Shirish Phatak is the Founder and CEO of Talon.
There’s a revolution taking place in corporate data centers today–and a bit of a tug of war, as well, as to who gets to decide the ultimate definition of software-defined storage.
With the amount of information companies must handle growing at explosive rates, traditional storage solutions, such as those based on SAN or NAS architectures, are being pushed to, and past, their limits. The old model, in which new storage capacity is added by purchasing more and more custom-designed dedicated storage appliances, is simply no longer financially viable; IT budgets just can’t keep up. Traditional storage systems are siloed and increasingly complex to maintain and scale, restricting enterprises to perform at their maximum efficiencies. Storage modernization is critical for global enterprises to sustain and thrive in the long run.
Additionally, the reduction of IT resources-to-user ratios, highlighted by the trend of enterprises transitioning infrastructure to cloud-based data centers, means that there is less distributed IT staff to manage disparate server and storage footprints.
One of the answers to this data management situation is that a growing number of data centers’ traditional storage paradigms are being displaced by software-defined storage (SDS).
SDS lifts a storage system’s intelligence out of hardware and into a layer of software that manages the entire storage infrastructure as a single entity. Users leave behind the need to deal with the specialized interface protocols and configuration complexities of individual storage units or subsystems.
A corporate imperative in 2017 is that enterprises need to manage all storage, which may consist of many different types of devices and media, located anywhere in the world, through a unified “single pane of glass” software portal. In addition, the increasingly global nature of the enterprise workforce, coupled with merger and consolidation activity, has driven an increased focus on data centralization across geographical boundaries. Software-defined storage management empowers enterprises to enhance their cloud storage environments and data protection capabilities.
SDS Enhances Scalability
Today, data centers are handling gigabytes and petabytes of data on a daily basis, and traditional SAN and NAS implementations are straining to keep up with soaring capacity demands. And the majority of that data is unstructured (file) data, with much of it stored in distributed locations.
Over the last five years, the industry has seen a strong move toward software-defined solutions to consolidate to a single footprint and modernize the storage and management of data.
Often these systems consist of high-performance storage arrays, each having one or two controllers that manage a number of disk drives. Storage capacity can be increased by adding more drives, in a scale-up approach.
When the amount of data being stored on the array is relatively small, that strategy works well. But as more disks are added to the array, the controller itself eventually becomes a bottleneck as it reaches its throughput limit. The exponential growth of unstructured data, estimated at over 20 percent per year, has led to a requirement for more scalable storage repositories, a far easier task to implement in a data center or a cloud environment than in distributed offices.
SDS is defining itself as the response, and redefining storage as scaling out rather than up. Instead of adding more disks to individual storage arrays, one of SDS’ incarnations expands capacity, adding storage available to the system as well as compute power. As a result, SDS implementations increase in performance as they expand in capacity.
SDS Enhances Flexibility
Traditional storage is custom designed to each manufacturer’s proprietary specifications, expensive to augment and usually incompatible with the products of other vendors. SDS has leveled the playing field in overcoming proprietary solutions..
SDS provides data storage flexibility without much physical restriction to accommodate the changing data storage needs of a distributed enterprise.
The core functionality of SDS is to provide maximize flexibility. SDS can also be used to add new features to old storage systems without investing in new hardware. Because intelligence resides exclusively in the SDS software, system-wide functionality is determined by customer-specified policies in the software, rather than by the unique characteristics manufacturers built into their hardware devices. This allows SDS implementations to include different types of storage units and media, such as hard disk arrays, flash drives, cloud-based storage, and, most notably, inexpensive commodity hard drives. This is done all without users having to be aware of where, or on which devices, data is actually stored.
SDS Enhances Agility
SDS’s value is in how it enables agility – the speed with which an organization can adjust to changes in technology or in its business environment. With traditional NAS or SAN installations, making significant changes to a company’s storage infrastructure takes time because of time required to fund, purchase, install, and configure new hardware. That lack of nimbleness can make it very difficult for a company to take advantage of new opportunities as they arise, or to effectively address problems before they become crises.
SDS adjustments to an organization’s storage operations can often be accomplished in software, reducing the time needed to adapt to new conditions to hours, minutes, or less. And modernizing distributed storage allows enterprises to manage their data center architecture more effectively and reduce the storage footprint throughout the enterprise. For example, when an application’s demand for storage unexpectedly surges, SDS software can, based on pre-programmed policies, automatically and within seconds allocate additional capacity from its storage pool to meet the requirement.
SDS Enhances Efficiency
SDS also substantially enhances data center efficiency. Because SDS allows an organization’s entire storage infrastructure to be centrally controlled through a software portal, the storage management task is greatly simplified. This brings us once again to the reduction in IT staff. No longer would it be required to spend time configuring and managing each individual device. Most storage management tasks, including data deduplication, replication, snapshotting, disaster recovery, and provisioning of new capacity, can be handled automatically by the SDS software. And increased speed saves time and money.
The Future of Storage Belongs to SDS
The amount of data that companies are required to handle is staggering, and ever increasing. Because traditional storage models are clearly no longer adequate to handle the deluge, the software-defined storage paradigm extends the value of SDS to the distributed enterprise, allowing it to benefit from technologies transforming the data center. SDS is the next step in the evolution to integrating and adopting modern storage architectures that have flexibility to scale, be agile and be cost-effective.
Opinions expressed in the article above do not necessarily reflect the opinions of Data Center Knowledge and Penton.