A provider can offer anything from raw ingredients for programmers to build online services and applications from scratch, all the way up to highly industry- and problem-specific applications, along with the management and maintenance required. There’s a place and market for these services. Anything from files to desktops and video conferencing can be run from the cloud. Clouds can even be hidden away from the public, and accessible only on private networks.
Each type of cloud offering will solve a need for a very specific market. It’s therefore crucial to do a little self-assessment before making the decision to virtualize or put a workload or process online, and prior to selecting a provider once you’ve made that decision. This is a very important step because it is related to how involved you are, and want to be, in the design and management of the service.
So many choices
Most public cloud providers are massive firms, offering rock bottom prices for the raw ingredients, namely the infrastructure. The infrastructure is shared by many clients and exclusivity is minimal. Also, at this end of the spectrum, it’s up to you to acquire all other required resources and the staff to build, manage, maintain, scale, secure and fix the service. Routers, switches, the connectivity and so forth. Technology-based firms find these type of cloud providers be a great fit. That’s why so many start-ups and dotcoms make use of public cloud services — they can assemble the raw pieces and manage everything themselves. Everything is done through code API calls. There aren’t custom SLAs or a live agent on the other end of the phone if and when you need help to fix and maintain, much less build, something. This is simply not their business model — labour is very costly. Simple, raw, and publicly accessible tools with a simple set of choices on hardware is what they excel at providing. Privacy, customizability and SLAs aren’t on their radar — that’s a different type of provider and business model.
At the other end of the spectrum, the assembly, maintenance, scaling, installation and management are taken care of by a different type of cloud services provider. Unsurprisingly, these are known as managed cloud providers and private cloud providers. They invest heavily in engineering and business analysis staff in specific verticals, redundant systems and SLAs. The sticker price of these services is of course higher, but at the end of the day, the work is done for you at very efficient rates. A single engineer can handle multiple setups. If you were to hire a team of engineers and support staff to maintain your infrastructure, it would make sense only if you wish to retain control of it in-house and/or have a sizeable infrastructure, or are a technology firm yourself. The infrastructure is made to fit the needs of the individual firms, with some providers even providing a high degree of privacy and low multi-tenancy through separate sets of hardware for each customer.[inline_divider type=”1″]
Letting the numbers finally decide
A costing spreadsheet helps make the choice clear. Unless you’re a technology firm like a big dotcom or social network whose day to day operations are focused on operating your cloud-based services, the TCO of running a medium sized infrastructure in-house or managing it internally by hiring extra engineers is typically much higher than the price of letting a managed cloud provider do it for you.
The choice of provider really depends on who you are, where your firm focuses its energy and resources, and the type of workloads being run. The storage of private and highly sensitive data on public clouds is a sensitive question. In fairness to the big providers, it’s not easy to maintain tens of thousands of machines with what amounts to a large automation to staff ratio — they’ve invested heavily in this type of delivery model. It’s also why there’s such a gradient in pricing — there’s a corresponding gradient in terms of number of staff and business engineers who adapt and apply the technology platform to your needs. Ask questions about how big the firm is and what the engineer to client ratios are like: it’ll help to normalize the question of pricing. It comes down to opportunity cost and what you’re currently doing, and the cost of where you want to go if you do it alone or get someone to help, and how much help you want.[inline_divider type=”1″] [column width=”1/3″ title=”undefined” title_type=”undefined” animation=”none” implicit=”true”] [linkarea background_color=”” hover_color=”accent1″ href=”/part-1-cloud-what-is-it-good-for/” target=”_self” icon=”” icon_color=”accent6″ icon_size=”62″ image=””] Part 1: Cloud, what is it good for?
First part in the series make the concept of cloud crystal clear.[/linkarea] [/column] [column width=”1/3″ title=”undefined” title_type=”undefined” animation=”none” implicit=”true”] [linkarea background_color=”” hover_color=”accent1″ href=”/part-2-isnt-everything-online-part-of-the-cloud/” target=”_self” icon=”” icon_color=”accent6″ icon_size=”62″ image=””] Part 2: Isn’t everything online part of the cloud?
This series deals with the concept of virtualization and the potential this idea has on business.[/linkarea] [/column] [column width=”1/3″ last=”true” title=”undefined” title_type=”undefined” animation=”none” implicit=”true”] [linkarea background_color=”” hover_color=”accent1″ href=”/part-4-the-money-question/” target=”_self” icon=”” icon_color=”accent6″ icon_size=”62″ image=””] Part 4: The money questions
This series ties up the final matter of price and ties the series together into a whole.[/linkarea] [/column]