North America

Hitachi Data Systems

Jumping into a discussion on Storage Economics is easy. We all want to talk about money. Getting it. Spending it. Saving it. Money (or finances) economics are part of our natural human nature, and the IT nature. Problem is, we often count the wrong types of money or look at money in too narrow of a focus.

I started dabbling in the economics of storage with a white paper on SAN ROI in the late 1990’s. At that time, justifying a SAN was technically acceptable, but the entry cost was often prohibitive (back when switches and directors were hundreds of thousands of dollars). Using economic principles to justify technology investments was a new dimension for me within a mostly-technical (Hitachi) company. The paper on SAN ROI was well received, and seemed to get me on the current path of various storage infrastructure & economic themes.

Some of my early influences on this topic came from an old light-blue book titled ‘Information Economics’ by Marilyn Parker and Robert Benson (Prentice Hall, 1988).

Some of my MBA books from the mid-80’s had to be dusted off as I combined topics around price elasticity to storage demand, IRR, NPV and predictive ROI. The IT slowdown of 2001-2004 provided some fertile ground to develop various proofs and persuasions to get IT people to ask for, and financially justify, computer spending during a time when IT capital spending dropped way off. Irrespective of the economic climate or country in which we may live, business principles of money and finance have to be included with technology strategies. I might need to define the meaning of Storage Economics:

  • Understanding the relationship of purchase PRICE and ownership COSTS of storage
  • Long-standing principles around the truth that price ≠ cost
  • Characterizations of what types of ‘things’ make up the costs of storage ( I have found 30 things so far)
  • What types of activities or investments companies make to reduce operating expenses (OPEX) - related to storage
  • Various comparative methods, using money, to assist in judgments that are made in storage plans

Storage Economics spans all topologies, vendors, architectures, mediums and countries. I find that many sales processes, from the client RFI/RFP to working with vendors was centered around price-per-MB. Being seduced into low price at the risk of higher long-term costs is a trap that we often get caught in. I often found that lower-priced solutions do not perform or produce the desired long-term business benefit, when you consider that business is centered around money.

My interest and passion is to connect the international language of business (that is money) to storage: storage architectures, storage strategies, topologies, decisions of every type around storage. People who learn the language of money, properly married to technology, gain an added measure of leverage in getting things done. And at the end of the day, that is what we all want to do, get things done.

I hope my ideas, suggestions and dialog through this format can help those who want to become economic heroes within their respective ‘spheres of influence’ . I find that economic heroes have a more interesting career within IT, than perhaps technology-alone heroes. Economic heroes are those people that learn the language of money and then connect these concepts with technology. These heroes can be influential and powerful in their own right in getting things done, and getting credit for the money.

5 Responses to “Jumping into Storage Economics”

  1. on 13 Apr 2006 at 6:13 am Mike Linett

    The best way to reduce the costs of storage is for vendors to allow customers to transfer their firmware licensing between systems within their data centers and other users. Just like cars. Can you imagine buying a used Acura or a used BMW and having to go buy a new firmware licnese to make the car run? A new car is a much of a platform or system as a storage unit, so why don’t vendors allow users to sell used software with their hardware?

  2. on 14 Apr 2006 at 9:00 am david merrill

    Thanks John for the questions about our future financials, this really isn’t appropriate for me to respond to, so you’ll just have to be patient and check out hds.com for the latest news.

    Now the cost of software, or software maintennace is only a fraction of the total cost of ownership. Check out the paper at http://www.hds.com/pdf/StorageEconomicsWHP-153.pdf to see all the things that can go into a storage TCO. Some of these are:
    Purchase price of HW and SW
    Maintenace costs
    Labor (up to 40% of TCO)
    Electricty and environmentals
    Risk of outage
    Cost of waste
    etc.

    Thanks for your suggestion though. Something to follow-up on.

  3. on 24 Apr 2006 at 10:42 am Jon Toigo

    I will be interested to see your future information on the economic gain rule applied to storage consolidation. There seem to be a lot of vendors telling customers to rip all of the storage out of branch offices and centralize it in one location. The result is almost inevitably slower response times for remote users and applications and a requirement to push out caching appliances of one sort or another to the distributed branches. At the end of the day, the costs and demerits of this model seem to override any cost-savings meted from the model. Am I missing something here?

  4. on 01 May 2006 at 10:31 pm david merrill

    John

    I will address this type of cost saving category in an upcoming post. You can download the paper that is referenced for more of a preview on this topic.

    You are correct in that IT can do things to optimize costs at the same time negatively impacting performance or availability or security etc. These negative impacts can then carry another cost for the enterprise. Cost or TCO is one dimension of the architecture/solution that we need to engineer in parallel with other qualities of the design. Consolidation can occur at the macro level, but what I see most of (for superior economics) is micro-level consolidation. Metro area, or within a data center domain is easier to realize consoldiation cost savings. Global consolidation of physical assets does cause other performance costs as you mention.

    In our work of defining 30 different types of cost, and looking at the types of activities (consolidation being one) needed to reduce cost, a holistic method for architecture and TCO is needed.

    You are spot on with your analysis.

  5. on 31 Dec 2006 at 10:09 am Costkiller

    Interesting story. Here’s a useful website to reduce the costs of storage :
    http://costkiller.net/centre-couts/cost-cutting-SAN.htm

Comments RSS

Leave a Reply

page top