User-centric computing. With the immanent launch of Windows 7 (the successor to Windows Vista), it's no surprise that the focus of Day 2 of the 2009 Microsoft Management Summit (MMS) is on Microsoft's client operating system. Actually, according to System Center General Manager Brad Anderson, starting at last year's MMS, there has been a shift in perspective and Microsoft now thinks about the client side of its business in terms of the user rather than the technology.

In a video interview, which you can watch HERE, Anderson discusses "the tangible progress" Microsoft has made "on the long-term vision of user-centric client computing." Because people are no longer tied to their desktops and prefer to work on whatever device is most convenient at the moment, the user-centric perspective makes sense. As Anderson says, the ability to "enable users to be productive across all devices starts with the operating system." And Windows 7 has several new features that address this new perspective. Anderson discusses these new features in the interview.

In the current economic climate, establishing the value proposition of Microsoft's upcoming releases is crucial. Microsoft's consistent management strategy is a key component of the value proposition. I'd be interested to hear your thoughts on the importantce of this unified management approach and the value of Windows 7. Send me an email at karen.forster@advaiya.com.

You don’t have to go very far to find more news about the economy and how businesses are looking to cut costs and reduce their exposure to risk. In many cases this means that companies are trying to reduce or eliminate spending and make do with the hardware and software that they already have. Indeed this is one of the key value propositions of SOA, but I will discuss that more in a future post. For now I want to focus on the value that can be gained from strategically upgrading existing applications.

Many organizations are looking at virtualization and server consolidation as ways of saving money. Indeed in recent years the trend toward virtualization has led to many changes in software and licensing. Applications have been architected to work in environments where they are not the only application running on the server. In many cases licensing terms have been clarified or changed to require licensing at the physical or virtual processor level. More sophisticated programs can be restricted to ensure they run on certain processors to enforce the licensing restrictions. By taking advantage of these changes an organization can consolidate applications onto fewer servers and still maintain the same levels of service. In a previous post on this blog Vikram Jain wrote about the results of research done by Advaiya that points to a cost of $6000-$6500 for each server in a datacenter. Reducing the number of servers could have an immediate impact on the bottom line for an organization. If the upgrade provides features that are currently provided by third party products, you could save additional licensing fees. Couple the advantages of upgrading software with the typically small training costs for software upgrades and you have the beginning of a really good business case for managing your software portfolio to provide better value.

In a recent study commissioned by Microsoft and conducted by Forrester Consulting found many benefits to support the case for upgrading to SQL Server 2008. The conclusion of the study was that for a particular customer the risk-adjusted return on investment for the upgrade was 162 percent. Of course your experiences will be different but it makes sense to regularly analyze your computing platform to determine the optimal applications and the benefits for optimizing your platform.

Take this true/false quiz to see if your company is making the same mistakes as the large, multi-national enterprises that have participated in Advaiya’s Business Value Assessment (BVA) engagements. Then view the video here to get the answers, as Karen Forster (Advaiya’s director of Platform Vision) and Eric Zinn (Advaiya enterprise solution strategist) discuss specific examples of common mistakes.

1.       Microsoft does a good job of communicating to its customers the value of Enterprise Agreements (EAs). True/False?

2.       It is not unusual for companies to have 18 or more directories. True/False?

3.       Tracking IT costs throughout an enterprise is nearly impossible because of redundant applications. True/False?

4.       Once applications are deployed by IT, the return on investment (ROI) begins immediately. True/False?

5.       By integrating new applications with existing systems, IT can motivate users to adopt deployed technology. True/False?

Check out the blog for more information: http://windowsitpro.com/article/articleid/101569/3-costly-it-mistakes-that-undermine-business-value.html.

What is your experience with these issues? Please use the comment function for this blog, or send me an email with your thoughts: karen.forster@advaiya.com.

I've worked in Big IT for 25 years. During that time I've worked with three major virtualization technology vendors: IBM, VMware, and Microsoft. I've learned the hard way about the implications of choosing one virtualization solution over another.

I'm forever grateful to IBM as their VM operating system was my ticket out of electrical engineering. I was working at Texas Instruments (TI), and  I was a lousy "double-e." But I'd stumbled across a cubicle near mine that had a big IBM 3178 terminal and keyboard attached to a VM/SP system. It looked so different from the cheesy little discolored plastic minicomputer terminals that encircled it.  It was industrial-strength looking, yet swoopy, with a keyboard so heavy it put a dent in your lap. Now this was computing. I started educating myself and eventually stepped over the line from knowledgeable VM user, to user consultant, and then to systems administrator. I eventually maintained all of TI's production VM systems that ran a number of company's factories.

I moved from the mainframe over to the client-server side of IT early on - back when it was still officially known as "the Client Server Project" - and left virtualization behind for a while. After all, Microsoft had Windows NT 3.1 in beta, their first real* operating system; it was too early in x86 architecture and OS development to think about virtualizing NT's resources . Before too long though, a company named VMware appeared on the market with its  Workstation product. It was great for testing and lab work, a real must-have, but no one would have ever considered using it in full production.

During my tenure at Intel, we evaluated and then chose a VMware-only virtualization strategy. At the time when we had to make the decision, VMware had the more-complete product offering to meet our requirements.  However, that decision left us with two issues to deal with: First, VMware was expensive, and sticking with the straight platform solution caused our virtualization deployment to be severely limited due to budget issues.

I was on the receiving end of this decision's negative implications: Some badly needed Active Directory recovery updates were put on indefinite hold because we didn't have the funding to pay for the VMware licenses. At the time, I resented not being able to implement a free solution (Microsoft Virtual Server)  that did a perfectly adequate job of fulfilling my tactical requirements.

The second issue with VMware was related to virtualization management. Intel uses Microsoft's System Center suite to monitor its production systems, and it was no problem to add VMs to this solution. But the virtualization platform itself - ESX - required a separate management solution, which of course wasn't free. Naturally, the expense of managing the environment was factored into the overall virtualization platform decision. But the soft costs of heterogeneous management weren't.  All in all, implementing a virtualization solution was quite expensive; it was only the fact that virtualization is such a huge cost-saver that made it workable.

Today, how much weight do you think you should given to unified management when you're considering virtualization solutions? How important is it? Are the disadvantages of supporting multiple vendors minor compared to the capabilities that having this heterogeneous environment provides? I'd like to know your opinion.

- Sean Deuby

* True 32-bit architecture, pre-emptive scheduling, with full virtual memory support