“The dictionary is the only place that success comes before work. Hard work is the price we must pay for success. I think you can accomplish anything if you’re willing to pay the price.”
— Vince Lombardi
Over the many years in this business I have tried to simplify the price/ performance ratio down to a simple graph. However, I have often been stymied by the problem that our customers value their labor resources and overhead costs much differently.
When you combine costing issues with the fact that companies like NetApp quadruple processor speeds and quintuple NVRAM but don’t even double their OPS when they come out with a new model, it makes it hard to do comparisons. Is it cost effective to run new equipment at 20% efficiency or 3 year old equipment at 50% efficiency? According to industry statistics most storage equipment runs at below 25% efficiency. And what is the real storage value of a hard disk and the right sized value. For instance a NetApp 144GB disk usually reads as 136GB, so you lose almost 10% of your disk space right off the bat. So what are you buying in usable space when you purchase a disk system from NetApp or EMC. Is it just a guess?
The tests we see mentioned and the white papers don’t seem verifiable or repeatable in any environment other than the specific test environment. And it often seems that the test systems are optimized .
So it all comes down to hard work to produce a verifiable solution to the storage hardware questions of :
1) How much does it cost to run my old storage compared with a new solution
2) Will a new storage solution provide more storage at a lower cost per TB?
3) Will Replacing my current storage solution cost me less per terrabyte to administer?
4) Is it more cost effective to add storage to my older system or buy a new system?
Zerowait is working on a way to provide our customers answers to these questions.