Editorials

SaaS – Not Working Out? Reader Feedback

Featured Article(s)
MySQL Basic SQL Part VI — Hierarchical Queries
It doesn’t matter if you call them hierarchical structures, tree structures, or self-referencing tables, the tables we will be dealing with in this article often pose quite the SQL challenge to traverse. Come JOIN along.

Do You Have Your Reservation Yet? Have a question?
SQL Server, Business Intelligence, .NET Development, SharePoint.
All the biggest name, most expert presenters. All to your desktop. Stop paying thousands for admission plus travel plus rooms, food and other things. This is a time to keep learning and stay on top of the technologies, stay focused and at the same time not have to worry about your budget for once. $110 gets you in the door (early bird registration). Get registered today and we’ll see you there. You can use Twitter to ask questions ahead of time – just use the hash tag "#SSWUGVC" and we’ll get the answers you need and address them in the keynotes or in direct email.

SaaS – Not Working Out? Reader Feedback
Dan wrote in with some thoughts on the Software as a Service discussion that has been in the editorials for a few days now. His thoughts:

"The Cloud is the equivalent of outsourcing in manufacturing. Outsourcing makes sense for those mundane business tasks that are not a core part of the business. For instance, ADP has been providing payroll services for businesses forever.

The downside of The Cloud is the loss of control. In other words, can your business tolerate the business practices of the provider?

One simple loss of control example is a small customer that chose to use a hosted Exchange email service instead of managing their own mail services. The servers at this nationally known provider went down. The customer could literally not receive or send email for over a week as the provider tried to get services back. In the end, the provider sent out an apology letter for the “inconvenience” and simultaneously gave kudos to their technicians for their dedication to solving the problem.

Now I don’t know about you, but I can setup an Exchange server easily in less than a day. Secondly, if the technicians were doing their jobs, the issue should have never occurred in the first place.


Every single major Cloud provider has had some sort of upgrade issue that has jeopardized customers. Salesforce.com had upgrade issues that caused pain to the customers. I’m not sure this model is working as well as you think.
"

This is probably the exact opposite experience of what I was talking about yesterday. Yesterday I was mentioning that the transparency and responsiveness to issues with Microsoft and Google offerings may be pointing to one side-effect of using software services and the social networking tools commonplace today. The transparency can wield heavy pressure on the vendor to respond. I do think that’s valid, but Dan’s point of non-responsive or unprepared service providers is a huge point.

I would say that the Exchange provider was unprepared, insufficiently experienced and, well, not qualified to provide the level of services required if you’re going to play in this space. I’m sure there were circumstances, I’m sure there were complications, but if you’re going to be the service provider in this type of solution, you better have full redundancy, full fail-over and complete recovery processes, period. If it’s down for longer than "X" amount of time (in your SLA, right?!?), then cut-over to standby boxes and get things back online.

To not have that type of plan in place isn’t a fault of software as a service, but that service provider.

This shows the caution and due diligence you’ll need to be doing, along with understanding the risks, as you consider these types of services. It won’t be easy, either, to determine just how qualified a given provider is. There is much to the management of these systems and determining whether a provider is "ready" is tough.

Start with the standard questions – what is your backup and recovery model, plan and approach? What time frames force hardware swap decisions? Do you have written recovery and triage processes? Do you have standby hardware and what is your recovery time? When did you last test a full recovery? This is just a start, if you get so much as a hesitation on answering these, go elsewhere.

Featured White Paper(s)
Protecting SharePoint for Windows 5.0
As the distributed workforce grows, the need for online collaboration and document management grows with it. MSFT SharePoint®… (read more)