14 Memory

Shared by: ratzbiswas

Tags

memory foam, Maximum Memory, Memory Foam Mattress, Memory Upgrade, Standard memory, Memory Upgrades, Memory RAM, Memory management, System Memory, memory foam mattresses

Stats

views:: 16
posted:: 1/19/2010
language:: English
pages:: 65

Public Domain

Document Sample

Revision no.: PPT/2K403/02

Memory
Revision no.: PPT/2K403/02

Classification

• Memory is classified primarily into two parts i.e. primary
storage and secondary storage.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Classification (contd.)

• Primary Storage is further classified into two types i.e. ROM
and RAM.

• Secondary storage contains the different devices such as
Floppy disk, Hard disk, Zip disk and DAT cartridge.

• Memory normally refers to the amount of RAM installed in the
computer.

• Leading companies which are the memory suppliers are
Micron, Siemens etc.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Primary Storage Devices

• Primary Storage Device types available are
– ROM (Read Only Memory)

– RAM (Random Access Memory)

• ROM (Read Only Memory)
– ROM is where data is stored permanently. Hence it is also called
as Non Volatile Memory.

– ROM chip works necessitates the programming of complete data
when the chip is created.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Primary Storage Devices (contd.)

– ROMs use very little power, are extremely reliable and, contain all

the necessary programming to control the device.

– Different types of ROM are

• PROM (Programmable Read only Memory).

• EPROM (Erasable Programmable Read only Memory).

• EEPROM (Electrically Erasable Programmable Read Only Memory).

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

PROM

• It is basically a blank ROM chip that can be written to, but only

once.

• A jolt of static electricity can easily cause fuses in the PROM to

burn out, changing essential bits from 1 to 0.

• It is much like a CD-R drive that burns the data into the CD.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

EPROM

• It is just like PROM, except that you can erase the ROM by

shining a special ultra-violet light into a sensor on top of the

ROM chip for a certain amount of time.

• The ultra-violet light used is at a particular frequency that will

not penetrate most plastics or glasses, and each EPROM chip

has a quartz window on top of it.

• EPROM eraser is not selective, it will erase the entire EPROM.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

EEPROM

• EEPROM chip does not have to removed to be rewritten.

• The entire chip does not have to be completely erased to

change a specific portion of it.

• Instead of using UV light, you can return the electrons in the

cells of an EEPROM to normal with the localized application of

an electric field to each cell.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Flash Memory

• Flash memory is actually a variation of electrically erasable

programmable read-only memory (EEPROM).

• Flash memory devices are high density, low cost, nonvolatile,

fast (to read, but not to write), and electrically reprogrammable.

• Big difference between EEPROM and Flash is that EEPROM

can be erased and rewritten at the byte level whereas flash

memory can erase or reprogram blocks of bytes, not individual

bytes, hence it is faster.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

RAM (Random Access Memory)

• RAM is considered "random access" because one can access
any memory cell, which is the basic unit of data storage, in the
same amount of time.

• RAM is a volatile memory, meaning all data is lost when power
is turned off.

• Programs are loaded before the CPU processes the
information into the RAM.

• RAM is used for temporary storage of program data.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

RAM Basics

• In dynamic random access memory
(DRAM), a transistor and a capacitor are
paired to create a memory cell, which
represents a single bit of data.
• The transistor acts as a switch that lets
the control circuitry on the memory chip
read the capacitor or change its state.
• Charge on the capacitors used in RAM is
constantly refreshed so as to keep the
information within it and hence the name
Dynamic RAM.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Writing Operation

• Initial phase of writing data to a particular cell in RAM consists
of first activating the address line that is connected to cell
through an electrical pulse.
• When the transistor is turned on, the operating system sends
bursts of signals along the consecutive data line that
represent 0 or one which is found in cells sequentially.
• When an electrical pulse from the data reaches a transistor
that is activated by an address line, the transistor switches on
and allows current to pass through, thus charging the
capacitor connected to it.
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Types of Memory Packages

DIP( Dual in line package)

• Found in older pc's, 286, 386.

• Each memory chip is fitted into the individual
socket.

ZIP (Zigzag Inline Package)

• All of the connectors were on one side, allowing the
memory package to rest on its side.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Form Factor

• Memory chips are mounted on green circuit
boards called memory module.
• These modules are fitted on memory packages.
SIPP (Single Inline Pin Package)
• SIPP is a small circuit board containing several
memory chips and has a single row of pins
across the bottom.
• SIPP memory has tiny pins instead of an edge
connector.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

SIMM (Single in line memory module)

• It comprises a little circuit board on which chips are mounted.

