Enterprises are struggling with an ever-increasing amount of data to store and manage. Across all industries, rapid data growth is outstripping the capacity of legacy storage systems and stretching the resources of understaffed IT organizations. A clustered, scalable storage area network (SAN) enables storage professionals to respond to data growth while preserving existing infrastructure investments and simplifying management. The idea is straightforward: With a scalable SAN, you can add nodes incrementally and still manage the entire storage environment as a single entity, as opposed to a traditional scale-up approach in which you grow a storage system to its limit and then add separate systems to increase capacity and performance.
Limits of a Scale-Up Design
All storage controllers have physical limits to their expandability. The number of CPUs and memory slots and the amount of space for disk shelves dictate the maximum capacity and performance of a single controller. If you need more storage or performance capacity, you can add CPUs and memory or install additional disk shelves, but ultimately, the storage controller will max out. When this happens, the only option is to buy another controller and scale up.
Analysts at Wikibon.org estimate that the cost of data storage and management at most organizations is 100% to 200% higher than needed due to the inability to dynamically balance workloads across storage systems, and the situation gets worse as the number of controllers
Benefits of a Scalable SAN
A scalable SAN enables you to seamlessly add new controllers as your storage environment grows — while maintaining access to a shared, back-end resource pool. Host and client connections, as well as data stores, can be moved anywhere within the shared pool without disrupting operations. As a result, existing workloads can be balanced across the available resources and new workloads can be optimized for performance. Technology refreshes, such as replacing disk shelves and adding or replacing storage controllers, are accomplished while the environment remains online and serving data.
Scalable SANs are designed specifically for enterprise applications that can benefit from nondisruptive operations and seamless scaling, such as Oracle, SAP, Microsoft Exchange, SQL Server and SharePoint. Perhaps the greatest benefit of a scalable SAN is that it allows IT organizations to begin with a small number of nodes and increase resources as needed. In addition, a clustered, scalable SAN addresses some of the most difficult challenges of your storage environment, including:
- Capacity: Support for more than 12 petabytes per cluster
- Performance: Flash acceleration to support hundreds of thousands of IOPS
- Efficiency: Deduplication, compression and thin provisioning
- Multi-tenancy: Logical storage containers for secure isolation of tenants
- Connectivity: Support for Fibre Channel and Fibre Channel over Ethernet, NPIV for virtualized fabrics, and over 150 I/O connections
- Availability: Intra- and inter-cluster replication plus nondisruptive operations for maintenance and upgrades
This blog is part of a series on clustered Data ONTAP.
Clustered ONTAP blog series: