Share what you know with millions of people
Focus is the best place to turn what you know into remarkable content
0
Ahead of the Curve: Is it too soon to consider post-cloud architectures?
Outsourced to home-grown and back, centralized to decentralized and back. Like most things, IT goes in clearly discernable cycles. As cloud adoption increases, on-premise options will become simpler, more powerful, and less expensive and the cycle will likely continue.
Is it too soon for IT leaders planning investments 5 or more years out to think about what comes after the clouds pass?
Events
- Dos and Don'ts of Small Business Marketing May 29 @ 11 am PT
- Lead Nurturing 202: The Next Generation May 31 @ 11 am PT
- The Tricks to Paid Media June 6 @ 11 am PT
- Display Advertising for Brand Awareness June 20 @ 11 am PT
Related Activity
3 Answers
Hi John,
For those who have the privilege/burden of planning 5+ years out, it is not too soon for considering what a post-cloud world will look like. For the sake of this discussion, I'm considering "post-cloud" to mean "the next big thing that is focused on after the cloud becomes largely ubiquitous" rather than "what we do once everyone gets rid of their private clouds and reduce dependence on the public cloud".
Just as there are all sorts of ways to use the web today, and different organizations use it to different degrees and with different levels of openness, so I expect cloud usage to look different for different types of organizations -- but I expect it to be largely ubiquitous in 2-4 years.
Increased cloud will not make on-premise options obsolete (on a macro level) for everyone. Some organizations will put lots into the cloud, some will put very little, and this will be based on the type of organization, the mindset of its leaders, the challenges of the industry, and other factors that will prevent neat characterization of who will and who won't.
Yes, I think that on-premise vendors will improve, and I think that some vendors will continue to offer both types of solutions, but there are significant advantages of the cloud models for vendors as well (ongoing payments), which will continue to spur changes in that direction.
Some classes of solution will continue to be really good and desirable in on-premise configurations, and many larger enterprises will continue to have a large mix of both solutions, since they have enough infrastructure and impetus to support both models.
But, I don't think the cloud is going away any more than the web is going away.
Even outsourcing hasn't gone away -- for the most part, it changes its destinations and its scope every now and then, but we're still doing more of it today than 5 or 10 years ago... Same will be true for cloud.
-ASB: http://XeeMe.com/AndrewBaker
For small companies that do not have in-house IT it is not an issue. They will basically hand over the Architecture to the vendor. They care but their business model is more important. The real rub is with large organizations and their extensive IT staffs. They have to understand SOA as a pre-req to Cloud Services. If they do not pay attention to the Architecture they will lose their IT edge.
John,
It's a very interesting point and trend you have identified. The pendulum's direction can switch with minimal notification in the information technology industry. In each of the trends you mentioned, it seemed that the switch was due to hitting a technological "wall". Centralized architectures were easier to maintain, but the bandwidth wasn't available to support the growing demand from PC deployments, so LAN servers were created to manage the load. When it became too expensive to manage all the deployed LAN servers, the pendulum again shifted toward centralization, and this time, not only was there enough bandwidth, but also Internet-enabled devices simplifying the deployment of printers, copiers, servers, etc.
So, when considering the post-cloud era, you must consider what is the wall that will change the pendulum's direction. Security complexity? Insufficient bandwidth? Migration failures? Cost is going to be a significant driver as it was in the two trends you mentioned. Also, control and ownership issues help spur these changes. Any attack on data ownership or lack of availability for applications and data due to cloud computing and the hoards of antibodies will come out against it. Even a private cloud effort can fail miserably leading naysayers to once again question why the enterprise needs self-service or on-demand interfaces internal to the enterprise.
JPM
http://about.me/jpmorgenthal
Answer This Question