Virtualization has many advocates and is the latest IT bandwagon rallying cry. The approaches to virtualization have taken two distinct paths. The supply side path is usually undertaken by IT operations in an attempt to mask hardware differences and minimize complexity; this approach does not go far enough. The other path involves a business demand driven approach. The demand driven approach needs to be taken to ensure that virtualization is executed in a strategic manner.
A recent article titled How to create a business-boosting virtualization plan http://www.networkworld.com/supp/2008/ndc5/081808-ndc-virtualization-strategic-planning.html?page=1 highlights the differences in the two approaches giving examples of each approach and highlighting the limitations of supply side virtualization. The blog titled: The $100 million dollar blade illustrates the dangers of unbridled server growth http://www.networkworld.com/community/node/28963.
When anyone examines IT from a business perspective, the principle question that comes to mind, is: “What is the ROI for IT?.” We must never forget that IT is an investment. IT competes for every dollar it’s allotted, against sales, other logistics, marketing, etc. The current runaway costs for IT in many organizations starts with infrastructure and datacenters because growth is managed from a supply point of view, where datacenter layout is dictated by server type and cooling restrictions.
A business driven virtual datacenter strategy would change any datacenter and infrastructure discussion from a cost centric mindset to a strategic partnership endeavor. In essence, the conversation would revolve around the following tenants:
- What drives revenue, risk and cost?
- Understand what the Quality of Experience (QoE) will be for clients and executives? How to you balance performance, cost, and efficiency?
- What are the key business drivers and pain points that the business experiences?
- What are the most significant business transactions and what is the impact of servicing them properly?
- How to define a measure of key performance indicators and critical success factors that can be used as metrics for goal and investment tracking?
These tenants provide the foundation for a business driven virtual datacenter strategy because a viable virtualization strategy requires rigor as well as an initial investment. That rigor must include business analysis and an engaging architecture practice. Data virtualization is an excellent business driver, because many organizations are still account driven and not customer centric as they know they should be.
The key lesson learned when building a business driven virtualization strategy is that you must focus on how you are going to model and measure demand, so that supply can be dynamically provisioned. The modeling exercise forces you to explore the many types of behavioral application patterns that a virtualization strategy must address. This includes numerical calculators, messaging systems, transaction systems, portals, and transformation engines, to name a few. These behavioral patterns provide operational requirement templates for applications that help stakeholders decide what kinds of resources they will need for a virtualized platform.
The key obstacles to achieving this strategy are not technical, but are individual and organizational behavior patterns, such as control, trust, responsibility, information sharing, proper documentation, and disciplined development practices.
An engaging architecture practice is critical to driving a business driven virtual datacenter because it will act as the bridge between all stakeholders (business, development, and operations), eliciting the characteristics of business demand and translating that into the implications for infrastructure change.
Tony & Shep
