Data Storage Technology Introduction A storage device consists of a read/write mechanism and a storage medium. The storage medium is the device or substance that actually.

Download Report

Transcript Data Storage Technology Introduction A storage device consists of a read/write mechanism and a storage medium. The storage medium is the device or substance that actually.

Data Storage Technology
Introduction
A storage device consists of a
read/write mechanism and a storage
medium.
The storage medium is the device
or substance that actually holds data.
The read/write mechanism is the
device used to read or write data to
and from the storage medium.
A typical computer system has many storage
devices
 Primary storage: Static RAM (SRAM)
Dynamic RAM (DRAM)
Flash RAM
ROM
 Secondary storage: Magnetic tape
Magnetic disk
Optical disk
Multiple devices are required because no one
device or technology can cost-effectively meet
all storage needs for a single computer system
or user.
Primary Characteristics
Speed
Volatility
Access method
Portability
Cost and capacity
Speed
Is the most important characteristic that
differentiate primary and secondary storage
With current technology, primary storage
speed is typically greater than secondary
storage speed by a factor of 105 or more.
Storage device speed is called access time.
 Is the time required to execute on read or write
operation completely.
 Access time of primary storage devices generally
are expressed in nanoseconds.
 Access time of secondary storage devices
typically are expressed in milliseconds.
A complete measure of data access speed consists
of access time and the unit of data transfer to/from
the storage device.
Data transfer units vary from one storage device to
another:
 The data transfer unit for primary storage
devices is usually a word.
 The data transfer unit for secondary storage
devices is block.
 The date transfer unit for magnetic and optical
disk drives is sector.
A storage device’s data transfer rate is computed by
dividing 1 by the access time and multiplying the
result by the unit of data transfer.
Eg. The data transfer rate for a primary storage device
with 15 nanosecond access time and a 32-bit word
data transfer unit can be computed as:
(1second / 15 nanoseconds) * 32 bits
= 266,666,667 (bytes/second)
Volatility
A storage device or medium is non-volatile if it
holds data without loss over long periods of time.
A storage device or medium is volatile if it
cannot reliably hold data for long periods.
Primary storage devices are generally volatile.
Secondary storage devices are generally nonvolatile.
Access Method
The physical structure of a storage device’s
read/write mechanism and storage medium
determines the way(s) in which data can be
accessed.
Three types of access methods:
 Serial access
 Radom access
 Parallel access
Serial Access
A serial access storage device stores and
retrieves data items in a linear or sequential
order.
Once written, data can be read back only in that
same order.
Serial access time depends on the current
position of the read/write mechanism and on the
position of the desired date item within the
storage medium.
Serial access devices are not used for frequently
accessed data but for holding backup copies of
data stored on other storage devices.
Random Access
A random access device is not restricted to any
specific order when accessing data but directly
access any desired data item stored on the
storage medium.
All primary storage devices and disk storage
devices are random access devices.
Access time may or may not be a constant.
 It is a constant for most primary storage devices.
 It is not a constant for disk storage.
Parallel Access
A parallel access device is capable of
simultaneously accessing multiple storage
locations.
If one considers the unit of data access to be a
bit, then access is parallel.
Parallel access also can be implemented by
subdividing data items and storing the
component pieces on multiple storage devices.
Portability
Data can be made portable by storing it on a
removable storage medium or device.
External disk drives are portable if they can be
added or removed easily from a computer system
and transported without damage to the device or
its data content.
Portable devices typically sacrifice access speed to
compensate for the lack of environmental control.
Cost and Capacity
Each of the
storage
device
attributes is
directly
related to
device cost
Characteristics
Cost
Speed
Volatility
Cost increases as speed increases
For devices of similar type, cost
decreases as volatility increases
Serial is the least expensive,
Random is more expensive than serial,
Parallel is more expensive than nonparallel
For devices of similar type, portability
increases cost
Cost usually increases in direct
proportion to capacity
Access method
Portability
Capacity
Primary Storage Devices
Storing Electrical Signals
Data are represented within the CPU as digital
electrical signals – the basis of data transmission
among all devices attached to the system bus
Electrical power can be stored directly by various
devices including batteries and capacitors
An electrical signal can be stored indirectly by using its
energy to alter the state of a device such as a
mechanical switch
Random Access Memory (RAM)
RAM is a generic term describing primary
storage devices with the following
characteristics:
 Microchip implementation using semiconductors
 Ability to read and write with equal speed
 Random access to stored bytes, words, or larger
data units
There are two basic RAM types
 Static RAM (SRAM)
 Dynamic RAM (DRAM)
Static RAM
Is implemented entirely with transistors
The basic unit of storage is a flip-flop circuit
A flip-flop circuit which contains two transistors for bit
storage is an electrical switch that remembers its last
position
SRAM is volatile unless a continuous supply of power
can be guaranteed.
