We all have our own personal work-related pet peeves. I've seen mild-mannered accountants curse up an alliterative storm directed as those pesky Windows updates that automatically restart the affected computer after the update is completed -- without first asking, and invariably while the owner is away from their desks, with important unsaved documents and emails. And who hasn't ever had to contend with a loud office colleague who talks at full volume (often about non-work related topics) at all hours of the day, annoying everyone within a 7 cubicle radius? My personal pet peeve however is what I refer to as "blog-spam". That is, blatant advertisements that masquerade as innocent blog posts. These poorly-disguised advertisements (usually for some marketing consultancy or systems integrator) contain all the appropriate buzzwords, but fail to deliver any real message or value; their main objective is simply to drive traffic to the blog owner's web site.
We've all come across these types of blog posts that, at first glance, seem quite promising. They contain the requisite tags and buzzwords: Facebook, Twitter, Social Media, and of course, Cloud Computing. But in the end, that's often all these pseudo-blogs have going for them — buzzwords. Let's take "cloud computing" for example. Could computing is hot right now. It's sexy-smoking hot. You could throw cloud computing into the title of almost any random blog and have a hit. A nondescript title like "Knitting patterns for socks (without flaps or gussets)" that might usually only get a few hits suddenly becomes a block buster when you slap on cloud computing: "Knitting in the Clouds: Extreme High-Altitude Sock Patterns". Presto, best seller.
Never mind that many people still don't seem have a firm grasp on what cloud computing actually is (more on that topic in a minute). People are practically willing to embrace cloud computing as a cure all for any corporate malady. Does your company suffer from poor customer engagement? Is senior management apathetic to customer concerns and focused solely on short-term profits? Do your products and services fundamentally suck? No worries, just apply our magical cloud-computing ointment topically, twice a day for three days, and you're good to go. Need to lose those extra holiday pounds? Want to re-grow that receding hairline? Can't get your teenager to pick up their room or take out the trash? Problem solved: cloud computing to the rescue. Voilà !
But let's stop joking and start getting serious. There's been a lot of talk in the blogosphere lately about cloud computing. At times it seems like anyone trying to sell anything feels compelled to adopt the cloud-computing mantra, whether we are talking about CRM software, Marketing Management consultancies, Search-Engine Optimization (SEO) specialists, or whatever. If it doesn't take place in the cloud, it's not cool.
Some of the cloud-computing conversation is driven by software vendors who peddle on-demand (i.e., hosted) versions of their software platforms via the software as a service (SaaS) delivery model; these vendors rightfully recognize cloud computing as an opportunity to lower their costs while positioning themselves on technology's cutting edge. And then of course, the conversation is further fueled by on-premises software vendors who claim, in response, that cloud computing is primarily a just marketing gimmick, which attempts to re-brand already existing technologies such as network storage, virtualization, and load-balanced on-demand web hosting. And there's the analysts and consultants who add to the conversation with their own research findings and insights. And then there's the bloggers, discussing and debating the opinions of everyone else. And so the conversation grows louder.
And as more voices join the conversation, it becomes harder to follow any one conversation. Side conversations develop. Topics shift and go off in new (sometimes strange) directions. Soon, it's almost as if no two people are even talking about the same thing anymore. I think that's actually where we are now with cloud computing. So let's press the reset button and start the conversation over from the beginning. What is cloud computing? Please feel free to disagree, but in my opinion cloud computing is essentially just the delivery of IT services over the Internet using elastic (often offsite, third-party) computing resources. The two main points here are: 1) the company using the software doesn't have to own or maintain the computers on which the software is run, and 2) the usage is elastic, meaning that the number of computer servers in use automatically scales up during peak usage and usage spikes while instantly scaling back down when usage declines.
So how is cloud computing any different than a regular SaaS delivery model? Both delivery models satisfy the first point above, in that the end-user customer doesn't need to work about installing, owning, or maintaining the IT infrastructure. And while its neither a direct result of SaaS or exclusive to SaaS, most SaaS provides charge their customers using a so-called subscription-based pricing model where you pay a monthly service fee, as opposed to a large, one-time capital outlay commonly charged by traditional on-premises software vendors (though to be fair, many on-premises vendors also offer monthly subscription-based pricing models for their on-premises software, so this is not a benefit unique to the SaaS delivery model.)
The main difference between SaaS and cloud computing then is the second point above: usage elasticity. If a traditional SaaS vendor hosts their software on their own servers, they are constrained to some extent as to the capacity they can offer their clients. Sure, they can add more servers as necessary to scale up as demand from their clients grow. And yes, they can install a bunch of extra redundant servers that are just sitting around doing nothing, in order to be ready for any sudden spikes in demand. But those extra servers cost money and take up space; they gobble up electricity and produce heat. It cuts into profit margins to install, run, and maintain a bunch of servers that aren't being used to capacity.
That's the main selling point of cloud computing. As a software vendor, instead of hosting the software you provide on your own servers, you outsource the hosting of the software you provide to some gigantic Internet behemoth (i.e., Amazon, Google, IBM, etc.,) who has a couple of hundred free acres of unused servers sitting in a server farm somewhere. Then, as a software vendor, you focus on writing your code and cashing your monthly subscription checks from customers, while the cloud providers worries about the details of load balancing, CPU cycles, data security, etc.
So clearly cloud computing provides SaaS vendors with obvious benefits -- namely it offloads the hassle of managing IT infrastructure and customer data, while also providing costs savings (as Google or Amazon can host the IT infrastructure more cheaply due to economies of scale). But what about the companies who license SaaS technology? Do they benefit from cloud computing as well? The answer is simple: yes, no, and well… it depends
Of course, the very mention of the words “cloud computing” will instantly win over even the most rebellious business users who have grown accustomed to think of their company's own IT department as fundamentally opposed to anything cool or pro-business. And, sure, customers might also benefit from lower subscription cost due to money the SaaS vendor saves by offloading the hosting of the IT infrastructure -- but that's assuming the SaaS vendor actually passes any of those savings on to their customers. And yes, usage elasticity in theory means less chance of servers getting bogged down and less chance that end users will experience performance problems.
But in practice, cloud vendors face the same architectural problems of trying to gracefully scale applications that any in-house IT department faces. Moving IT to the "clouds" doesn't eliminate servers and databases -- it just moves them to someone else's backyard. Which brings up another point that makes some cloud-computing skeptics nervous; who the heck knows were your valuable customer data actually is and who is guarding it (or if they are guarding it). Just like the old rule of thumb that you shouldn't write anything in an email that you wouldn't want to see on the front page of the Wall Street Journal, some cloud skeptics would probably also advise not to send any data off into the clouds that you wouldn't want to end up on the desk of your main competitor.
This post is also availalbe at TheSocialCustomer.com.
Comments