HP A7143A RAID160 SA Controller Support Guide, February 2007

RAID160 SA Controller Overview
Board Components and Features
Chapter 2
27
IMPORTANT The batteries on a new RAID160 SA controller may have a low charge when the controller
board is first installed. No action is required on your part, since the internal circuitry
automatically recharges the batteries and enables the cache. The recharge process takes less
than four hours. The controller will function properly during this time, but without the
performance advantage of the array accelerator. When the batteries are charged to an
acceptable capacity, the array accelerator is automatically enabled.
PCI System Interface
RAID160 SA controller interfaces with the system through a high-performance 64-bit PCI bus that:
• Runs at 66 MHz
• Provides a high-speed path (up to 528 MB/s) between the system board and the controller
• Includes two parity protection signals
The RAID160 SA controller is a PCI Bus Master device conforming to Rev. 2.2 of the PCI Local Bus
Specification. As a bus master device, it takes control of the PCI bus during high-speed transfers, freeing the
system processor to handle application processing or other types of tasks.
For maximum performance, HP recommends that you use only 66-MHz devices on any given 66-MHz PCI
bus. Combining 66-MHz and 33-MHz devices on a PCI bus will decrease the overall bandwidth to 33-MHz
speeds.
SCSI Support
The RAID160 SA controller supports disk drives that conform to Wide Ultra160 and Wide Ultra2 standards.
Although Wide Ultra2 devices operate at a different maximum speed from Wide Ultra160 devices, operating
speeds are unaffected if they are connected to the same SCSI bus because they both use low voltage
differential (LVD) signaling.
Fault Management Features
The RAID160 SA controller and the HP-UX operating system support several fault management and data
reliability features that minimize the impact of disk drive defects on your system.
• Auto-Reliability Monitoring (ARM) is a firmware process that operates in the background scanning
physical disks for bad sectors in fault-tolerant logical drives. ARM also verifies the consistency of parity
data in logical drives that are using RAID 5 or RAID ADG. This process assures that you can recover all
data successfully if a disk failure occurs in the future. ARM operates only when you select a fault-tolerant
configuration (RAID 1 or higher).
• Dynamic sector repair by the RAID160 SA controller automatically remaps any sectors that have
media faults (detected either during normal operation or by auto reliability monitoring).
• S.M.A.R.T. is an industry-standard diagnostic and failure-prediction feature of physical disks, developed
by HP in collaboration with the disk drive industry. It monitors several factors that can be used to predict
imminent physical disk failure due to mechanical causes. Such factors include the condition of the
read/write head, the seek error rate, and the spin-up time. When a threshold value is exceeded for one of
these factors, the disk sends an alert that failure is imminent. Thus, the user can back up data and
replace the disk drive before failure occurs.