Software-defined storage (SDS) is an approach to data storage management in which data services (software functions) are abstracted from the underlying physical storage hardware. This approach is drastically changing the landscape of traditional data storage. Rigid hardware silos are now transforming with the help of software-defined architectures into a more flexible infrastructure that is easier to manage and change.
Many of the challenges that IT/storage administrators face are being addressed and overcome by SDS. In fact, Gartner’s Strategic Roadmap for Storage states that 50 percent of global storage capacity will be deployed as SDS on-premises or in the public cloud by 2024 (up from less than 15% in 2020).
SDS enables organizations to have a free choice of hardware and build their storage infrastructure the way they want. By virtualizing, and aggregating storage capacity across an infrastructure, SDS helps to manage resource provisioning, data services, and data migration centrally and remotely. Other benefits include:
- The ability to easily switch devices between manufacturers and upgrade/change existing technologies (for example, from HDDs to SSDs, classical SAN to hyperconvergence, expensive NAS systems to lower-cost object storage, etc.)
- Freedom to customize and optimize the storage infrastructure based on budgetary needs, to meet performance requirements, adhere to data protection norms, and improve availability SLAs.
- Leveraging uniform and consistent data services across diverse storage media.
With SDS adoption growing significantly and many hardware vendors and third-party storage software companies offering SDS capabilities, it is challenging for IT departments and service providers to select the right SDS solution for their needs. There are differences in deployment models, limitations in data services, differences in support for storage models and access protocols, and so on.
There are specific criteria that IT teams should consider as they evaluate a software-defined storage solution for their storage environment.
Ensure Continuous Business Operations: Storage downtime and associated disruptions are a common challenge in all data centers. The SDS solution you are evaluating must be able to create data redundancy locally and over remote sites along with integrated fail-over, resynchronization and fail-back support. Additionally, point-in-time data recovery techniques (snapshots, backup integration, eventually continuous data protection (CDP)) should be included to return to normal operations based on the last good known data status in the event of data loss.
Enable Fast Data Access: Responsiveness of storage devices is critical for seamless application access and, in turn, a positive user experience. Make sure the SDS solution you choose covers inherent functionalities to improve storage response times. These include caching mechanisms, parallel I/O operations, and fully automated and transparent tiering of data to the right storage based on AI/ML-based algorithms.
Integrate New Technologies Seamlessly: To save future implementation time and costs, the selected solution should be able to provide simple ways to include new technologies (e.g., NVMe, Intel Optane, cloud, etc.) into the existing environment. Without having to upset data accessibility or cause downtime, it should be possible to add/switch/decommission storage equipment during business operations.
Achieve Non-Disruptive Data Migration: Migrating data is typically a complex process which requires tremendous manual effort and incurs additional expenses. The SDS solution you are evaluating should be able to control data migration seamlessly and remotely between disparate storage media.
Automate Storage Management: To enable IT teams to be more productive, they must be freed from spending too much time on unimportant tasks. The easiest way to achieve this is leveraging automation. Consider only SDS solutions that help you to automate repetitive manual storage administration and management tasks.
Centralize Capacity Management: Look for SDS solutions that can aggregate resources across diverse storage hardware, regardless of how they are connected. The ability to pool resources across disparate storage media helps to centrally manage capacity provisioning and load balancing.
Deliver High-End Data Services Across Diverse Storage: Regardless of the type of storage gear being used, the selected SDS solution must allow for uniform execution of all data services across heterogeneous storage infrastructures. Without having to toggle multiple management consoles, you should be able to manage data services for your storage infrastructure from a single pane of glass.
Gain Free Choice of Hardware Assets: Being bound to a single hardware supplier has many disadvantages. Among others, your choice of products/services would be limited, your price negotiation shows room for improvement, and technical dependencies limit your ability to act according to your needs. By having maximum compatibility to work with any storage equipment from any manufacturer, you get the necessary freedom of choice. Ensure the SDS solution works with hardware from all storage manufacturers for maximum flexibility.
One of the key highlights of a SDS solution is being hardware-agnostic and compatible to work with equipment from any storage or server manufacturer, giving organizations the ultimate flexibility to customize storage infrastructure. Maximizing the collective value from a storage infrastructure by delivering fast, uninterrupted data access, while reducing storage costs and increasing flexibility and scalability is critical in meeting business continuity and disaster recovery objectives.
Vinod Mohan is a senior product marketing manager at DataCore Software.