Seagate thinks about going private (again)
Once is not enough
Seagate is talking with private equity firms about going private for the second time. Why?
The company is the leader by revenue in the hard disk drive (HDD) market - although Western Digital overtook it in unit shipments earlier this year.
One suggestion is that Seagate must rationalise its supply chain as it responds to market trends. This will affect share price, revenue and profitability. So private ownership would be a curtain for the company while it restructures to emerge stronger and revitalised.
Under such a scenario, Seagate execs could be asking questions such as: "Do we need to make disk platters in Northern Ireland? Why not make them closer to the plants that build the disk drives?" They might also look at the i365 business and wonder if the money invested in it might get a better return invested in the core HDD and coming core SSD (solid state drives) businesses.
For Seagate to consider going private a second time there must be serious problems that are better resolved out of sight of Wall Street. We think Seagate could make significant restructuring moves in the next couple of years, whether it goes private or not.
Let's cast a quick eye at the product range and look at the main trends affecting the HDD market, identifying where Seagate is on-trend or falling behind the curve.
Seagate makes a range of 2.5-inch and 3.5-inch drives for netbooks, notebooks, desktops, servers, storage arrays, external storage for desktops and notebooks, and for consumer electronic devices.
Basically there are two 2.5-inch disk drive brands: Momentus for notebooks and Savvios for enterprise arrays. There are two 3.5-inch disk drive brands: Cheetah for arrays and Barracuda for desktops. The Constellation products are for near-line storage in arrays and come in 2.5-inch and 3.5-inch form. Pulsar is the new line of 2.5inch SSDS for server and array use. The CE area has three specialised brands: Pipeline, SV35.5 and EE25 Series.
Seagate offers a FreeAgent line of external and portable drives using either 2.5-inch Momentus or 3.5-inch Barracuda drives inside the casing. There is also a BlackArmor network-attached storage (NAS) range of 4-bay desktop products. Lastly, Seagate's i365 subsidiary supplies data protection for small businesses with a range of eVault appliances and cloud services.
HDD market trends and moves
The main storage market trend is unremitting data growth, particularly in business unstructured information and in consumer content generation. Also, a number of specific trends affect Seagate.
First, the 3.5-inch Fibre Channel drive enterprise array technology is under threat from two directions. Solid state drives (SSDS) and SAS drives are replacing Fibre Channel drives, and fast (10K) 2.5-inch drives are replacing 15K 3.5-inch drives in storage arrays so as to get more IO/s and drive density - the number of spindles in an enclosure.
Multi-level cell (MLC) flash is beginning to appear as a second flash tier in enterprise storage arrays, as it is cheaper than faster single level cell (SLC) flash. Seagate has signed a deal with flash manufacturer Samsung to co-develop MLC flash controller technology. In notebooks the fitting of 3-bit MLC flash threatens HDD sales as battery life will go up and response time go down. When capacity is big enough, this will surely affect desktops as well, with bulk desktop data storage transitioning to USB 3.0-connected external drives.
Seagate's flash-embedded Momentum XT hybrid drive is a halfway house to SSD and has promise. However, no other supplier has followed suit except Hitachi-LG with a flash-embedded Blu-ray drive.
The Constellation drives need a 3TB product if Seagate is not to fall behind WD. Current PMR (perpendicular magnetic recording) disk drive technology is running out of steam and a vastly expensive shift to HAMR (heat-assisted magnetic recording) or BPM (bit-patterned media) technology is needed with an interim shingled write possibility to increase track density on a drive at the expense of longer write update times.
Seagate appears to have fallen behind the areal density curve with 3.5-inch SATA drives. Its 3TB Barracuda XT, as used in the 3TB FreeAgent GoFlex external disk product, is a 5-platter product with 600GB/platter. WD's 3TB MyBook external disk product has a 4-platter drive inside it with 750GB/platter.
Tablets such as the iPad use flash and not hard disk drives and are beginning to cannibalise netbook sales and may slow growth rates for notebook sales as well, both fates which will lessen demand for netbook and notebook hard drives.
Commentators such as Barclays Capital analyst Ben Reitzes suggest that Seagate may need to go private because the Tablet Effect will could see disk drive sales into the overall notebook/PC market fall, rendering forward unit shipment and revenue forecasts unreliable. They liken Seagate's wish to go private now to its riding out the dotcom bust by being private then. Let's translate this sentiment: the tablet flash effect on the netbook-to-desktop market will be as drastic in its effect as the dotcom bust was. Are they serious? Really?
Commentators who take this line say Seagate's enterprise revenues mean it can better withstand a netbook-to-desktop HDD demand crash than Western Digital. In that case WD's executives and board should be panicking, and there is no sign of that.
We think that 3.5-inch hard disk drive sales into enterprise arrays could fall as the performance tier goes to flash and 2.5-inch SAS, although the bulk capacity tier should remain a 3.5-inch SATA preserve. Preserving SATA's capacity lead over flash and satisfying future capacity needs depends upon the expensive development of HAMR or BPM. Where is the cash coming from?
The notebook market will have an increasing flash component and Tablet sales will rise and they will mostly use flash storage. Where does this leave us?
We think that the main problem areas are flash and the HAMR/BPM transition, with Seagate falling behind in 3.5-inch areal density presenting a subsidiary issue. The company has to claw its way back and then transition to HAMR or BPM in the 3.5-inch SATA product area as well as the 2.5-inch SAS area to preserve capacity growth rates.
It has to get into the enterprise storage array SSD space with a line of SLC and 2/3-bit MLC products. (4-bit down the road) and also get into the tablet and notebook space with SLC and 2/3/4-bit MLC product. ®
Sponsored: IBM FlashSystem V9000 product guide