Storage virtualization has been a thing for decades. The early EMC storage arrays that lay at the heart of storage area networks (SANs) were an early attempt to create a virtualized pool of storage. So successful was the concept, that EMC became the dominant force in storage for decades and is now part of Dell Technologies.
Since then, more and more areas of storage have become virtualized. Here are some of the top trends in storage virtualization:
1. Cross-cloud and Cross-regional Storage Virtualization
Organization data is largely decentralized across data centers, clouds and different geographic locations. With data spread across cloud and regions, using virtualization to unify data access through storage abstraction simplifies many aspects of data and artificial intelligence (AI) platforms, including security, governance, and vendor lock-in prevention.
“While different types of virtualization (data, data, storage, and network) are touching upon newer avenues, 2022 will see an increased investment in the virtualization of storage systems,” said Adit Madan, Director of Product Management at Alluxio.
2. Storage Versus Data Virtualization
With that trend in mind, Madan recommends that enterprises prioritize storage virtualization over data virtualization. The abstraction of storage instead of computation, as in data virtualization, helps migrate away from legacy computing technologies instead of creating redirection to new and old computation. With storage virtualization, organizations can extend the life of existing systems and reduce the time spent managing and migrating data from storage silos.
3. Increasing Consumption of Unified Scale-out Storage
Consumption of scale-out storage with both file and object capabilities is increasing as more applications require it. First proven in larger environments, small and medium size enterprises now have confidence that choices in scale-out storage are becoming affordable and easier to manage.
“It is important that file or object workloads can be easily supported together as the needs of applications often and usually change quicker than the infrastructure,” said Morgan Littlewood, Senior Vice President of Product Management, iXsystems.
As such, the classic definition of NAS has evolved from network-attached storage to network, applications, and storage, suggested Littlewood. Data center environments will always have a mix of old and new technologies, where the requirement of architecturally supporting a diversity of workloads trumps standardization.
“A unified storage architecture that scales-out in support of a shifting mix of containers and virtual machines that use a mix of file, block, and object offers the opportunity of infrastructure standardization as the diversity of workloads only increases,” said Littlewood.
4. Kubernetes Adds Greater Virtualization
Kubernetes with persistent storage and edge computing is changing the way IT architects will think about storage and virtualization. Kubernetes and container storage interface (CSI) are becoming the way that storage is generally controlled for significant simplification and integration of applications and storage.
New workloads like AI, for example, require massive datasets, a high degree of parallelization, and high-performance compute and storage. The global enterprise artificial intelligence (AI) market is anticipated to grow at a compound annual growth rate (CAGR) of 39.7% to $309.6 billion by 2026. This requires scale-out computing and storage performance, GPUs, and much more storage capacity. No single vendor provides all the pieces today.
5. Open-Source Software-Defined Growth
IT costs have gotten out of control once again. This may come back to bite established storage software providers in the enterprise market. The storage market, after all, is mature with ample choice among products for demanding requirements. At the same time, businesses are grappling with storing more data across several tiers.
Yet, features such as corruption prevention, data/snapshot efficiency, and write persistence in the face of power outages and hardware failures have traditionally been restricted to proprietary enterprise storage solutions. Littlewood believes that open enterprise storage software is gathering momentum even for demanding applications as a way to lower cost up front and over time. Software technologies such as OpenZFS, Minio, Samba, Ceph, and HDFS are mature and empower choices that can dramatically lower storage costs.
6. Hybrid Storage as a Service
According to John Scaramuzzo, CRO of Nyriad, the three key factors that are driving sustainability and efficiency improvements are concern for the environment, supply chain issues are leading to a desire for smaller hardware unit count, and inflation.
“Hybrid Storage as a Service will see continued growth as a virtualized consumption model between cloud and on-premise storage,” he said. “The use of containers with disaggregated resources will continue to grow, blurring the line between storage and physical hardware. Block, file, and object will all be residing on common hardware.”
As supply chain issues continue, businesses will need more flexible storage solutions managed and operated by hybrid service providers, minimizing downtime.
“The need for data centers to be more efficient and sustainable will force storage platforms to become universal, servicing multiple storage types and workloads,” said Scaramuzzo. “Gone are the days of specialized appliances.”