Consolidation and Quagmire
Seagate: Can There Be Only One?
What's this
Today, after all of those scores of exits and mergers, we are left with three hard drive manufacturers: Seagate and WD, each with roughly 45% of the market, and Toshiba with the remaining 10%. (In October 2011, Indian firm Simmtronics opened a hard drive fab in Dubai, but with essentially no market share, it’s unjust to rank the company alongside the other three.) Perhaps it’s worth noting that Toshiba has exactly one 3.5” desktop SATA drive.
Two primary forces have propelled this consolidation. On one hand, there are inevitable cases of strategic acquisition for the sake of building market share. The bigger fish eat the smaller fish in order to get bigger still. However, the less obvious factor is that it has become increasingly difficult not only to stay in the storage game but to advance it. Magnetic storage has its own version of Moore’s Law, with capacity—as measured by areal density—doubling or increasing by 10X every so often. Areal density is the number of data bits able to fit within a given square inch of platter space.

As you can see in this late 2008 chart from the Journal of Magnetism and Magnetic Materials, areal density has progressed in a fairly linear manner for most of the last 50 years. In the 1980s, thin film magnetic recording allowed for improvements in areal density, but not at the pace of rival technologies, hence the slackening of ascent in the chart. Magnetroresistive and giant magnetoresistive head technology put hard drives back on their former path and bought the industry enough time for perpendicular recording technology to evolve, which dominates today’s hard drive products.
From 2004 to 2010, perpendicular recording maintained the storage industry’s usual 40% annual areal density growth rate. Then progress slowed, like had suddenly run into muddy ground, just as thin film had decades before. The optical lithography processes behind chip fabrication are running into similar issues. As the individual units of the technology (magnetic particles or transistors) become ever smaller, eventually they begin to run up against the limits of physics. Ultimately, magnetic particles can only be made so small and packed so densely before they can no longer be reliably written to and read by head mechanisms. With each step forward, the path to the next density advance becomes more difficult and costly.
Perhaps more than anything else, this increasing difficulty lies at the heart of why we only have three hard drive manufacturers left. Smaller players simply lacked the resources and vertical integration necessary to make areal density advances feasible. The laws of natural selection ran their course, and the lesser players gradually exited the field.