Virtualization and Cloud executives share their predictions for 2017.  Read them in this 9th annual VMblog.com series exclusive.

Contributed by Noam Shendar, Chief Operating Officer, Zadara Storage

The Uberization of the Data Center Continues

In tech industry parlance, it's easy to fall prey to binary thinking.  We either buy enterprise IT equipment, or we get it as-a-service and share.  We either choose a public cloud or a private cloud.  We've grown so accustomed to limitations and either/or choices that we fail to see this bright new world made possible by our collective innovation.

In 2017, we predict that "flexibility" will reign - new options for "sharing" will arise without former performance or other penalties - and the cloud will continue to attract enterprise-scale workloads and change the business model of how IT buys its solutions.  Here's a few specifics.

1)  Hybrid clouds will flourish as the great Unclouded join in.

At AWS re:Invent in late November 2016, we were amazed how many of the attendees were not current AWS customers, but were sent to the show to figure out how to leverage the cloud for their organizations.  Hybrid clouds will flourish in 2017 as those un-clouded companies opt to keep the crown jewels of corporate applications on-premise (as-a-service), but move other workloads to the cloud.

That's why hybrid clouds using On-Premise as-a-Service (OPaaS) are a great move.  They let enterprises enjoy cloud economics and scalability, yet retain control over what's most sensitive.  It's also why so many vendors have debuted On-Premise as-a-Service, private cloud solutions - we counted six in the past three months.  Keeping IT assets on-premise, yet letting someone else manage and maintain them, and paying for them as OpEx, as a Service, will be the next flavor of managed services."

2)  The Uberization of the Data Center will continue.

One might consider Amazon Web Services as the first Uberizer of the data center world.  What Amazon did was to allow customers to divide resources into increasingly thin slices and do unheard-of procurement practices - buying exactly what they want - from a simple Web browser. Like Uber, AWS changed not only in how IT is procured - but how it is consumed.  

We're already seeing? the same change for networking and storage domains.  With storage, IT Brand Pulse predicted that by 2020, 50% of all storage will be provided on an OpEx, pay as you consume, as a service basis.  By 2023, that figure jumps to 80%.

Like any transition, the Uberization of data centers will never be an all or nothing event.  Latency or bandwidth requirements may suggest a public cloud isn't the best way to run every application.  Architectures still will need to be built with forethought so they perform to SLAs. But at the end of the day, people running data centers are the same as people who have chosen newer ride-sharing options over traditional choices like buying a car and depreciating it for 5 years, or hailing a cab from the street corner.  When the market makes it easy for them to buy exactly what they want, they vote with their wallets.

3)  Flash will achieve price parity to HDD, or less.

CIOs who haven't taken a fresh look at flash drives owe it to themselves to do so in 2017.  A new generation of 3D NAND-based SSDs from Intel and others is delivering stronger price/performance and increased competition - such that we're seeing price parity to Hard Disk Drives (HDDs) for the first time. Modern flash drives last longer and have a more predictable lifetime than traditional rotating media, thanks to wear-leveling and other advanced controller techniques. This translates to less management overhead tied up in replacing failed drives and keeping inventories of spares - and thus less demand for IT management time.

The old saying used to be ‘All-Flash means No-Cash,' but as capabilities increase and prices drop, analysts suggest that transition to SSDs will accelerate.  Flash may be the CIO's best friend in 2017 and beyond.

4)  Bionic, hyperscale cloud storage will arise to meet the Big Data onslaught.

Big Data means Big Storage - and it will demand new, vastly scalable architectures that can support the data explosion that analysts foresee, both from the estimated 75 million Internet-connected devices that analysts expect by 2020 alone - and from very large data sets over longer time frames that data scientists will need to crunch.  Legacy IT systems were simply not architected to work at this scale. 

We'll see more enterprises build hyperscale cloud architectures like AWS, Microsoft Azure and Google Cloud Platform, but for the sole use of the enterprise itself.  Big Data capable storage will be built on software-defined storage, meaning they'll have a common interface to the storage system so IT can form them into flexible "mix and match' configurations.  They'll leverage cloud storage, object storage and even select on-premises storage systems as part of a seamless network run remotely.  

As result, storage will become more "bionic" - more autonomic, self-configuring, self-healing, and self-administered as AI assists in automating decisions on what to store, where. IT involvement will be more "strategic" rather than "tactical".  Data services will be performed remotely by third-parties who run, manage and upgrade this Big Data-scale storage.  Data services will transform into utilities following the same path as other common utilities that we take for granted, such as electricity, telephone, cable TV and Internet services.

5)  This year's sleeper of a trend - On-Premise as-a-Service - will accelerate as more vendors follow early innovators.

Since September 2016, six major vendors have announced On-Premise as-a-Service initiatives where private clouds keep the storage asset on side, yet it is remotely managed, provisioned and billed on a pay as you consume basis, after it is consumed. (In case you're counting, they include Oracle, IBM, HPE, Nimble, Dell/EMC/Microsoft Azure, and Zadara Storage).  Of course we're proud to have pioneered this approach way back in 2012 when Zadara Storage went live with its enterprise storage-as-a-service platform.  We see recent moves to embrace the ability for someone else to deploy, operate, maintain and upgrade the gear that sits on the enterprise' premise, and simply send a bill for what storage is consumed, as affirmation that the business model underlying storage itself has badly needed innovation.  For years storage systems forced IT managers to spend huge chunks of CapEx for unknown storage volumes and architectures to support the organization 3- to 5-year out, after which the investment was largely worthless.  The market has spoken, and sanity is starting to prevail.

##

About the Author

Noam Shendar, @NoamShen, is Chief Operating Officer (COO) at Zadara Storage, provider of enterprise Storage as a Service (STaaS) solutions for private, public, hybrid and multicloud settings.  Noam has over 20 years of experience in storage and chip-level engineering, product development and business planning.   He was formerly senior director of business planning and product management at LSI, leading a new product group in the Engenio storage unit (now a part of NetApp).  Before that he was director of corporate strategy for LSI, a strategy director for MIPS Technologies, and VP of applications engineering for iBlast, an entertainment technology startup.   Earlier he held research and engineering positions at Intel Corporation's Microprocessor Products Group.

Noam has a B.Sc. (with Honors) in electrical engineering from the University of Wisconsin-Madison and executive MBA from Santa Clara University. He holds a US patent in the field of digital television.