Dynamic RAM
Uses transistors and capacitors.
The capacitors are the dynamic element
DRAM chips include circuitry that automatically
performs refresh operations
Each refresh operation is called a refresh cycle
SRAM is more expensive to fabricate
than DRAM due to its more complex
circuitry.
DRAM is slower than SRAM due to its
required refresh cycles.
Neither RAM type can match current
microprocessor clock rates which range
from 500MHz to 1.5GHz.
The fastest DRAM is at least 25 times
slower than modern microprocessors.
SRAM is at least 2.5 times slower.
A number of technologies are used
to bridge the performance gap
between memory and
microprocessors:
– Read-ahead memory access
– Synchronous read operations
– On-chip caching
Synchronous DRAM (SDRAM)
Enhanced DRAM (EDRAM)
Ferroelectric RAM
Read-Only Memory (ROM)
Is a random access memory device that
can store data permanently or
semipermanently.
Is used to stored programs such as
computer system boot subroutines like the
system BIOS.
Two ROM technologies are currently in use
are: Electrically Erasable Programmable
Read-Only Memory (EEPROM) and Flash
Memory.
EEPROM can be programmed, erased, and
reprogrammed by signals sent from an external control
source, such as a CPU.
 The primary drawback of EEPROM is that read and
erase operations require relatively high voltages
and repeated cycles to cause permanent change.
Flash Memory can be erased and rewritten much more
quickly and is used for programs and data that aren’t
frequently updated.
 It tends to wear out after 100,000 or more write
operations, making it unsuitable for primary
storage.
Memory Packaging
Memory circuits are embedded within microchips and groups of
chips are packed on a small circuit board that can be installed
or removed easily from a computer system.
Early RAM and ROM circuits were packaged in dual in-line
packaged (DIPs)
Later, single in-line memory module (SIMM) is adopted
which incorporated multiple DIPs on a tiny printed circuit board.
The double in-line memory module (DIMM) is a newer
packaging standard. It is a double-sided SIMM with memory
DIPs and electrical contacts on both sides of the module.
Dual in-line package (DIP)
Single in-line memory module (SIMM)
CPU Memory Access
The physical organization of memory,
The organization of programs and data within memory,
And the method(s) of referencing specific memory
locations
are critical design issues for both primary and secondary
storage devices and processors.
Physical Memory Organization
The main memory can be regarded as a sequence of
contiguous or adjacent memory cells.
Cell 0
Cell 1
Cell 2
Cell 3
Cell 4
In a byte sequence, the leftmost byte is called the
most significant byte and the rightmost byte is called
the least significant byte.
Big endian describes architectures that store the most
significant byte at the lowest memory address.
Little endian describes architecture that store the least
significant byte at the lowest memory address.
The addressable memory of a CPU is the highest
numbered storage byte that can be represented.
– Addressable memory is determined by the number of
bits used to represent an address.
– Physical memory is the actual number of memory
bytes that physically are installed in the machine.
– Physical memory is usually less than addressable
memory.
Memory Allocation and
Addressing
Memory allocation
describes the
assignment of
specific memory
address to system
software, application
programs, and data.
Unused
High memory
Program data
Program code
Program
offset
FFFFH
Operating
system
3000H
0000H
Low memory
Absolute addressing describes memory address
operands that refer to actual physical memory locations.
Usually, the operating system calculates and stores the
program offset in a register when the program first is
loaded into memory.
During program execution, the CPU automatically adds
the program offset to all memory address operands
before accessing memory.
The method of automatically computing physical
memory addresses is called indirect addressing.
The register that holds the offset value is called offset
register.
