+

Chapter 5 Internal Memory

William Stallings, Computer Organization and Architecture, 9 th Edition
+
Objectives

How are main memory structured?

Whether main memory may cause errors?
How many types of memory?
After studying this chapter, you should be able to:
 Present an overview of the principle types of
semiconductor main memory.
 Understand the operation of a basic code that can detect
and correct singlebit errors in 8-bit words.
 Summarize the properties of contemporary advanced
DRAM organizations.
+
Contents

 5.1 Semiconductor Main Memory

 5.2 Error Correction
 5.3 Advanced Dram Organization

Semiconductor- Chất bán dẫn  (silic, germanium) là vật liệu

trung gian giữa chất dẫn điện và chất cách điện. Chất bán dẫn chỉ
hoạt động như một chất dẫn điện ở một điều kiện nào đó. Chất
bán dẫn được dùng để tạo ra các transistor (transfer-resistor).
+
5.1- Semiconductor Main Memory

 Organization

 Semiconductor Memory Types

 Dynamic RAM and Static RAM
 Types of ROM
 Chip Logic
 Chip Packaging
 Module Organization
 Interleaved Memory
+
Organization

 Basic element of a semiconductor memory is the

memory cell.
 Cell properties:
 1-They exhibit two stable (or semistable) states, which
can be used to represent binary 1 and 0.
 2- They are capable of being written into (at least once),
to set the state.
 3- They are capable of being read to sense the state
Semiconductor Memory Types

All of the memory types that we will explore in this chapter are
random access. That is, individual words of memory are directly
accessed through wired-in addressing logic.
+
Dynamic RAM (DRAM)

 RAM technology is divided into two technologies:

 Dynamic RAM (DRAM)
 Static RAM (SRAM)
How Dram cell works? Read by yourself.
 DRAM

 Made with cells that store data as charge on capacitors (tụ điện)

 Presence or absence of charge in a capacitor is interpreted as a binary 1 or

0

 Requires periodic charge refreshing to maintain data storage

 The term dynamic refers to tendency of the stored charge to leak away,
even with power continuously applied
+ Static RAM
(SRAM)
 Digital device that uses the same logic
elements used in the processor

 Binary values are stored using

traditional flip-flop logic gate
configurations

 Will hold its data as long as power is

supplied to it
 SRAM versus DRAM
SRAM
 Both volatile: Power must be continuously supplied to the
memory to preserve the bit values

 Dynamic cell
 Simpler to build, smaller
 (smaller cells = more cells per unit area) DRAM
 Less expensive

 Requires the supporting refresh circuitry

 Tend to be favored for large memory requirements

+ Used for main memory

 Static
 Faster
 Used for cache memory (both on and off chip)
+ Read Only Memory (ROM)
 Contains
a permanent pattern of data that cannot be
changed or added to
 No
power source is required to maintain the bit values in
memory
 Dataor program is permanently in main memory and
never needs to be loaded from a secondary storage device
 Data is actually wired into the chip as part of the
fabrication process
 Disadvantages of this:
 No room for error, if one bit is wrong the whole batch of ROMs
must be thrown out
 Data insertion step includes a relatively large fixed cost
+
Programmable ROM (PROM)
 Less expensive alternative
 Nonvolatile and may be written into only once
 Writing process is performed electrically and may be
performed by supplier or customer at a time later than the
original chip fabrication
 Special equipment is required for the writing process
 Provides flexibility and convenience
 Attractive for high volume production runs
 Read-Mostly Memory

Flash
EPROM EEPROM
Memory
Electrically erasable Intermediate between EPROM
Erasable programmable read- programmable read-only memory and EEPROM in both cost and
only memory
functionality

Can be written into at any time

without erasing prior contents
Uses an electrical erasing
Erasure process can be
technology, does not provide byte-
performed repeatedly
level erasure
Combines the advantage of non-
volatility with the flexibility of
being updatable in place

More expensive than PROM but Microchip is organized so that a

it has the advantage of the section of memory cells are erased
multiple update capability More expensive than EPROM in a single action or “flash”
 Typical 16 Mb DRAM (4M x 4)

MUltipleXer is a device that selects one of several  input

signals and forwards the selected input into a single line

Address lines

Data lines
Chip Packaging
 MAR

Figure 5.5

256-KByte
Memory
Organization

+ 1chip: 512*512= 218 bits

=256kb
8 chips  256KB
( 8bits/word)
 1MByte Module Organization

Select column

Select row
Data
Enable buffer

E: enable, signal permits the chip operating or not

Interleaved Memory
Composed of a collection of
DRAM chips

Grouped together to form a

memory bank

Each bank is independently

able to service a memory read
or write request

K banks can service K requests

simultaneously, increasing memory
read or write rates by a factor of K

If consecutive words of memory are

stored in different banks, the transfer
of a block of memory is speeded up
+ 5.2- Error Correction
 Hard Failure
 Permanent physical defect (khuyết tật).
 Memory cell or cells affected cannot reliably store data but become stuck at 0 or 1 or
switch erratically between 0 and 1
 Can be caused by:
 Harsh (khắc nghiệt) environmental abuse(sự ngược đãi)
 Manufacturing defects
 Wear (hao mòn)

 Soft Error
 Random, non-destructive event that alters the contents of one or more memory cells
 No permanent damage to memory
 Can be caused by:
 Power supply problems Alpha particles: Phenomenon in which 2
protons and 2 neutrons  bound together into a
 Alpha particles  particle identical to a helium nucleus (Wiki for
more details).
Error Correcting Code (ECC) Function

