I've heard that companies have SAN islands, even within a single data center/campus. Why is that? What is the challenge with consolidating all servers and storage systems into one big SAN footprint?
Companies use SAN islands for many reasons. It may be as simple as different political environments within the same company where one organization does not want anyone else "touching their stuff." It may be that they are using separate SANs for mainframe and open systems or they want to segregate out the Windows environment from the Unix environment.
Many companies have different management groups for different OS environments or application groups and in many instances, the Unix admin and the Windows admin staff are under different constraints and may be managed by different people. Then comes the network guys who are sometimes off in still another group with different responsibilities and management.
These groups can sometimes be working off separate budgets and may want to make their own buying decisions when it comes to what to buy for data storage. I have seen this much too often in my travels. What you end up with is no clear policies, standards or guidelines in place across the entire organization. Everyone is off doing their own thing creating their own little "fiefdoms."
The way to get around this mess is for upper management to get involved and create clear directives from the CIO's office that drive standards throughout the company. The creation of a "Storage Management" group is essential in larger companies if they want to reap the benefits of the combined buying power from all the segments of the business. Choosing best of breed technology and driving standards across all business functions also reduces complexity, increases interoperability and lowers the learning curve for new employees entering the organization. The end result is increased business flexibility and the ability to bring new application to market in a more cost effective and timely fashion.
What you don't want to do though, is force the system admins into a solution that simply does not fit the application environment. Your company should be able to provide storage resources to new servers based on the needs (an budget constraints) of the application in question. This means you WANT to have different classes of storage available to choose from with each class having different properties in the way of cost per megabyte, performance and reliability. This way you can provide a "tiered" storage utility to end users in your organization, based on the service level agreement needs of the applications being supported.
CLICK for part 2.
Dig Deeper on Server-based storage
Related Q&A from Christopher Poelker
SAN expert Chris Poelker discusses how to change the size of a LUN in a Microsoft cluster server environment. Continue Reading
SAN expert Chris Poelker compares connecting a SAN with wavelength cabling and dark fiber and discusses the pros and cons of each. Continue Reading
Storage expert Chris Poelker outlines WWN basics in order to answer the question: "Why do HBAs in a SAN have same base?" Continue Reading