1.4 Chapter Summary
Using the SNIA Shared Storage Model
The SNIA Shared Storage Model provides a framework for understanding the relationship between applications and underlying storage resources.
At the high level, the SSM separates applications from the supporting storage domain.
The storage domain includes the file/record subsystem, block aggregation layer, block subsystem, and auxiliary services subsystem.
The file/record subsystem defines whether data is passed to the upper-layer application as files or as records.
Database applications typically use records.
Files and records ultimately reside on storage disks as data blocks.
A data block is a contiguous group of data bytes stored on disk.
Data blocks are grouped into files or records at the block aggregation layer.
Block aggregation can be performed on the host, within the SAN, or at the storage device.
The block subsystem is composed of physical disk or tape devices that store data blocks.
The services subsystem includes discovery, monitoring, management, security, backup, availability, capacity planning, and other storage-specific functions that enhance the operation of the storage domain.
A SAN resides between the file/record subsystem and the block subsystem.
SANs enable connection between multiple servers and storage resources in a peer-to-peer network.
SANs serve data blocks to the host; NAS serves files to the host.
The SAN infrastructure may be Fibre Channel, IP SAN, or some combination.
A NAS device may span the file/record subsystem to the lower-level block subsystem.
The Shared Storage Model can be used to overlay customer applications and storage and to facilitate planning for shared storage alternatives.
The SSM is a convenient tool for both data center and remote site SAN design.