What is Data Tiering?
Posted in tech
While the concept of storage tiering is not new, the DataSphere metadata engine takes tiering to a whole new level. DataSphere enables IT to view storage capabilities as services that can be put to work specifically for the requirements of data, then tiering data to the right resource to meet business objectives across performance, price, activity level, and more. Let’s take a closer look at how.
Tier Data Across Any Storage System, Including the Cloud
Many vendors offer storage systems that can tier data between different media within a single storage silo, such as flash or disk. Some vendors even offer the ability to tier data across different storage systems within their own product ecosystem. DataSphere’s vendor- and protocol-agnostic architecture enables data to tier across storage of any type, from any vendor, from flash to shared to cloud and object storage. This gives enterprises much more choice when purchasing storage systems, enabling them to select storage based on the capabilities their applications require versus what a single vendor has to offer. DataSphere can even tier data across on premises object stores and single or multiple cloud stores.
DataSphere automates the flow of data to the right storage resource for changing application and business demands.
Scale Any Tier Linearly, On Demand
Most tiered storage systems require enterprises to purchase capacity in bundles that include cache, flash, and different tiers of disks. As data grows, one or more tiers typically becomes increasingly underutilized, making additional investments less cost effective. Also, these systems typically can’t scale beyond certain workloads and capacity limitations before performance begins to degrade. With DataSphere, storage of different types can be simultaneously available to applications through a global namespace. Enterprises can add more capacity of any type, on demand. The DataSphere architecture is immensely scalable, capable of supporting petabyte capacities and the management of billions of files, without degradation.
Tiering is Intelligent, Non-Disruptive and Application-Aware
Conventional tiering is blind beyond the storage container, meaning the tiering software has no awareness about how data’s placement on a tier is affecting end application performance, which is what businesses really care about. DataSphere has visibility into data on the client so it can make an informed decision about which tier will best meet business objectives, including performance and protection requirements. DataSphere uses these objectives to intelligently align data on the ideal storage tier in its global namespace.
This ability to manage data by business objectives transforms data management, since storage capabilities now become a service that data consumes as required. DataSphere’s unique architecture allows for data movement to occur while files are open and data is actively being accessed. DSX Data Movers ensure all reads and writes to data are completed atomically even when the data is in flight to the target tier. See the Managing by Objectives with DataSphere white paper for detailed information about using objectives to automatically align data to the right storage resource.
Put Storage to Work for Your Data
DataSphere raises the bar for tiering. It enables enterprises to deploy exactly the storage they need, from any vendor, and to scale any tier, on demand. This reduces costs, increases flexibility, and improves service levels. DataSphere makes tiering application-aware, through policy-based, automated data mobility that puts storage capabilities to work in the service of data. This closes the gap between application requirements and storage capabilities, enabling IT to directly deliver critical infrastructure services to the business.