Home > Storage > PowerScale (Isilon) > Product Documentation > Protocols > PowerScale: Home Directory Storage Solutions for NFS and SMB Environments > Planning for disk-pool performance and management
This section provides an overview of SmartPools node pools and file pool policies with respect to planning for optimal storage performance and data protection. It is not intended to be the definitive source of best practices information on SmartPools.
Node pools are groups of disk pools, spread across similar PowerScale storage nodes. Each node pool only contains disk pools from the same type of storage nodes and a disk pool may belong to exactly one node pool. For example, F series node with 1.6 TB SSD drives would be in one node pool, whereas A series nodes with 10 TB SATA drives would be in another.
Multiple groups of different node types can work together in a single, heterogeneous cluster. Each node type is optimized for a different capacity-to-performance ratio, so it’s common for organizations to leverage all these architectures into one cluster in order to ensure an optimum match between their different data sets and the storage nodes that they reside on. For example, on node pool of F series for I/Ops-intensive applications, one node pool of H series nodes, primarily used for high-concurrent and sequential workloads, and one node pool of A series nodes, primarily used for nearline and/or deep archive workloads.
File pool policies control three principal attributes of files: node pools/tiers where files reside, I/O optimization settings, and the protection level of files. The placement of a particular file or directory is automated by the PowerScale storage cluster, based on a series of standard and customized file pool policies, and occurs entirely transparently to users and clients within the logical filesystem hierarchy. Moving a file or directory from one node pool to another does not alter the file’s location within the file-system tree, nor does it require clients to be reconfigured to access the file in a new location.
File-pool policies are created and applied at either the directory or file level on a PowerScale storage cluster based, on one or more attributes of the data. Attributes that can be used to determine inclusion or exclusion by a file-pool policy may include:
Storage administrators and business units can leverage these attributes to ensure that a particular data set is optimized for both capacity and performance. In creating a file-pool policy based on one or more of the above criteria, the following actions can be applied by the policy:
Data-access performance and protection can be adjusted and optimized any or all of these available policy-based settings. High performance data, as identified by one of the above attributes, can be placed on a pool of more powerful nodes, or its metadata moved to SSD (also known as ‘metadata acceleration’, or both, to increase throughput rates and decrease latency levels.
Alternately, older data, as identified by last-modified or last-accessed date, can be moved to a pool of archive nodes to free up capacity on the more powerful pools for newer, higher-performance data.
Note: The use of metadata acceleration in conjunction with one or more SmartPools file policies requires that the node(s) on which the targeted data is stored be SSD equipped.