Take telephone communications for example. 10+ years ago, we all relied on land lines for our primary communications. But, the improvement in mobile networks has changed this. Many people don't even have a land line phone in their home any more. And now, with the improvement in VoIP networks, much of the voice traffic goes over the internet. Like many IT professionals, I use VoIP software on my laptop to talk with colleagues, friends, and family while I'm travelling around the world.
But, if I have a really important business call while I'm travelling, I still use a phone because VoIP through most hotels' internet is just not reliable enough for closing an important deal. Every winter, at my home in New Hampshire, the combination of snow, ice and wind causes some electrical power outages. Usually, the power is out for a few hours or less, but this year, a particularly bad ice storm took out power for five days. No lights, no heat, no internet...but my land line phone still worked. Don't ask me how, but the phone always seems to work. In fact, we've come to expect this level of reliability, availability, and performance from our phone system and other business critical infrastructure. When evaluating new options for business capabilities, we need to understand these requirements and expectations.
Business drivers for cost reduction and realignment toward outsourcing are having a profound impact on how and where business infrastructure and applications are provided. Hosted services and SaaS have reduced costs and replaced or enhanced capabilities across the enterprise. But SaaS has done much more than lower costs. It has made it possible for medium and small businesses to have powerful communications and applications. Today, even my two person company can afford teleconferencing and application sharing functions, capabilities that were available to only large corporations ten years ago.
At the same time, Web 2.0, Enterprise 2.0, Wikipedia, Facebook, Twitter, and so on, have raised expectations, pressuring companies to provide better user experiences. The combined business drivers for reduced costs and better collaboration and user experiences are driving enterprises toward SaaS solutions across the board. But, like the trusty land line phone, we still expect, and business requires, a minimum level of quality and reliability. In addressing these business drivers, we need to understand business operational requirements and user expectations. Savings from bringing in a low cost VoIP phone system won't be appreciated if the voice quality is poor or if availability is limited. Adding new collaboration and social networking features might be cool, but they are no excuse for degrading application performance or introducing compliance or security issues.
Don't get me wrong. I'm not saying that there aren't good applications available in the cloud, because there are. But, using applications in the cloud introduces added complexity. To mitigate that requires a good implementation architecture that supports enterprise requirements and integrates with enterprise application architecture styles. Many SaaS providers don't really understand these issues. So, it's up to you, as the architect, to understand these challenges and evaluate whether a particular solution is up to the task or not.
Yes, the internet changes everything. But, the more things change, the more they remain the same. Without reliability, availability, performance, and security, new capabilities can't be used for business critical activities. So, make sure you understand how a SaaS solution fits within the complete, end-to-end application requirements, and have accounted for reasonable user expectations when evaluating cloud solutions.