2014 AD: A year in storage. STILL not enough nails for the disk coffin lid

Beard Stroke Spending some of the festive season in San Francisco has given me a chance to look back at the wonderful world of storage and to ponder which may have been the biggest overall storage stories over the past twelve months or so.

Here's what I think are the strongest contenders:

• Splintering of SAN/NAS
• Cloud storage in public and hybrid clouds
• SSD capacity rise and foundry output limitations
• Unstructured data increase and rise of object storage
• Server SAN and hyperconverged systems
• Incumbent restructuring
• Rise of activist investors

The sharding of SAN and NAS technology was covered here and my thinking us that the sheer number and type of technologies proposing an alternative to legacy NAS and SAN arrays constitutes a secular change; having a central NAS/SAN no more, so to speak.

One alternative is the public cloud and virtually all legacy storage suppliers have either bridges to or abstraction layers for the public cloud. Amazon, Azure and Google are causing hybrid cloud havoc as incumbent vendors try to protect their customer data-centre installed base, and startups like Avere, Nasuni, Panzura and the acquired SteelStore, StorSimple, and TwinStrata provide on-ramps to the public cloud.

Are hybrid clouds buggy whips for an automobile era?

Flash has hit performance drive arrays and disk drives hard, and looming capacity rises, like SanDisk's claimed 16TB SSDs in 2016 (24TB TLC ones) could hammer more nails into the disk drive coffin lid - only there aren't enough nails as foundries are limited in number. For flash to fulfil its potential we need another 10-20 foundries, meaning a collective investment - by an industry bearing the scars from profit-killing gluts in the past - of +$100 billion.

That's highly likely NOT to happen quickly.

In opposition to the SAN/NAS-eviscerating public clouds are server-side SANs and hyperconverged systems, equally determined to cut through what they see as the SAN/NAS crap. With EMC and VMware on board, plus all the server vendors except Cisco riding the EVO:RAILs this seems to be a presently unstoppable technology trend that has legs.

Cast your eye at Symantec and HP splitting in two, Dell going private, and IBM selling off its X86 server business and you will realise that incumbent vendor stability is not a given. Will Cisco get more serious about storage? Will IBM get out of the storage business? Is another round of storage startup acquisitions going to get under way?

None of these three questions are unrealistic to ask, which is evidence of incumbent instability in the face of secular storage market and technology changes.

Lastly, a maturing SAN/NAS and storage networking market, with low-to-no growth and slow-to-react vendors, has created opportunities for activist investors like Starboard Management and Paul Singer's Elliott Management to take positions in companies as diverse as EMC, Emulex, NetApp and Riverbed and demand change up to and including selling the company.

The storage industry is passing through a multi-faceted transition as startups, incumbents and customers wrestle with the effects of these seven forces affecting them.

It ain't over yet, no fat lady having started singing. Year's end in 2015 will probably provide the opportunity for another view looking backwards and trying to make sense of it all.

That's the storage business for you, full of things that spin and stream and flash technology that - no matter what engineers tell you - has is by no means in a solid or stable state yet. Merry Christmas and a happy New Year. ®