Magnetic Storage
A magnetic storage device
– converts electrical signals into magnetic charges,
– captures the magnetic charge on a storage medium,
– and later uses the stored magnetic charge to
generate an electrical current.
Magnetic storage devices must control or
compensate for some undesirable
characteristics of magnetism and magnetic
storage media
Magnetic Decay and Leakage
Magnetic decay:
Is the tendency of magnetically charged particles to lose
their charge over time.
Is constant over time and proportional to the power of the
charge.
Will cause the stored charge power to fall below the
threshold required for a successful read operation
 the data content of the storage medium effectively is lost.
Magnetic leakage:
The strength of individual bit charges also can decrease
due to magnetic leakage from adjacent bits.
Storage Density
Coercivity is the ability of a substance or magnetic
storage medium to accept and hold magnetic charge. It
varies widely among elements and compounds.
The amount of surface area allocated to a bit is referred
to as the recording density, which is expressed in bits,
bytes, or tracks per inch.
TIP: Designers and purchasers of magnetic media and
devices must find a suitable balance between high
recording density and the reliability of the media,
especially over relatively long periods of time.
Media Integrity
The integrity of magnetic storage medium depends on
the nature of its construction and the environmental
factors to which it is subjected.
Loss of coercible coating represent a loss of strength in
stored magnetic charges  data becomes unreadable
when the remaining charge falls below the threshold of
readability.
TIP: To extent the life of magnetic media, they must be
protected from physical abuse and temperature and
humidity extremes
Magnetic Tape
Is a ribbon of plastic with a
coercible surface coating.
Tapes are mounted in a
tape drive for reading and
writing.
Tape drive contains motors
that wind and unwind the
tape.
Tapes primarily are used to make back up copies of
data stored on faster secondary storage devices and
to physically transport large data sets.
There are two basic approaches to record
data onto a tape surface:
– Linear recording places bits along parallel
tracks that run along the entire length of the tape.
– Helical scanning reads and writes data to or
from a tape by rotating the read/write head at an
angle to the tape and moving from tape edge to
tape edge.
Magnetic disk
Magnetic disk media are flat, circular platters with metallic
coating that are rotated beneath read/write heads.
Data normally are recorded on both sides of a platter.
A track is one concentric circle of a platter.
A sector is a fractional portion of a track.
A read/write head is placed for each side of each platter.
Read/write heads are mounted on the end of an access
arm.
Access arms are attached to a positioning servo.
The servo moves the access arms so that the read/write
heads can be positioned anywhere between the outermost
track and the innermost track.
Hard disk:
Is a magnetic disk media with a rigid metal base.
Typical platter size is between three and five inches
in diameter.
High speeds is up to 10,000 revolutions per minute.
Drive capacity depends on the number of platters,
platter size, and recording density.
Multiple hard drives often are enclosed in a single
storage cabinet and referred to as a drive array.
Floppy disk (diskette):
Uses a base of flexible or rigid plastic material.
Rang from 2.5 to 3.5 inches in diameter.
Contains a single flexible plastic platter coated with an
iron or other metallic compound on both sides.
The platter is mounted permanently in a plastic case
that can be removed from a diskette drive.
The case has an access door that is closed when the
diskette is removed from the drive and opened
automatically when the diskette is inserted into the
derive.
Disk access time depends on several factors including:
Time required to switch among read/write heads
Time required to position the read/write heads
Rotational delay – the time that the disk
controller must wait for the proper sector to rotate
beneath the heads
Optical Mass Storage Devices
Came of age in the 1990s
Primary advantages: higher recording density and
longer data life
Can retain data for decades
Not subject to problems of magnetic decay and
leakage
Popular due to standardized and relatively
inexpensive storage media.
Store bit values as variations in light reflection
Optical storage devices are available
currently in a wide variety of storage
formats and write technologies including:
–
–
–
–
–
CD-ROM
WORM (CD-R)
Magneto-optical
CD-RW
DVD
Guideline for the proper care of
Compact Discs
Do’s
– Store the disc in a jewel box when not in use
– Hold a disc by its edges
Don’t
–
–
–
–
Expose the disc to excessive heat or sunlight
Eat, smoke, or drink near a disc
Stack discs
Touch the underside of the disc