While virtualization technology has become a more common practice in today’s data centers for managing IT assets efficiently, there are still struggles when it comes to establishing effective policies. Ideally, organizations can dynamically map and model their business applications to the appropriate underlying infrastructure—helping to reduce business risk and cost associated with that application’s support. Companies not taking advantage of virtualization infrastructures are also increasing the likelihood of unplanned downtime for business applications.
This is where storage virtualization can play an important role. Today, you can virtualize through hosts, storage arrays, or networks—either Fibre Channel or Ethernet. Additionally, you can have either block or file-level virtualization solutions. But before you get around to implementing storage virtualization there is one exercise we recommend businesses take – setting your scope and objectives.
Storage virtualization could be one of the most cost-effective and strategic decisions a company makes. And like any strategic business decision, you should make it very carefully. Removing a storage virtualization solution can easily disrupt the application-computing infrastructure, so you should begin storage virtualization implementations with a thorough understanding your scope and overall objectives.
One thing to keep in mind from the outset is that a virtualized environment completely abstracts the relationship between the data presented to the server as storage and the location of the physical storage. This means storage administrators will not know precisely where storage capacity is allocated. This can present difficulty for IT departments when they troubleshoot application performance issues.
Ask yourself this question: “Is this project part of a broader virtualization deployment strategy? Or is it designed for a specific use, such as tiered storage, disaster recovery, or basic resource management?”
You’ll then want to determine if your virtualization project requires a specific platform integration. For example, does it need to support AS/400 systems, UNIX servers such as Solaris, AIX, HP-UX, special clustering considerations or data sharing requirements, or specific protocols such as CIFS, NFS, FC, iSCSI, FCoE?
Armed with these basic requirements, you’ll be better prepared to select a vendor that can most fully meet your scope and objectives. It’s likely you’ll live with your chosen vendor’s virtualization solution for at least 3-5 years, so continue your vendor selection process with specific goals and objectives—you want to select all your vendor partners carefully and diligently.
We’ll be discussing some additional things to consider when implementing storage virtualization in subsequent blogs, so come back often!