Storage systems have gone from being simple to being swamped: There’s more data than ever before, from more sources than anyone could have imagined just a few years ago. At the same time, research, regulatory, and compliance requirements are making the challenge of data management more complex.

“Storage is at a breaking point across the enterprise,” says Simon Robinson, research analyst for 451 Research. “Most organizations have an ‘accidental architecture’ of storage solutions that were put in place to meet specific needs, but the result is a costly, complex, and unscalable hodgepodge of systems that consume management time and can’t support changes in the business environment.”

Today’s rigid storage systems can’t grow or change as needed. Every hardware-based storage solution ultimately has a size limit—and so far, every one of them has proven too small. The only type of storage that has a chance of handling the data deluge is one that’s structured and defined by software, with the ability to create a single, efficiently managed storage infrastructure that’s not constricted by hardware capacity and instead manages data across a vast and (if necessary) rapidly expanding array of commodity and even re-purposed hardware.

Software-defined storage, or SDS, coupled with two other trends—the evolution of two-tiered storage and the drive towards convergence—offers a new approach to storage that seems ideally suited to the continual radical change and data growth that characterize today’s business environment.

Here are key trends in storage that organizations can look to as 2015 gradually gives way to 2016:

1. SDS: Virtualization for storage infrastructures

Server virtualization is transforming compute capabilities and software-defined networks are changing networking capabilities, enabling organizations to scale up and improve agility while saving money on hardware and operating expenses. Software-Defined Storage (SDS) is doing the same for storage, helping organizations manage vastly growing data stores efficiently and cost-effectively, so that data can support their business decisions and add to competitive differentiation, rather than absorbing all their IT resources.

2. SDS will continue to dominate the industry narrative

Research suggests that storage accounts for an average of 30% of organizations’ total IT budgets (CapEx and OpEx) – and as much as 60-70%, so it stands to reason that the price of storage will be key in determining the cost-effectiveness of IT. SDS is proven to help organizations handle the data deluge while avoiding massively expensive storage systems (accurately called MESS) in favor of right-sized, scalable, and easy-to-deploy solutions. From a CapEx, OpEx, and performance perspective, SDS delivers the lowest cost per desktop with the highest IOPS. So, companies will definitely be able to achieve innovation and value by adopting SDS.

SDS supports innovation and open business models with the lowest TCO, making it good for both customers and partners. While “openness” varies from one SDS solution to another, the best of them can run on any standard x86 hardware and HDD/SSD and are hardware-independent, both as a storage platform or OS, installed on a standard server or as a virtual storage appliance that runs in a hypervisor. It can support all platforms, all stacks, and all workloads, making it a good fit for any vertical, public or private cloud, workload or app, cloud platform or stack, server platform, or sub-platform. For organizations, this delivers the agility to scale and deploy storage in response to changing business needs. It also invites open-source collaboration, innovation, and flexibility. Organizations can cut their storage costs by 50 percent or more, reducing capital expenditure by purchasing commodity hardware and reducing operating expenses through streamlined management and ease of use of SDS.

There is still no commonly agreed definition of SDS, so individual SDS approaches vary widely.

In addition, OpenStack is likely to be the basis for offerings going forward, as it promises further development and hardening of block-level storage support, includingCinder, as well as the Swift object storage engine and API. Meanwhile, storage providers will continue to offer support and compatibility with other leading cloud-based APIs, especially Amazon S3.

3. The evolution of storage architecture will accelerate

Predictions continue to point to the evolution of two-tier storage architectures, where ‘hot’ data is served from host-resident nonvolatile memory, while all other data—older data, backup/snapshot copies, etc.—resides on large-capacity HDDs, most likely running an object-based storage system.

To get the fastest performance for their‘hot’ data, organizations are turning to all-flash arrays (AFAs), and the market for these is growing quickly. But though predictions say that flash will become the dominant primary storage method as falling prices drive continued adoption of flash in multiple form factors during 2015, it’s not the whole story: There’s also that “long tail” of data that needs to be stored, protected and retained, but is less time- or mission-critical. SDS helps organizations to address the full range of their storage needs by speeding and simplifying the process of moving data to the most appropriate hardware, while consolidating access and management of the data with the rest of the company’s data stores.

In response to changing storage requirements, the mix of storage types—object, block, and file—looks set to evolve too. In particular, demand for object-based storage is likely to continue to grow.  Partly this is because more organizations now have the very large environments—with more than 1PB of unstructured data—that are most appropriate for object-based approaches. But also, object-based approaches are inherently software-centric and designed to scale massively on commodity hardware. It’s entirely possible for organizations to implement software-only scale out block and object storage solutions that can deliver high-performance global inline deduplication on petabyte scale clusters.

4. Convergence and hyperconvergence will continue to ramp up

Convergence is increasingly becoming the name of the game in all aspects of IT, because it promises simplicity, seamlessness, and high effectiveness. But while all forms of IT convergence are attracting attention, the benefits of convergence are potentially the greatest in the storage domain.

Clearly, IT-wide convergence is still some distance over the horizon, but hyperconverged storage environments are being built today. These combine the simplicity of a converged storage-server architecture with SAN-like data resiliency and other services, making it easy to share files across storage pools and users, for maximum seamless scalability. Robinson predicts that “adoption of such hyper-converged offerings will be concentrated chiefly in small and midsized enterprises, although there will also be growing penetration in enterprise departments and remote office and branch office (ROBO) environments.”

5. The storage revolution is happening now

To ensure they’re on the right side of the storage  ‘tipping point,’ organizations need to make sure that the storage solutions they buy today both fit into their current models and also give them a path forward to providing storage at a scale they don’t yet believe they’ll need: a bridge,as it were, into a fast-approaching future.

“Change will not happen evenly, and the rate of change will vary greatly between organizations, depending on their attitude and approach, as well as size,” says Robinson. “But we believe change will happen; the storage challenge within many organizations has reached a tipping point, and even organizations that previously proved most resistant to change are contemplating new approaches. In this regard, we think 2015 will be the tipping point for many of these emerging approaches.”

While it may be true that every organization is now in the storage business, it’s also true that the vast majority don’t want storage to be their primary business. They just want storage to be simple again—and it’s up to the vendors of today to adopt and adapt SDS into turnkey solutions that can deal with the staggering volume and complexity of today’s storage environments while making storage management look simple to users --  especially if they want to compete in the storage landscape well into the future.

Tarkan Maner is executive chairman and CEO of Nexenta, which develops open source-driven software defined storage solutions.

Register or login for access to this item and much more

All Information Management content is archived after seven days.

Community members receive:
  • All recent and archived articles
  • Conference offers and updates
  • A full menu of enewsletter options
  • Web seminars, white papers, ebooks

Don't have an account? Register for Free Unlimited Access