After the sale is when the costs become real
Years ago I bought a car that seemed like a great deal. It was a relatively fancy car, but I soon discovered it had a lot of hidden costs.
After owning it for several years, I tried to upgrade the GPS system that was included with new maps. I quickly discovered that I could only purchase the special software from the dealership for more than the cost of buying an entire new GPS system. Considering I was planning on selling the car within a year, it did not make any sense to invest any more money in it. When I went to replace the car I was encouraged to buy even more special accessories “up front” as it would be cheaper than trying to add them after the fact. I saved my money and bought a smartphone to stick in my car, and apps as I needed them instead. I wanted to avoid overbuying and being stuck with “more car” than I needed.
Car sales, much like enterprise storage, are predicated on the idea of getting the best deal up front, and then extracting high margin services or for anything not purchased up front. In this way the actual cost is often not discovered until after the initial purchase price.
HyperConverged Infrastructure (HCI) powered by VMware vSAN helps optimize costs in many ways. It is time to stop facing unpredictable costs, stop overbuying, and aligning storage costs with the needs and timing of the business.
Problems with not controlling purchase timing
One Big Buy – For me, it was not uncommon to discover that while an array would be sold at a 75% discount to me off of list price, adding an additional shelf of drives would only see a 25% discount. The end result of this pricing model was a tripling of prices for expansion of an existing array. This model encouraged IT departments to buy more up front to maximize savings. Purchasing with vSAN was a much less complicated situation. There was no need to stock up on storage for the next four years as servers and storage in servers do not have punitive price increase if purchased on a slower trickle.
Poor Chargeback – I’ve met plenty of storage administrators who had to keep alive out of support storage systems, or who teetered on an out of space problem. This was commonly caused by the infrequent, and variable substantial replacement costs associated with traditional storage systems. Storage would widely be viewed as a “sunk common cost” until one small project would push a system into an out of space condition, or over the performance capabilities of a system. Explaining that a project that accounted for 2% of revenue was going to require a 100% recapitalization of the storage platform generally caused problems with the CFO.
vSAN powered HCI avoids this, by allowing you to expand storage capacity and performance in granular amounts. Tying hardware purchases that are proportionate to the size of a project allows IT to “feed” the capacity of clusters without any sudden million dollar requests.
Poor Timing – I live in Houston which is home to a lot of Oil and Gas IT departments. Seasonality of storage purchases and support renewals does not always align with the seasonality of cash flow to industries that experience large swings. Oil companies that can “scale up” cost effectively do well in the good times, and companies that can “scale down” rapidly are the only ones who survive when oil gets to $25 a barrel. In the down years, I’ve heard CIO’s say they couldn’t spend $50 to save $200 as budgets become frozen. Underbuying means the expansion is not efficient. Overbuying means massive replacement and support renewal costs that may break a company in the down years. Many industries operate this way.
vSAN powered HCI allows scaling up one node at a time, and the ability to remove only the oldest nodes as they become obsolete provides for much more granular spin downs. Combined with vSAN powered VMware Cloud on AWS, storage can become completely open based if needed.
Vendor Relationships – Traditional storage often involved only seeing your “account team” every few years when the “big buy” was coming. In many cases it felt like a hit and run relationship that operated on the principal of the first of the Rules of Acquisition. Switching to a VSAN powered HCI purchasing experience gives you more frequent but smaller bites. This can change the relationship to more of a partnership than an infrequent transactional relationship.
Idle Resources – In my last blog, I discussed how assets that are not yet deployed can sit idle and cost you a fortune. A storage array’s capacity that is deployed but not in use is even more expensive than an array sitting in a box. It should be noted that predicting the needs for storage for a three to five year period is a complex task. You must balance the performance, and capacity needs that each has multiple dimensions. For performance, you have to balance controller, cache, storage network throughput. For capacity, you have to balance expansion costs against support costs and the timing of replacing the system. It’s a careful multi-dimensional juggling game where you try to buy resources, at the right time, and use them for the maximum amount that is economically feasible. Given that this 3D chess game is plaid with new applications and business requirements coming in all the time, there are always going to be significant wastage with this “buy upfront model.”
vSAN powered HCI changes this model to allow just in time expansion of the right resources when needed.
Time Value of Money (TVM) – Spend any time around a finance department and you will quickly realize that they would much rather have $100 today than $100 tomorrow. Their department’s purpose is to find the highest return on the money that the company has today and prioritize spending based on this model. Money that is saved in the form of upfront overbuying off of an MSRP number that is significantly inflated against what the cost of a drive can be purchased in a server is a false dillema.
Switching to vSAN powered HCI enables a department to get competitive and realistic prices for drives at any time.
Removal and replacement Limbo – The cost and time associated with migrating off of a storage platform are amplified by consolidating storage purchasing. As discussed in a previous blog, With vSAN disk groups and individual drives can also be evaluated for removal. Decommissioning and replacing a single server is a relatively trivial exercise for a single individual to perform quickly. Decommissioning large storage arrays can require a forklift.
Overbuying has costs
It’s time to stop overbuying storage to “Save” money on storage costs that were artificially created to force more massive, more wasteful purchases. vSAN powered HCI solves this problem by allowing cust in time purchasing in the form of expansion as granular as a single drive per host, or one server at a time.