No error/Correctable

bits

Read
Write M+K

• No errors are detected. The fetched data bits are sent out.
• An error is detected, and it is possible to correct the error. The data bits plus error
correction bits are fed into a corrector, which produces a corrected set of M bits to
be sent out.
• An error is detected, but it is not possible to correct it. This condition is reported.
Next slide: An example for ECC function.
ECC Function: Examples
• The XOR operation is ussually used in ECC functions
• The most simple data for checking is the original data  A copy of
original data is written to memory . 8-bit data: 00001111, ECC
data: 00001111  Memory must be increased to double size
• XORs some bits of M-bit original data to K-bit ECC will decrease
memory size.
• Examples:
8 bits  3 bits: 01010110  101
8 bits  2 bits: 01010110  00
8 bits  1 bits: 01010110  0
- Main memory bank usually includes 9 chips. Why?
+ Hamming Data: 4 bits
1 XOR 1 XOR 1 = 1

Error
Correcting
Code
Richard Hamming at
Bell Laboratories

Parity bit (P) =1 if

number of 1s is odd.
Based on parity bit,
data can be corrected.
01
+ Increase in Word Length with ECC

Data 4 bits (22)  At least 3 bit ECC (2+1)

Data 8 bits (23)  At least 4 bit ECC (3+1)
+
Layout of Data Bits and Check Bits

Check positions: 23 22 21 20

(Algorithm for computing Ci bit is pre-defined)

Check Bit Calculation
Error

ECC write: 0111

ECC read: 0001

0111 XOR 0001 != 0  Error

 + Hamming SEC-DED Code
Single-Error Correcting/Double-Error Detecting

The sequence
shows that if
two errors occur
(Figure 5.11c),
the checking
procedure
goes astray –
chệch hướng (d)
and worsens the
problem by
creating a third
error (e).

To overcome the problem, an eighth bit is added that is set so that the total number of
1s in the diagram is even. The extra parity bit catches the error (f).
 Table 5.3
Performance Comparison
DRAM Alternatives

Table 5.3 Performance Comparison of Some DRAM Alternatives

 5.3- Advanced DRAM Organization
SDRAM
 One of the most critical system bottlenecks when
using high-performance processors is the interface to
main internal memory
DDR-DRAM
 The traditional DRAM chip is constrained both by its
internal architecture and by its interface to the
processor’s memory bus
 A number of enhancements to the basic DRAM
architecture have been explored: RDRAM

Table 5.3 Performance Comparison of Some DRAM Alternatives

Synchronous DRAM (SDRAM)

One of the most widely used forms of DRAM

Exchanges data with the processor synchronized to an external clock signal

and running at the full speed of the processor/memory bus without imposing
wait states

With synchronous access the DRAM moves data in and out under control of
the system clock
• The processor or other master issues the instruction and address information which
is latched by the DRAM
• The DRAM then responds after a set number of clock cycles
• Meanwhile the master can safely do other tasks while the SDRAM is processing
SDRAM
+
SDRAM Pin Assignments
+
SDRAM Read Timing
 RDRAM Developed by
Rambus
Rambus Dynamic Random Access Memory 
Bus delivers address and control
information using an asynchronous
block-oriented protocol Adopted by Intel
Gets a memory request over the
high-speed bus for its Pentium and
Request contains the desired
address, the type of operation, Itanium processors
and the number of bytes in the
operation

Protocol: pre-defined rule

Bus can address up

Has become the
to 320 RDRAM
main competitor to
chips and is rated
SDRAM
at 1.6 GBps
Chips are vertical packages with
all pins on one side
Exchanges data with the
processor over 28 wires no
more than 12 centimeters long
+
RDRAM Structure
+
Double Data Rate SDRAM
(DDR SDRAM)
 SDRAM can only send data once per bus clock cycle

 Double-data-rate SDRAM can send data twice per clock cycle, once
on the rising edge of the clock pulse and once on the falling edge

 Developed by the JEDEC Solid State Technology Association

(Electronic Industries Alliance’s semiconductor-engineering-
standardization body)
+
DDR SDRAM
Read
Timing
+
Cache DRAM (CDRAM)

 Developed by Mitsubishi

 Integrates a small SRAM cache onto a generic DRAM chip

 SRAM on the CDRAM can be used in two ways:

 It can be used as a true cache consisting of a number of 64-bit lines
 Cache mode of the CDRAM is effective for ordinary random access to
memory
 Can also be used as a buffer to support the serial access of a block of data
+ Exercises
 5.1 What are the key properties of semiconductor memory?

 5.2 What are two interpretations of the term random-access memory?

 5.3 What is the difference between DRAM and SRAM in terms of application?

 5.4 What is the difference between DRAM and SRAM in terms of characteristics such
as speed, size, and cost?

 5.5 Explain why one type of RAM is considered to be analog and the other digital.

 5.6 What are some applications for ROM?

 5.7 What are the differences among EPROM, EEPROM, and flash memory?

 5.8 Explain the function of each pin in Figure 5.4b. 182 CHAPTER 5 / INTERNAL
MEMORY

 5.9 What is a parity bit?

 5.10 How is the syndrome for the Hamming code interpreted?

 5.11 How does SDRAM differ from ordinary DRAM?

+ Summary Internal
Memory
Chapter 5

 Semiconductor main memory

 Hamming code
 Organization
 DRAM and SRAM  Advanced DRAM organization
 Types of ROM  Synchronous DRAM
 Chip logic  Rambus DRAM
 Chip packaging  DDR SDRAM
 Module organization  Cache DRAM
 Interleaved memory
 Error correction
 Hard failure
 Soft error