When evaluating potential storage vendors and solutions, one method I have seen quite often is the feature compare. You might be familiar with this process. A listing of all a products features and functionality listed side by side against the competing alternatives is created. Typically the offering with the most check marks is deemed the best and then, budget willing, it finds itself inside your data center.
But what if the best answer for you business was at the opposite end of the spectrum? Bare with me here, this may sounds crazy, but too much functionality can be a bad thing. And by bad, I mean expensive.
A variety of storage offerings exist that may not have as many check boxes as their competitors, but provide a nice blend of affordability, robustness, and capability. HP, IBM, Dell, and NetApp all have affordable SAN storage options in the same space. While much of the storage press focuses on a number of organizations that are seeing exponential data growth, a significant number of companies are just not seeing the same dramatic capacity demands. These organizations may have more traditional business models, and while they use and rely on technology, their data needs stay in the 10s of terabytes and grow at a 10-20% pace. For these organizations, the best solution may be to right size your storage and buy just what you need. From multiple conversations with IT leaders, I have pulled together a quick checklist to see if you are paying too much for storage.
1. Are you using all of your paid functionality? See some common examples of underutilized capability below:
- Thin provisioning: Overprovisioning is important when capacities are tight and multiple applications are demanding more capacity than they actually plan to use. But today, with 6TB SATA drives available, you need to ask if this feature still applies to your specific environment. For example, if your organization has 10TB of storage, it might be more affordable to load up on some high capacity drives than pay for a feature you will never use.
- Replication: Many arrays offer a replication capability, but are you using it? Do you have a secondary site for replication? Or even a secondary system? If not, then you probably don’t need external replication. If you do, do you use a 3rd party software program for replication? If so, then the replication capability on the array is likely redundant and unnecessary.
2. Does the feature provide you the advertised value? Some areas to double check:
- Solid state and teiring: Performance is a difficult thing to measure and bottlenecks are even more difficult to isolate. If you have a smaller environment, then SSDs or even tiering between FC and SATA drives may be overkill. Storage processors and memory are also always coming down in price, and a next generation entry level array may have the same or better performance of a previous generation mid-range alternative. Prior to investing heavily in improving storage performance, it is prudent to make sure you identify the bottlenecks in your existing environment to ensure you will benefit from the additional spend.
- Scale-out technology: Scale-out architectures promote liner performance, scaling capacity with performance. While that functionality is useful, adding capacity comes with the added cost of more performance. Also, the feature is only applicable as you add shelves of storage. So if you don’t need the added performance or if you only need one or two shelves of storage, you may be better served with a scale-up option.
- Power and cooling: Numerous solutionsthat promote savings to power and cooling, and benefits tend to be more applicable to environments with lots of servers and storage hardware or where space and power are tight (think New York City). If you only have one or two shelves of storage, any savings from power cooling will likely be nonexistent.
- Ease of use: I am still waiting for a product to advertise itself as hard to use, but so far no luck. There are definitely a number of features that make life easier, but are they worth the cost? If it significantly simplifies a common activity to the extent that resources can be freed up for something else, then it provides value. If it reduces the number of clicks from 9 to 2 or simplifies a one-time activity, then chances are the benefit may not justify any additional cost.
3. What are your maintenance costs in the 4th and 5th years?
- I am sure that you may be asking, “What if the features are included in the array or part of a bundle?” Bottom line, features take developers and support personnel. You are paying for them one way or another. It may be with a higher base storage price, it may take shape with higher drive prices, but a good place to look is in the later years, typically 4th and 5th, for software maintenance pricing, which is often a refuge for the hidden storage costs.
The bottom line is that you are paying for every feature and function in your storage one way or another. If you are not taking advantage of all of it, you may be over paying and it may be time to find another option.
As always, I want your thoughts and feedback. Let me know if this was helpful.