Tuesday, June 7, 2011

Fear, Uncertainty, and Doubt in the Cloud

Cloud-based computing has ushered in a new period of Fear, Uncertainty and Doubt - FUD - in the halls of corporate IT (Big IT).

Fear that our company data is not secure in the cloud. Despite the undeniable fact that my PayPal, Amazon and bank accounts are safe.

Uncertainty that the cloud can possibly be more reliable in delivering up-time than in-house, on-premise, systems can - yet you have to admit that the Internet has proven to be more resilient and available than the best run in-house systems in your company.

Doubt that the disaster recovery that is available on the web and the cloud could possibly be better and faster than that of an in-house hosted system. Though an automated cloud-based failover can kick in within minutes of a triggering failure, while an in-house failover system requires a huge amount of expensive setup and testing time with experts that may not be available or even on the company payroll when the failure occurs.

Let’s be honest, it sounds like absolute common sense to have your data and systems on your system. That way, you KNOW that you have the data and the system - no matter what happens. But let’s drill down a little further into this seemingly logical and self-evident statement.

In the early days of personal computers (1980s) the volume of data generated and the hard disks available were pretty small. A 10-megabyte hard disk was a big deal. A lot of data were stored on 5.25” floppy disks that held no more than 1 Mb of data. Beginning in this era, IT departments evolved sophisticated backup systems. Twenty five years later Big IT is still thinking the same way: Get more and more storage and servers to meet the needs of data recovery and backup. In fact, Big IT just keeps getting bigger. Data volumes have skyrocketed and will continue to do so. Continuing along this path of in-house management of data will continue to cause a growth in Big IT and its expensive assets. So, what is the alternative?

Around 2006, I started using Gmail from Google to support my communication needs. Overnight, my email communications and attachments were no longer stored on my system or the company’s corporate systems in MS Outlook, MS Exchange or Lotus Notes. Yet, every single email that I have saved since then is available to me. Google’s vaunted search engine is my email search tool! I have upgraded my personal laptop, which is also my desktop, each time that Microsoft launched a new operating system - five or six times since then - and prior to using Gmail, I would’ve had to consolidate my MS Outlook appointments, contacts and emails on to each new machine before I could stop lugging two laptops on my road trips. Now, every single email and contact that I have saved since 2006 is available to me, from any workstation - not just on my laptop. And, to make this even more convenient, all of this data is also available to me on my phone device.

By simply moving my data to the cloud (inadvertently, I’ll admit), I have made it work for me far more effectively and hassle-free than having it stored on the ‘company’ system. So, what was the cost for all of this great convenience? Zip. Nada. Zilch. Google provides this service at no cost - other than sliding a few relevant advertisers’ links past me while I review my email. (On a side note, I have lost count of the number of times that I actually did learn quite a bit from these links - so it really is a win-win scenario.) At a nominal cost, Google would suppress these links.

This is a simple example of the allure of the cloud versus in-house Big IT. Inexpensive versus expensive. The cloud is awash in reputable companies (such as Microsoft, Oracle, Amazon, IBM, Google) that provide low-cost storage and even virtual machines - all accessible via the web. This is a compelling business model for companies that are trying to reduce costs and be more efficient with their data. Sure, it means less work, which means that Big IT’s empire shrinks a bit - but it costs a lot to run that empire. So a savings would be a good thing.

So what are the potential downsides of living in the cloud? A glitch in the Internet could cause problems. The Internet is a huge entity and it seems to be running well enough for many businesses to have already made the leap into the cloud. Recently, Amazon’s EC2 system hiccupped and many businesses were seriously impacted, some for as long as a day or so. But this could have been avoided had these businesses designed their systems as recommended by Amazon - mirrored their systems on two different domains - so that one would have acted as a backup to the other. Disaster recovery requires a failover system. In a cloud context, this could be as simple as having your primary host and your failover host be provided by different cloud vendors.

Thus far this discussion has focused on data. How about applications delivery?

Until 2006 or so, applications were primarily written to run as thin or thick clients on people’s desktops or laptops. Big IT installed applications on their servers and applied update patches from the vendor as required. Again, this requires more resources and helps Big IT to stay big. More applications and solutions require more servers and hardware and more people to manage them.

The cloud provides a significant new opportunity. Let the vendor bear the cost of managing the updates. This means less cost to the company while obtaining a higher quality of focused attention - vendors know their product and its behavior much better than Big IT could ever hope to understand. Consistent, contract-based, change management procedures ensure that the vendor complies with reasonable and logical processes that, in turn, ensure that the company gets a consistent and reliable product from the vendor. Another huge advantage is cost. The web-based access model for applications is typically based on the number of users and the applications that they use. Per-user costs are in the range of $100 to $1,000 per user per month depending on the application and the volume of data. Compare this with the millions of dollars required to even contemplate the deployment of an ERP system onto an in-house system.

To me, it seems obvious that getting your head into the cloud is a smart and savvy thing to do. The sooner, the better. All that you have to lose are your overhead costs.

Note: All trademarks and trade names used in this article are the property of their respective owners.

No comments:

Post a Comment