DEC DS-HSZ22-AA, User Manual

Page: 36 / 94

RAID Array 3000 Pedestal Enclosure
2–8 EK–SMCPO–UG. A01
•
Each RAID set can be partitioned into smaller redundancy groups.
•
The controller’s host LUN Mapping feature makes it possible to map RAID
sets differently to each host port. You make the same redundancy group ap-
pear on different LUNs to different hosts, or make a redundancy group visi-
ble to one host but not to another.
•
Any drive may be designated as a hot or warm spare. Spares are global,
meaning that in the event of a drive failure, the controller will search for the
first available spare on any channel or SCSI ID and automatically begin re-
building the failed drive’s data.
2.5 Performance Enhancements
The controller employs a number of techniques to achieve as much performance
as possible from its design.
2.5.1 Custom Components
To increase performance and reliability, the controller’s core functions have been
encapsulated in four custom ASIC (Application Specific Integrated Circuits)
components as follows:
XOR ASIC: Used in the Exclusive -Or parity calculations employed by RAID
levels 4 and 5.
DMA ASIC: Controls the data path hardware for the various I/O ports
CPU Interface ASIC: Supports the controller’s MIPS R3000 RISC central
processing unit.
Memory Controller ASIC: Controls the memory system and supports data
movement on the internal bus at a maximum burst rate of 80 MB/second and a
maximum sustainable rate of 60 MB/second.
2.5.2 Efficient Write and Read Algorithms
Standard RAID write operations that involve parity, such as those in RAID levels
4 and 5, require multiple, time-consuming steps:
1. Read data from the parity drive.
2. Read existing data from the target data drives.
3. Exclusive-Or the old parity, old data, and new data to generate new parity
data.
4. Write the new parity data to the parity drive.
5. Write the new data to the target data drives.