You can’t coach speed; either you have it or you don’t. Coaches in most sports readily subscribe to that statement.
Same thing in the IT business. Either your networks and devices are fast, or they’re not. Sometimes they can be tuned to become a little quicker, but like athletes, that fundamental ability has to already be there.
In 2015, the data storage sector of IT, like all its counterparts, saw a continuous amount of general improvement in both speeding up the flow of data from node to node through more intelligent networking and in the increased use of NAND flash, which runs rings around hard disks in terms of performance.
The year also was marked by the continued “race to the bottom” in cloud storage pricing. Amazon, the unquestioned world leader in storing data in its own data farms, isn’t even sure how much data it currently holds for businesses and individuals. It’s doing so well at this that it has dropped its pricing 45 times since 2008, and we may well see further reductions in 2016.
Others, including Google, also have had to compete by dropping their prices. If this all continues to go according to trends, cloud storage will be completely free of charge for everybody in a couple of years. Well, maybe not quite that cheap.
Enterprises now trust the cloud for storage
This, of course, all revolves around the fact that people and businesses are trusting cloud storage and accompanying services more and more with their important business and personal data — knowing that Amazon, Google, Microsoft, HPE, Dell and all the other cloud storage vendors have multiple backup systems in place so that nothing ever gets lost.
In fact, it’s almost impossible for a file, a log, an image or a video to get lost in a cloud storage service. If it does disappear, you can bet that 99 percent of the time the owner accidentally deleted it himself.
Finally, 2015 — as most previous years — was marked by ever-increasing capacities of hard disk drives. Samsung last August came out with a 16TB drive — at a cost of $7,000 each — to leapfrog over WD, which came out with a 10TB drive in 2014 to jump over Seagate and HGST’s 8TB drives.
Data protection is the one factor that wasn’t improved markedly in 2015, although some vendors will argue to the contrary, blaming virtually all data breaches on the human factor. And there is a lot of truth in that.
So what were the key data points in data storage in 2015? eWEEK’s Darryl K. Taft published a slideshow based on IBM’s viewpoint of the year gone by, and it’s very relevant. But it’s strictly IBM’s point of view.
No. 1 Data Storage News of 2015: 3D Xpoint
Perhaps the biggest, most important data storage news of the year was Intel and Micron’s announcement last July of 3D XPoint (pronounced 3D cross point), an entirely different chip architecture that ostensibly improves data movement by orders of magnitude.
3D XPoint is the first completely new memory form factor since Toshiba brought NAND flash to the market way back in 1989. It literally takes the memory chip up into new, uncharted territory by becoming three-dimensional, instead of flat.
The parents of the newborn non-volatile memory, which has been several years in the development womb, claim it is 1,000 times faster than NAND flash, and if that guess is only 10 percent accurate, it’s still pretty fast. Intel and Micron predict a wide range of business benefits and use cases, ranging from speeding up stock markets, scientific research and oil and gas exploration to making more realistic video games—if that indeed is possible.
3D Xpoint will be coming into availability in Q2 of this year. When it gets into production processes, then we’ll see what kind of impact it will have on bottom lines.
Software-defined storage gains momentum
It’s surprising that many people still do not understand the concept of “software-defined” anything. If a technology is controlled by IT automation, then it is “software-defined.” Humans set the policies. It’s not that complicated. All software defines something or other.
In 2015, software-defined storage began to make huge inroads in daily IT production, with intelligent systems from IBM, Microsoft, Dell EMC, HGST, Datacore, Nexenta, Brocade, Fusion-io, Asigra, NetApp, Barracuda and others selling well into the markets.
Software-defined storage will automatically look at incoming files and stick them in the storage tiers in which they’re supposed to be stuck. For example, hot data involving point-of-sale transactions will be sent to a hot data flash node for a couple of weeks, to be sure the customer is satisfied with the purchase. Thereafter, it will be relegated to a slower hard disk node for perhaps a year, depending on policies.
Finally, it may go into cold storage on digital tape somewhere in the world to be held for seven years, or as long as the local and national laws require.
None of that drudgery has to be done by humans anymore; new-gen storage software takes care of all of it. More of this will be taking place in 2016.
Businesses of all sizes now trusting the cloud for storage
More and more small and medium-size businesses are moving to cloud-based storage in pursuit of lower costs and greater agility. Yet most enterprises need to pursue a hybrid strategy, retaining a significant amount of storage capacity on-premises for reasons of performance, security, regulatory compliance, cost and the avoidance of cloud service lock-in.
Typically, an enterprise will keep mission-critical data in-house, using the cloud for lower-priority data and for coping with episodic capacity requirements that fall outside the normal run of business. In 2016, we’ll see more solutions based on software-defined scale-out storage to provide enterprises with more flexible management capabilities.
Even though there was news just about every week about data breaches in companies large and small, most users realize that humans — not the technology — is to blame for most of them.
Security is the biggest issue in all of IT going into 2016, and that goes for all of IT.