Download the authoritative guide: Cloud Computing 2019: Using the Cloud for Competitive Advantage
Like everything else, when it comes to adding storage, there is a right way and a wrong way to go about it. According to Jim DeCaires, storage product marketing manager at Fujitsu America, companies add new storage for four primary reasons: capacity, a technology update, performance and availability. No matter what the reason, installing it the wrong way can lead to a host of problems.
"There are key questions that need to be answered for each of the four reasons for considering additional storage," said DeCaires. "If you do it wrong, you end up with the sprawl of storage islands, over-provisioning, etc."
Prior to storage expansion, DeCaires advocates carrying out an audit of current capacity utilization. This will point up areas of under-utilization and could potentially reduce or eliminate the need for the planned purchase.
"Companies must understand the utilization of their existing capacity potential," said DeCaires. "If it's possible to consolidate existing storage to capture unused capacity, do so."
Another way to reduce storage purchases is to understand the performance profile of existing and planned applications. By knowing these profiles and where the hotspots are located, storage can be matched up to accommodate peak loads. In some cases, this might make it unnecessary to buy more disks. The motto according to DeCaires is to know your performance bottlenecks well before you go. Sometimes you might have to add software or hardware; other times you might need to just tweak some settings. Throwing storage at the problem might not solve anything, so be sure you know what the actual situation is before submitting an embarrassingly expensive request for something that doesn't improve the user experience.
What about availability? Once again, DeCaires recommended thorough homework to really understand service-level agreements (SLAs).
"Availability requirements are driven by business SLA demands," said DeCaires. "Do you understand your outages, and have you targeted the correct issues for resolution? Understand the MTBF [mean time between failures] and MTBUI [mean time between unscheduled interruptions] of your environment."
Last but not least, technology updates are generally easy to see. If you don't have the options you want and performance is suffering simply because of outdated technology, the answer is obvious update it. But don't just keep accumulating the latest and greatest storage hardware and software. That might make you popular with your vendor or reseller, but it's a recipe for disaster. DeCaires said new technology requires a high level of patience and commitment to get everything running smoothly. For mission-critical applications, in particular, plenty of testing that focuses on proven solutions should be done in advance.
Go in With Your Eyes Open
On the same line, he made it clear that every major purchase should have its exit plan carefully evaluated before signing on the dotted line. If the exit process is really painful after it has been implemented, ask yourself if it's worth the trouble.
Read the rest at EnterpriseITPlanet.