Reliability - Compaq BL10e - HP ProLiant - 512 MB RAM Technology Overview

Table 2. Strategies to improve single disk capacity, performance, and reliability
Mechanical Magnetic
Increase platter
rotation rate
Increase areal
density of data
Reduce platter size
Decrease seek times
Several mechanical design strategies are used to reduce the physical distance that the read/write
heads must travel to reach the target segment:
 smaller diameter platters
 multiple platters per drive
 increased speed of platter rotation
 increased areal density of data
 decreased seek time per track
Magnetic storage strategies are used to increase the amount of data in each track. Increasing the bit
density per unit length of track also helps increase the rate at which data can be written to and read
from the drive.
Disk I/O strategies are used to reduce the time that a logical read/write spends waiting for the
physical read/write operation. These strategies seek to effectively decouple the logical and physical
operations of the disk. Increasingly sophisticated approaches become practical as the embedded
processing power and memory incorporated into the drive increases. There are several approaches
that can be used:
 buffering the data to be written to disk (write caching)
 queuing read operations
 read-ahead caching
 queuing write operations
 write caching

Reliability

Disk drive reliability is measured in terms of Annual Failure Rates (AFR). The AFR is the percentage of
disk drive failures occurring in a large population of drives in operation for one year. For example, a
population of 100,000 drives with an AFR of 1.5% would experience approximately 1,500 failures
per year. An AFR calculated from a small number of drives is subject to large statistical variations that
render it inaccurate.
Major factors in determining drive reliability are the duty cycle and the I/O workload to which the
drives are subject. Duty cycle is simply power-on time, which is calculated as the number of hours that
the disk drive is powered on, divided by the number of calendar hours. I/O workload is disk working
time, which is calculated as the number of hours that the disk drive is aggressively reading and
writing data, divided by the number of calendar hours.
Disk I/O
Increase bit density per
unit of track
Write cache buffer data to be written to disk
Queue read operations
Queue write operations
Reorder read and write operations to execute the next
operation physically available on drive
5