being implemented by multiple infrastructure vendors. They may have slightly quirks in implementation but if you read the labels carefully you should be able to move between infrastructure vendors of your chosen cloud eco-system. Or between public and private. This sounds similar to a where you could have only 2 or 3 big eco-system players which comes close tot the argument that lock in is unavoidable except the the infrastructure vendors would be offering different support packages underneath the ecosystem from fully managed 24x7 phone to no-support-diy. This will then create some choices and avoid complete lock in.

    2) Abstraction from the eco-system - this is where it gets more complicated but truth is that if you do the hard work at the beginning you will have flexibility later on. You need to get into the "infrastructure-as-code" mentality in a big way. Get familiar with a configuration management tool like Puppet / Chef / Ansible / Salt and write descriptions for each piece of infrastructure. The annoying bit is then having to find the right orchestrator and bootstrap tools for the ecosystems, but once done you can affectively move between eco-systems, as a migration, as a burst, or even as failover or load balancing between data centres.

    Some hard work at first to understand the abstraction but potentially no lock in. - except maybe for your configuration management tool ;)