beautypg.com

Technology overview, Built-in virtualization, Block-level data management – Dell Compellent Series 40 User Manual

Page 6: Figure 1, Fluid data architecture

background image

Cost and Performance benefits of Dell™ Compellent™ Automated Tiered Storage for Oracle® OLAP
Workloads

6

Technology Overview

Dell Compellent Fluid Data Architecture – feature description

Dell Compellent created an enterprise storage solution that automates the movement and management
of data throughout its life cycle with Fluid Data Architecture. Fluid Data Architecture enables true
storage virtualization, providing a single pool of storage resources that spans across all disks and RAID
levels. It also enables a dynamic block architecture that actively and intelligently manages data at
more granular level. With the automated storage tiering, data is placed intelligently and efficiently
among the multiple tiers and RAID levels.

The following features describe Fluid Data Architecture from Dell Compellent.

Built-in virtualization

A virtualized storage environment provides an ideal foundation for tiered storage. With a truly
virtualized environment, there are no constraints on where data can reside, as data is not confined to
particular disk group. You can move data without limitation to a tier with a particular drive type or
performance level, or to a particular RAID level within a storage tier, depending on usage patterns.

Fluid Data Architecture

Figure 1.

Block-level data management

Fluid Data Architecture enables true storage virtualization and facilitates a highly granular approach to
storage tiering. Dell Compellent divides data into pages, which are preformatted collections of
allocated and unallocated disk blocks. The pages are 2 MB by default, though administrators have the
option to manage storage as 512 KB or 4 MB pages, and details about each block are captured in action.
This granular approach optimizes tiering. Using real-time system intelligence, the Dell Compellent SAN
identifies very small data blocks that are eligible for movement based on access frequency and then
moves just the small pages that include those blocks to the appropriate tier or RAID level, creating a
finely tuned tiered storage environment. Using small pages also increases the efficiency of data
movement since it is more efficient to read, write, and migrates smaller pages rather than larger ones.
Static storage architectures are limited to much larger blocks of data, ranging from 16 MB to 1 GB.