Jessica Meats, or Geeky Girl, claimed 99.999% availability for SQL 2005 today on a  Technet blog post, “Mythbusting: SQL Server isn’t reliable.” She cites a customer quote from a Microsoft Fujifilm case study, and adds a 99.99% uptime claim from a Barbox case study.

What’s behind this?   Take a closer look and you’ll see that Geeky Girl is a Microsoft employee who refers to a 5 year old case study on Microsoft’s web site.   Using one of the oldest marketing tricks in the IT world, she misinterprets a customer quote to make the ridiculous claim that SQL (2005 no less) is 99.999% available.  No proof, no data – and what does Microsoft say?

“Customers can have confidence in SQL Server on ftServer for local provisioning of database servers with exceptional uptime, rapid asynchronous data mirroring for disaster recovery, rolling upgrade style patch management with minimal operational impact, and for any database application where underlying hardware with near-six nines availability is the preferred choice.”

- Claude Lorenson, Director SQL Server Marketing, Microsoft Corp

Myth busted?  Yes, with SQL Server running on ftServer.

Like every good son, I called to wish my mom a happy Mother’s Day. Disappointed at not hearing the big ‘thank you’ for the flowers, I asked how she liked the arrangement. The florist never delivered the order. On Mother’s Day. No flowers. Why? “Our computer has been down for two days,” the proprietor told me when I called.

Mother’s Day is the florist industry’s Christmas season. But like many businesses, from the smallest to the largest, the excuses for not anticipating the havoc resulting from an outage are familiar: my business can ride through an occasional outage; investing in uptime assurance is too expensive; what I have is good enough. What they’re really saying whether they realize it or not is that their customers are not their first priority.

This florist has no idea what this misadventure has cost in lost business this past Mother’s Day and for many more to come. I, for one, will never go back

Having just returned from the eClinicalWorks Northeast Users Conference – a really helpful and informative conference of 1200 enthusiastic users – the IT challenges that face small to mid-sized EHR practitioners were never so apparent. Unlike larger facilities and hospitals with IT staffs and expertise, the small practitioners are jumping on the EHR bandwagon without a parachute – in otherwords moving to an EHR system which, no matter how great the software is, is going to fail.

Why is that?

A standard computer is not designed for high availability and is going to fail. That’s annoying if it’s your family pictures on the disk drive that crashed, and having your data backed up is critical. It’s different however for EHR – your system is down, the records are not available at the point of patient care, back up data may not be current and introduces the very risks that EHR is supposed to avoid like prescription error, false diagnosis, never mind lost appointments, lost productivity and costs.

Larger facilities who understand IT design their systems with high availability to reduce or eliminate the risk of system downtime in the first place. Stratus helps them do so and, as we discussed last week, we can help small and mid size practices also implement uptime assurance for their EHR systems to avoid the risk of system downtime to their patients and their practices.

It’s like health care really – yes, go to the doctor when you are sick and take your medicine (back up and recovery), but focus on preventive medicine (Stratus and uptime assurance) and not be sick in the first place.

I had the chance to speak at the Microsoft Connected Health Conference in Chicago this week, where Stratus was among a number of Microsoft partners sponsoring this event. This was a well received conference with good insights on technology and services. Talking with practitioners and other partners also laid bare some limitations with Microsoft products – and why Microsoft embraces their healthcare partners so readily. A few important gaps in healthcare IT are solved with Stratus capabilities in line with key Microsoft products:

Microsoft SQL Server: It is the foundation of just about every EHR system. And few if any practice under 25 doctors or 100 users need Enterprise edition. However the Standard edition does not support high availability. So many practices that ask for high availability see their costs skyrocket because of this, along with the other costs involved with Windows Clustering. And yet many of these practices don’t have the IT expertise to install and manage Windows Clustering in the first place. The result? Most small to mid size EHR installations have NO high availability and are extremely vulnerable to these systems going down.

Answer? Stratus Avance and Microsoft SQL – provides high availability for any version of SQL Server, ease of implementation and operation so little IT knowledge is needed, and far lower costs to make this approach much more affordable to small and mid size practices.

Microsoft BizTalk: Is widely deployed for information exchange requirements so think HL7, EDI, CPOE, etc – critical healthcare applications which need to be available. Downtime for these systems can have serious consequenses. BizTalk is widely deployed for these kind of critical data interchange needs yet isn’t really supported by Windows Clustering – talking with several consulting partners who specialize in this area it’s a critical application need that doesn’t have a viable Microsoft high availability solution.

Answer? Stratus Avance or ftServer and Microsoft BizTalk – viewed as a single system, either of these Stratus solutions provide higher availability than Microsoft Clustering could hope to achieve, and give a powerful, yet cost effective solution for BizTalk high availability.

Microsoft has it right with it’s partner ecosystem.

“Good enough” service is only acceptable until something bad happens, like the Amazon downtime debacle. Then everyone demands compensation for business interruption and lost revenue. A couple of industry experts in a Networkworld article went so far as to say that customers actually shared in the blame for their losses, stating they should have anticipated failure and made plans to minimize impact. Silly customers. They thought they were doing that when they contracted for cloud services.

Cloud service providers are not experts in uptime assurance. They know it, which is why their availability SLAs are empty promises. Uptime assurance is sophisticated technology. Running virtualization software on armies of servers doesn’t magically create it. Purpose-built software and hardware, proactive availability monitoring and management, and best-practices oversight does. These solutions – industry-standard with compelling ROI — are available today. But until cloud customers demand SLAs with real teeth, and service providers own up to their responsibility to protect their customers’ interests, good enough will remain the unacceptable standard in the cloud.