Western Digital Acquires sTec; SSD Shakeout Begins
June 24, 2013
Today Western Digital, market-leading HDD manufacturer, announced that its HGST subsidiary is acquiring struggling enterprise SSD vendor sTec for $340 million--about a 90% premium over sTec's share price last Friday.
This is just the beginning of a consolidation in the solid state disk (SSD) market, which, according to Storage Newsletter, currently has 110 vendors all hunting for a valid market niche.
- Are You Really Ready for a Software Audit?
- Want Information Fast or Want it Right? Learn How to Have Both
White PapersMore >>
In 2009, sTec was a high flyer because it was the exclusive provider of SSDs to EMC. Its stock traded at $40. At that time, sTec's SLC ZeusIOPS SSD was both the fastest on the market and just about the only Fibre Channel interface SSD around.
However, as array vendors, including EMC, shifted their controller designs from Fibre Channel disks to SAS, they could source SSDs from a wider range of vendors. Many chose to use suppliers such as Intel, Samsung and HGST, rather than rely on sTec.
When EMC announced it had over-ordered SSDs and would fulfill customer orders from inventory rather than buying more from sTec, the company's stock crashed and sTec's founding Moshayedi brothers were left facing shareholder lawsuits and an SEC investigation that they had sold stock at the top knowing the bad news was coming. The sharks were circling, with hedge fund Balch Hill pushing for change at the top.
HGST, which before the WD acquisition last year was Hitachi Global Storage Technology, is operated by WD as an arm's length subsidiary due restrictions imposed by China on anti-trust grounds. HGST has been a leader in enterprise SAS SSDs, due in no small part to a joint development deal with Intel that leaves HGST as Intel's path to the enterprise SAS market.
My spies tell me that HP/3Par's latest 7450 all-solid-state array, which I wrote about recently, is powered by HGST's SSDs.
HGST, and therefore WD, when the arm's length restrictions are removed in a year or two, will have sTec's IP portfolio. This includes over 100 issued or pending patents and an internally designed SSD controller that's at the heart of its current SSDs. It'll also get immediate access to the PCIe SSD market and sTec's EnchanceIO caching software.
I, for one, could tell that the end for sTec was near when it fiddled with its official corporate name, changing it from STEC to sTec. While there may be times where camel case makes some sense (see DeepStorage.net), wasting valuable corporate resources on a meaningless name modification when the ship is taking on water is a good indication that management hasn't got a real solution.
I expect WD/HGST will refocus sTec's products and marketing back to the OEM market. The company's recent attempt to sell directly to the data center market hasn't been terribly successful. In addition, a few of its latest products, like the Windows Storage Server-powered s3000 appliance, haven't had a visible raison d'etre.
The bigger question is what will it take to survive in the SSD market over the next few years. The flash vendors (Intel/Micron, Toshiba/SanDisk, Samsung and SK Hynix) have the cost advantage of vertical integration and a few others (LSI, OCZ, Virident, Fusion-io) own their own controller IP, but the vast majority of the players simply assemble SSDs from the same set of merchant components. How can Kingwin, Mushkin or even Kingston differentiate their products enough to stay in business? There may be no good answer.