back to article IaaS is OVER, ladies. Time for OpenStack to jump clear

Data. Was ever such a single word filled with such promise – and heartbreak – for IT firms? In the cloud, succeeding in getting hold of people’s data means these people are incentivized through inertia to stay with you. That’s the promise, at least. The rub? Actually making a successful business from people’s data. The act of …

  1. K

    Finally an article that realises the reality..

    But even then the article tries to put a positive spin on Cloud..

    I've said it before, Cloud is what the name suggests - mist and vapour with a sprinkling of BS mixed in.

    Perhaps we can now stop getting it funnelled down our throat as a panacea to all IT woes - Just because the fanatics scream it over and over, doesn't make it true, and beside brainwashing doesn't actually work.

    I won't deny it, Cloud has a market and will be an excellent solution to some problems that require cheap storage or flexible hyper-scale computing. But for 90% of SME server/computing has little appeal.

    1. Anonymous Coward
      Anonymous Coward

      Re: Finally an article that realises the reality..

      This won't be a popular opinion on here, but I'd turn your figures on their head - I'd say 90% of SME server/computing requirements will (maybe not today, maybe not tomorrow, but in a couple of years when cloud infrastructure has matured a little more) find that cloud is a great fit for them. It's the 10-20% of "unusual" workloads that cloud will struggle with. To coin a phrase, businesses are in the business of running their businesses, NOT of running an IT infrastructure. Let those who are good at it do that and let the businesses concentrate on their core competencies. Everything gets commoditized over time, and cloud is just the next step in that direction for IT.

      1. K

        Re: Finally an article that realises the reality..

        There is a lot of if and when in your argument. But there is a reason the bulk of people in a position of responsibility for IT don't jump for it wholesale, because they understand it has some strengths, but likewise a lot more weaknesses.

        There will be a time when its more popular, try 5+ years, but even then I can pretty much guarantee its not THE next direction, its simply yet another tool in the toolbox, that will be used as and when it offers the best solution.

        Additionally I state my previous argument, Cloud doesn't exist, its mist and vapour, but hosted services on the other hand...

        1. Anonymous Coward
          Anonymous Coward

          Re: Finally an article that realises the reality..

          'The bulk of people with responsibility for IT" will be out of a job. Infrastructure people will be first in line. Which is an awkward message on El Reg. People who consider 'cloud' as being the same as 'hosted service' will almost certainly be right at the front of the exit line...

      2. Nate Amsden

        Re: Finally an article that realises the reality..

        I think cloud is a good future for the SME, but that cloud is SaaS, not IaaS which is what this article is referring to. IaaS cloud (even as a customer) still requires pretty significant knowledge to operate, of course SaaS does not.

        Another thing that is driving cloud adoption I'm sure is just lack of supply of (good) IT folk. If you can't find someone(s) to run infrastructure right then well you are probably better off not trying. Unfortunately (most) companies seem to think it's not worth paying decent money ($200k+) for people to do these kinds of things when in the end it could(depending on person etc) save them million(s) per year. It'd be different if there was more supply (of people) on the market, but it seems clear to me at least on the west coast of the U.S. there is not. I keep seeing and hearing stories about companies paying high six figures per month to cloud providers and either don't care that they can save a lot, or they just think that is "normal".

        There was one good article here from someone that said something along the lines of when your cloud spend crosses $20k/mo then you need to start looking at alternatives. For my org cloud spend was about to explode above $100k/mo when we moved out 3.5 years ago, and my previous company was one of the ones spending $300-450k/mo on cloud (which I showed could be done in house with a 3 month ROI, but ultimately the company wasn't interested, I left, and it collapsed not long after).

    2. randhawp

      Re: Finally an article that realises the reality..

      Cloud for a SME means, 0 capital costs, 0 IT maintenance costs, 0 IT hardware upgrade costs.

      We are already seeing cloud based services in accounting, time keeping, inventory, document management flourishing.

      For a small mom and dad company, having a server running in the back room is a hassle.

      IT infrastructure and maintenance as local down the street business - well its days are numbered !

      1. This post has been deleted by its author

      2. K

        Re: Finally an article that realises the reality..

        Definitely agree on the very small company points. I suspect actually that would spur future growth in cloud as well, as the small company grow, why invest in onsite, when they already have a working solution they can expand. But let's face it, the SME's with real spending power that most cloud providers target are not 1-2 employees, they are the 20-200 and up, who would typically have onsite IT expertise and infrastructure.

        But your other points are misplaced, cost is only 1/4 of the driving factor in IT, its ensuring it delivers the necessary service.

