M.

Tech(CSE)-Advanced Computer Architecture-CSE507

Unit-III: Memory Organization

3.0. Introduction: The goal of the memory system is to provide the programmer with a
large storage capacity for programs and data and to provide this information on demand
to the CPU in a timely fashion. The limiting factor in determining how fast the CPU can
execute instructions is often the speed with which it can fetch instructions and operands
from memory. Techniques can be employed to give the illusion of having an essentially
unlimited fast memory even when such a memory may not actually exist within the
computer system.
There are two types of memories that communicate directly with the CPU: read/write
random-access memory (RAM) and read-only memory (ROM). Read-only memory is a
programmable logic device in which the stored information is embedded in the solid-state
circuitry either during its manufacture or by being programmed in later via electrical
impulses. Once the information has been stored in ROM, it can be read but not altered.
For both RAM and ROM the time required to access a specific word is independent of its
location. This is in contrast to a sequential memory such as magnetic tape in which the
access time to a particular data item depends on its relative location on the tape. Reading
the nth word requires that the previous n-1 words be read first.
A memory unit stores binary information in groups of bits called words. Most computer
memories use words that are multiples of 8 bits. A block
n data diagram
input lines of the memory unit is
shown below: b address
Memory unit

2b words
Read n bits per word

Write

n data output lines

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09
 M.Tech(CSE)-Advanced Computer Architecture-CSE507

3.1.Memory units
The capacity of a memory unit is usually stated as the total numbers of bytes that
it can store.

Term Normal Usage Usage as Power of 2

K ( Kilo) 103 210 = 1,024
M (Mega) 10 6 220 = 1,048,576
G (Giga) 109 230 = 1,073,741,824
T (Tera) 1012 240 = 1,099,511,627,776

3.2.Key Characteristics of memories:

i. Location
• CPU
• Internal(main)
• External (secondary)
ii. Capacity
• Word size
• Number of words
iii. Unit of Transfer
• Word
• Block
iv. Access methods
• Sequential access
• Direct access
• Random access
• Associative access
v. Performance
• Access time

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09
 M.Tech(CSE)-Advanced Computer Architecture-CSE507

• Cycle time
• Transfer rate
vi. Physical type
• Semiconductor
• Magnetic surface
• Optical
vii. Physical characteristics
• Volatile/non volatile
• Erasable/non erasable
i. Location: With most visible aspect of memory, the memories are located in three areas
of the computer. One is in the form of CPU registers, which are used by CPU as its local
memory.
The other two are internal and external to the computer. The internal memory is often
equated with main memory.
The external memory consists of peripheral device, such as disk and tape that are
accessible to CPU via I/O controllers.
ii. Capacity: The capacity of the internal memory is typically expressed in terms of bytes
or words. External memory capacity is expressed in terms of bytes. This capacity also
depends on number of word in memory.
Total capacity of internal memory can be calculated as follows
Total memory = Number of words X word length
Number of words = 2 address bus width
Word length = Data bus width
iii. Unit of Transfer: For the internal memory, the unit of transfer is equal to the number
of data lines into and out of the main memory module. This is equal to the one word
length, but it may not be. The common word lengths are 8,16,1nd 32 bits.
For external memory, the data often transferred in much longer units than a word, and
these are referred to as Blocks.
iv. Access Methods: The method of accessing data can be categorized into four types
• Sequential Access: Memory is organized into units of data, called Records.
Access must be made in a specific linear sequence Stored addressing

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09
 M.Tech(CSE)-Advanced Computer Architecture-CSE507

information is used to separate records and assist in the retrieval process shared
read/ write mechanism is used , and this must be moved its current location to
the desired location , passing and rejecting each intermediate record .Thuds the
time to access an arbitrary record is highly variable.
• Direct access: As with sequential access, direct access involved a shared read
write mechanism. However, individual blocks or records have a unique address
based on physical location. Access is accomplished by general access to reach a
general vicinity plus sequential searching, counting, waiting to reach the final
location. Again access time is variable.
• Random Access: Each addressable location in the memory has unique, physically
wired- in addressing mechanism .The time to access a given location is
independent of the sequences of prior access and is constant. Thus any location
can be selected at random and directly addressed and accessed .The main
memory systems are a random access.
• Associate Access : This is Random access type that enables one to make a
comparison of desired bit locations within a word for specific match and to do
this for all words simultaneously .Thus a word is retrieved based on a portion of
its contents rather than its address .As with ordinary random access memory
each location has its own addressing mechanism and retrieval time is constant
independent of location or prior access patterns.
Cache memories
V .Performance : There are three parameters are used to measure the performance of the
computer.
• Access time :The time required to read/write the data from /into desired record is
called Access time. The access time is depends on amount of data to be read/write
in the desired record .If the amount data is uniform for all records then the access
time is same for all records.
• Memory Cycle time : Is the combination access time plus time required to locate
desired record.
For Random access method ,this memory cycle time is same for all records
,because the time to access a given record is independent of the location. where

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09
 M.Tech(CSE)-Advanced Computer Architecture-CSE507

as for the sequential access and direct access ,the memory cycle time is different
because the time to locate a given record is depend of the position of desired
record with respect to first record and access time same for all records..
• Transfer rate : The rate at which the data can be transferred into or out of a
memory unit.
For random access memory it is equal to 1/cycle time
For non random access memory the following relationship holds:
Tn =Ta +(N/R )
Where
Tn = Average time to read or write N bits
Ta =Average access time
N = Number of bits
R = Transfer rate ,in bits per second(BPS)
vi. Physical type of memory :The two most common memories are semiconductor
memory and magnetic surface memory. and third one is optical memory this is so
popular .
vii. Physical Characteristics : In a volatile memory ,information decays naturally or
lost when electrical power is switched off. In no volatile memory ,the information once
record remains without deterioration until deliberately changed; no electrical power is
needed to retain information
3.3. Memory Hierarchy: Processors use memory of various types to store the data on
which a program operates. Data may start in a file on a magnetic disc ("hard disc"), be
read into semiconductor memory ("D-RAM") for processing, transformed and written
back to disc. As part of the transformation process, individual words of data will be
transferred to the processor's registers and thence to the ALU. Transfers from
semiconductor memory to registers will usually pass (transparently to a programmer)
through one or more levels of cache.

Memory designers are able to trade speed for capacity - the fastest memory (registers)
having access times below 10ns but the lowest capacity (10s of words) and the slowest
(magnetic tape) having access times of several seconds but the highest capacity (10s of G

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09
 M.Tech(CSE)-Advanced Computer Architecture-CSE507

Bytes). Thus the memory in a system can usually be arranged in a hierarchy from the
slowest (and highest capacity) to the fastest (and lowest capacity).

CPU Registers Capacity access time cost /bit

Cache memory Increasing Decreasing

Main Memory

Auxiliary memory

M.Rajasekhara Babu/Faculty/School of Computing Sciences/VIT University/Velloe-14

http://sites.google.com/site/mrajasekharababu/mtech09