Table of ContentsView in Frames

How the AFF platform differs from the FAS platform

AFF platforms differ from FAS platforms in a few significant ways. You should understand these differences before using your AFF cluster.

The following table describes the differences between AFF and FAS systems:

Difference Description
SSD drives Only SSD drives are supported on AFF platforms. If non-SSD drives are attached to an AFF platform, the disks do not go online. If the root aggregate is HDD-based, the nodes do not boot. HDDs and FlexArray are not supported.
Performance Using an AFF80xx provides maximum I/O performance. For example, AFF systems running Data ONTAP 8.3.1 have experienced 200 to 300 microsecond reductions in FCP latency for read-intensive workloads already at or around 1 microsecond in overall response times.
Volumes All volumes created in aggregates owned by an AFF node have inline compression and inline deduplication of zeroes enabled by default.
SVMs The storage efficiency polices for SVMs differ from those on a FAS system.
Root-data partitioning The root-data partitioning capability is enabled by default on an AFF node.
Mixed platform support You cannot pair an AFF node with a regular FAS node in an HA pair. You can mix AFF HA pairs with FAS HA pairs in a cluster. For more information, see the Clustered Data ONTAP Storage Platform Mixing Rules.
Zero block elimination Instead of writing incoming 4 KB blocks that contain all zeroes to storage media, AFF80xx systems indicate the block's inode as all zeroes. With deduplication enabled, zero block elimination is enabled by volume by default, with no user configuration required.
FC SAN Optimization An FC SAN optimized AFF configuration is available. The FC SAN optimized cluster enables you to quickly complete a configuration that is ready for demonstration and testing.