        1. Anonymous Coward
          Anonymous Coward

          Re: Finally an article that realises the reality..

          You're right, delivering the necessary service is a huge driving factor. That's where cloud can give major gains. For example, historically if I want to use new hardware for a new project, service, whatever then I had to get agreement for the capital costs; get the purchasing team to order the equipment;get the infrastructure team to configure and commission the kit, and as often as not I'd find that we were short of space in racks for the new hardware or our our SANs were hitting limits. Now on cloud, I just specify the kit I want and within minutes it's there for me. That is a major boost in productivity, over and above the cost advantages. And that's in a company of over 200 people with a sizable IT operation.

          And as has been mentioned by others previously, the zero capital outlay, ability to scale up and down on demand and simplified DR experience - have a hot or cold standby in a different datacenter which exactly matches my live one - all add increasing value again.

          Cloud isn't going away. People may fear it's going to put them out of a job - and for some it will. But for the good ones, there is still plenty of work to be done, it's just moving up the food chain. Relatively low-skilled (but historically fairly well paid) jobs are in jeopardy. Highly skilled jobs will be just as in demand as before - because on-premise isn't going away anytime soon and hybrid network and application architectures are complex but greatly needed.

          By the way, one of the things you're forced to do in moving to cloud is just as beneficial in a non-cloud environment. Automation of EVERYTHING deployment and configuration related. That means source controlling your configuration scripts and settings, builds, setup and start-up tasks etc. There is no reason why you shouldn't do this for on-premise stuff now, but even with the best of intentions that work usually gets put on the back-burner in the face of having to fulfill day-to-day requirements. Being able to automatically, rapidly and repeatedly create and tear down servers makes your testing and deployments much more valid and thorough. There's nothing intrinsically cloud-y about it, but going to the cloud just forces you to do it.

  2. All names Taken
    Paris Hilton

    A natural evolution?

    How many digital images do you have of your family?

    How many scanned images do you have of your family?

    How many years, generations, variations in technologies do these images (and soon to be added in the archival sense?: movies) do you have?

    Are you willing to store them to one machine (laptop, notebook, oompah-loompah broo ha-ha? mobile phone, ... ?) and be prepared to lose them when its storage chemistry and physics breaks down causing a data loss of sorts?)

    That is only trivial (financial sense) example of critical (emotional sense) data that most pholks will have and be in the process of having and building even though they may not be cognatively aware of the data acquistion acquiring and building and ... OK! I'll shut up now :-)

  3. Will Godfrey Silver badge

    Not what I'm seeing.

    SMEs I know of have been backing up and syncing machines for ages. Private individuals are also learning (often the hard way) to sync across different machines. I don't know anyone who has just one Desktop and/or laptop. The messages about internet blackouts, security etc. are very slowing sinking in.


    The problem with cloud...

    At last an article that addresses the biggest, but not only, problem with cloud. It fixes a capacity problem, i.e. expense, that companies don't have. Managing and billing for IO at a tenet level is the hard nut to crack and where there is money to be made.

  5. miketheitguy

    The most interesting thing about this movement is how misinformed people are about what IT actually does in the world of computing. We aren't just hardware monkeys. I'd say I spend well less than 1% of my time actually doing anything with physical hardware. It's almost pretty much set it and forget it at that point.

    99% of my time otherwise is spent in software. Whether it be provisioning systems, hardening systems, dealing with our 'InfoSec folks' over what constitutes 'hardened' systems, and dealing with our devs infecting development VMs with malware.

    Sure, there is a cost associated with that infrastructure, but for my workloads the cost projection for AWS vs just buying locally isn't actually there. A quick ballpark number so far is that the local infrastructure versus forking everything to AWS (and beefing the internet connection to compensate) is that the local infrastructure pays off in about 2 years. After two years, AWS becomes the more expensive platform short of any price changes.

    But you still need actual systems and platform engineers to get anything done. While that person might not have to know how to set up ISCSI volumes, or how to build redundant networking; those concepts are so old and so well known by most IT professionals we can do it instinctively. The more challenging part is balancing infrastructure security needs with developers who can't write for a platform for shit.

  6. whiz

    'Walmart dumped the legacy Oracle ATG e-commerce platform..' for Openstack.

    ATG uses middleware, CRM and eCommerce as enablers to deliver CX.

    So what happened to the ecommerce. The statement sure as hell makes the assumption that somehow Openstack does middleware and ecommerce.

    I think the author has to re-phrase. Its seems Walmart used Openstack to implement a private cloud and thus save on infrastructure. The applications for eCommerce are NOT replaceable by using openstack.

    To imply as such is ridiculous.

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