The End of Tier 1 Storage as We Know It?
Flash-based storage in its many forms is poised and positioned to cause a major disruption to what is currently viewed as tier 1 storage. As we move into an era in which business agility requires greater application performance, we will see application demands continuing to pull flash up the storage stack and into the server.
As flash becomes more and more integrated — and critical — in the storage stack, we will see the end of tier 1 storage as we know it. Instead, we will see a new paradigm that is basically a two-tier storage infrastructure that breaks down between storage optimized for performance and storage optimized for capacity.
In the near future, users can expect enterprise storage to look more like this model, with two layers that are seamlessly integrated:
- Performance (I/O operations per second [IOPS]) layer. This is what is emerging in the storage stack and storing an increasingly large set of hot data, often very close to the application itself running on the server.
- Capacity layer. This layer will complement the performance layer and seamlessly integrate with it. The capacity layer will directly address the economics of storing, protecting, and managing the orders-of-magnitude increases of capacity fueled by big data, while utilizing storage technologies that are more appropriate to maximizing this data based on its value to the organization.
This changing paradigm — performance versus value — became obvious to me when I served as vice chair of the Storage Networking Industry Association’s Solid-State Storage Initiative almost four years ago. As the industry rapidly evolved, we could see dynamic random-access memory (DRAM) storage arrays quickly giving way to NAND Flash-based successors. Our cross-vendor team of storage specialists realized that a major disruption was about to take place and that it would last for at least a decade. The result: Two distinct camps emerged, along lines similar to the layers outlined earlier:
- Performance camp. This group is focused on consistently low-microsecond response times and multiple gigabytes per second of throughput.
- Value camp. This group offers a rich software layer on top of NAND flash and sacrifices ultimate performance in favor of very high performance, combined with advanced storage efficiency and data management.
Future-Ready Approach to Tiering
At NetApp, we created the virtual storage tier architecture as a result of our unique position in the middle ground between these two camps. This virtual tier of storage extends to server hosts, is updated in real time, and is self-managing. It is optimized for performance and efficiency, has a unified SAN and NAS architecture, and reduces the effective cost of NAND flash-accelerated storage using deduplicated and thinly cloned blocks that deliver high performance. As a result, NetApp customers are utilizing flash at some of the highest rates in the industry.
But the flash disruption doesn’t stop at the array, as we recognized long ago. Application performance demands will continue to pull flash up the storage stack and into the server. Over time, even storage semantics themselves will give way to persistent (non-volatile) memory semantics, enabling simpler and faster high-performance, real-time applications.
In the meantime, big data capacity trends and requirements for enterprise-level data management, availability, protection, and efficiency will further entrench non flash spinning disk media into a scalable capacity layer. This layer will serve as the foundation of this new storage stack, one that safeguards and manages an organization’s “single source of truth”: the data required to run the business.
When the flash disruption is complete over the next few years, the once-lucrative tier 1 frame array market might simply disintegrate. Although ESCON (Enterprise Systems Connection)/FICON (Fiber Connectivity)-attached arrays will continue to exist for the important — but shrinking — mainframe market, fault-tolerant data persistence functionality will move away from tier 1 frame arrays and up the stack, all the way to the application layer. Performance capability will move to the server/host layer, leaving data management, protection, and efficiency to a shared storage architecture. Expect a capacity-optimized layer to dominate here, leaving precious little room for archaic tier 1 frames.
In the future, users shouldn’t expect arrays of spinning disks to be purchased primarily based on performance characteristics and shouldn’t expect high-performance solid-state storage (flash or otherwise) to provide cost-effective, end-to-end data management and protection capabilities. Being pulled apart from both the performance and capacity ends, the all-encompassing tier 1 storage frame array is rapidly approaching the end of its run.
In contrast, unified and extensible storage architectures such as the NetApp® virtual storage tier provide a framework for integrating new flash-based media in a manner that delivers scalable performance and capacity in anticipation of a continued evolution toward a two-tier storage infrastructure.
Val Bercovici, Big Data and Cloud Czar at NetApp
Val Bercovici joined NetApp in 1998 and leads the Strategic Planning Team within the office of the CTO. Working with customers, analysts and alliance partners, Val focuses on next-generation research projects and is responsible for NetApp’s product vision.
NetApp, the NetApp logo, and Go further, faster are trademarks or registered trademarks of NetApp, Inc. in the United States and/or other countries.