I often blog about the challenges of managing rapid data growth in a world where budgets are growing at a slower pace than our information because this is a problem everyone faces. Literally, I hear about it in virtually every customer meeting. Plus in a recent study, 62% claimed limited funding for data protection initiatives, so there is no doubt that this is a top of mind concern for IT practitioners. Recently, I ran across an interesting post by David Merrill on the HDS storage blog. In the blog entitled “2014 Prediction #3 – Price Erosion Will Not Keep Up with Growth Rates”, David talks about some important economic concepts related to data growth and information which I think are very informative and thought provoking.
In the piece (and illustrated in the graph below), David shows how price/TB for disk storage has been decreasing consistently at a rate of about 25% which roughly equates to 25% customer data growth. The result is relatively flat capital expenses (CapEx) on disk storage (e.g., customers need 25% more storage to accommodate their growth, but since price decrease 25%, their total expenditure is approximately unchanged.) This trend is consistent in the chart (the green line) until 2012 where CapEx starts increasing due to slowing price/TB declines. This is a troubling trend for IT teams and implies that disk storage CapEx must increase which can be a real challenge in today’s capital constrained environments.
The big challenge facing IT is how to reduce storage CapEx in the context of a larger trend of slowing price/TB declines. One obvious answer is tape. Tape brings completely different cost metrics than disk (or even cloud) offerings and The Economist suggests that it is 2.5x cheaper than disk. Now, of course, tape access profiles are different than disk, but for many environments, it is an ideal medium for older infrequently accessed information. Most industry experts agree that approximately 80% of corporate data is older and inactive. Imagine what would happen if you stored this information on tape instead of disk – disk usage would plummet by 80%, backup windows would shrink by 80% and total price/TB would drop by about 50%! The entire CapEx price/TB equation would radically change and you would free up a large amount of CapEx for investments in other technologies such as flash storage, Software Defined Storage or any other technology du jour. The possibilities are amazing!
The key takeaway is that we live in a world of rapid change and growth. As you look to the future and think about investments and technologies, you must consider all options. Clearly the graph implies that investing in traditional disk arrays will become more capital intensive over time. There is an opportunity to think holistically about the technologies you might use to compliment your disk investment to maintain service levels and improve economics. Tape is clearly an important tool to consider when looking to the future especially when you consider its roadmap of increasing density and the maturation of LTFS.
Chart used with permission of author. Original source.