Three top misconceptions
Three top misconceptions
Outline some common misunderstanding of setting up a cloud deployment, such as misunderstanding the need to architect the infrastructure, dangers of cloud lock-in (Cloud orchestration will solve this), and support for automation. Conclusion: Hybrid cloud works best in most scenario.
There is no question that public cloud providers get a lot of attention these days as the cloud matures and gain greater acceptance within businesses. It’s no wonder though, seeing how cloud computing has contributed to substantially faster rollout of new services and greater efficiency in IT deployments.
Unfortunately, many of the misconceptions from the early days of cloud never did get corrected, resulting in a great deal of misunderstanding about what the cloud can, and cannot do. Understanding this requires taking a step back and viewing the public cloudscape through the various layers of infrastructure-as-a-service (IaaS), PaaS (platform-as-a-service) and SaaS (software-as-a-service).
We look at the three top misconceptions below.
Much of the misconception on this stems from the days where people said “cloud computing” when they really meant IaaS. The widespread use of open source software also gave the cloud an overarching perception of openness and interoperability that does not necessarily reflect the proprietary nature of many cloud platforms today.
While open source software is widely used, their implementation at the various layers do not automatically lend themselves to interoperability. Indeed, cloud services from different cloud providers will not typically work together, and migrating apps built for one cloud to another may entail significant re-architecture and recoding.
Of course, many of the cloud services do expose an extensive array of APIs (Application Programming Interface) to facilitate integration. In practice, a not insubstantial amount of effort and time is required for any integration efforts.
Another common misconception of the cloud is that it is easy to deploy in the cloud with little or no setup necessary. While this is undoubtedly true for most SaaS offerings, it must be remembered that most enterprise cloud deployments are hardly monolithic, standalone affairs, but which often entails weaving both cloud and on-premises deployments together.
And the irony of the cloud’s popularity means that the average organizations is likely to be using more than one cloud, too. Some of these would undoubtedly be PaaS offerings that require various platform specific setup and configuration. Expect substantial work too, to test and migrate existing applications to the cloud on the IaaS front.
Again, while it is true that SaaS offerings such as hosted email or CDN (Content Distribution Network) are essentially self-contained and self-maintaining, this is not true for a large number of cloud offerings. Moreover, even self-contained SaaS offerings do require the occasional – though rare – downtime to fix a newly discovered bug or security flaw.
Moreover, Linux compute instances on Amazon Web Services (AWS) need to be regularly updated by the customer, while CMS and ERP servers hosted on cloud servers do need to be updated and maintained. So don’t go off firing your IT department yet, but leverage some of their freed up capacity towards strategic initiatives to bolster the company’s bottom line.
Despite the proprietary nature of public cloud platforms, a new solution known as cloud orchestration promises to let enterprises use more than one cloud platform in an effective fashion. This is done by utilizing the various APIs to create a single pane of glass view across multiple cloud resources, greatly simplifying the task of management.
In addition, it also facilitates the creation of a new fabric that developers can tap into to enable the development of cloud services at greater cost efficiency than ever, and which could potentially span across clouds. [To find out more, read more about StarHub’s cloud orchestration solution here.]
Why Should CIOs Move Away from Traditional Data Centres?
Break away from the tradition and get a first-person Point-of-View (POV)..
4 key considerations of a data centre strategy
As IT budgets shrink, IT challenges grow – rising capacity requirements, endless demands to ‘do more with less’, and the need to leverage new technologies in order to stay competitive.