• The circuit board fits into a Memory Slot on the Motherboard in

the same way as graphics card or any other card is fitted.
• SIMM's are of two types 30 pin and 72 pin,

• 386 and 486-SX used 30 pin SIMMs, 486-DX PCI chipset and

Pentium use 72 pin SIMMs.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DIMM (Dual Inline Memory Module)

DIMM has connectors on both sides of the module
• They are of 168 pin.
• Used for SDRAM.

SO-DIMM (Small Outline DIMM)

• Commonly used in notebook computers.

• It is smaller than the 168-pin DIMM and is
available in either 72 or 144-pin configurations.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

RIMM (RAMBUS Inline Memory Module)

• Implemented for RDRAM.
• It is proprietary of ASUS motherboards
• It is a 184-pin module offering faster access and
transfer speed, and thus generate more heat.
PC Cards, SmartMedia etc.
• These are small, thin modules that plug into a
special socket found mostly on notebook
computers, digital cameras, and Personal Digital
Assistants (PDAs).
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Memory Speed

• When the CPU needs information from memory,
it sends out a request that is managed by the
memory controller.
• The memory controller sends the request to
memory and reports to the CPU when the
information will be available for it to read.
• Entire cycle - from CPU to memory controller to
memory and back to the CPU - can vary in
length according to memory speed as well as
other factors, such as bus speed.
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Memory Speed (contd.)

• Memory speed is sometimes measured in

Megahertz (MHz), or in terms of access time

• The actual time required to deliver data -

measured in nanoseconds (ns). Is called as

access time.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Access Time (Nanoseconds)

• Access time measures from, when the memory
module receives a data request to, when that
data becomes available.

• Memory chips and modules used to be marked
with access times ranging from 80ns to 50ns.

• With access time measurements lower numbers
indicate faster speeds

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Example
• The memory controller requests data from memory and
memory reacts to the request in 70ns.
• The CPU receives the data in approximately 125ns.
• The total time from when the CPU first requests
information to when it actually receives the information
can be up to 195ns using a 70ns memory module.
• It takes time for the memory controller to manage the
information flow, and the information needs to travel from
the memory module to the CPU on the bus
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

MEGAHERTZ & SYSTEM CLOCK

MEGAHERTZ (MHZ) [millions of cycles per second]
• Beginning with Synchronous DRAM technology,
memory chips had the ability to synchronize
themselves with the computer's system clock
SYSTEM CLOCK
• A computer's system clock resides on the
motherboard.
• It sends out a signal to all other computer
components in rhythm, like a metronome.
• This rhythm is typically drawn as a square wave

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

SYSTEM CLOCK (contd.)

• If a system clock runs at 100MHz, that means
there are 100 million clock cycles in one second.
• Every action in the computer is timed by these
clock cycles, and every action takes a certain
number of clock cycles to perform
• It's possible for the CPU and other devices to
run faster or slower than the system clock.
• Components of different speeds simply require
a multiplication or division factor to synchronize
them.
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Types of RAM

– Two categories

• DRAM (dynamic RAM)

• SRAM (static RAM)

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DRAM

• Uses tiny capacitors to store charge corresponding to digitals
0s and 1s since capacitors always required dynamic
refreshing.
• Types of DRAM

1) FPM 5) DDR RAM
2) EDORAM 6) RDRAM
3) SDRAM 7) SGRAM
4) ECC DRAM 8) VRAM

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

FPM (Fast Page Mode) RAM

• A type of RAM that allows faster access if the data being called

is in the same row as the data previously requested.

• Also called page mode memory.

• First memory chips to use the burst mode timing, wherein data

is read 32 bytes at a time one after the other.

• Typical of processors from 8088/86 - 486.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

EDORAM

• It send data while data was being written in to it independently.

• This RAM is used from 80286 class machine till Pentium class

machines.

• It can not operate on a bus speed faster than 66MHz.

• It works at 3.3V.

• It is available with pin configuration of either 30 pins or 72

pins.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

SDRAM

• Synchronous DRAM is best suited to PII / PIII class computer

due to its 100 and 133 MHz operating speed.

• This RAM consists of two separate internal bank of transistors

for storing data.

• One bank of data can be accessed while the other is getting

ready thus streamlining the data

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

ECC DRAM

• Many higher-end systems use a special type of RAM called
error correction code (ECC) DRAM.

• NON-ECC is normally used by the end users.

• NON-ECC RAM checks out for any error occurred in parity bit,
but does not correct it, which is performed by ECC

• ECC detects problems in RAM quite well and can fix most of
them on the fly.

• ECC RAM are costly as compared with NON -ECC RAM

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DDR (Double Data Rate RAM)

• Similar to SDRAM operating at double speed of system bus of

SDRAM

• high data transfer rate at 1.066GB/sec.

• DDR doubles the throughput without increasing the clock

frequency.

• The maximum clock frequency remains at 133 MHz

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DDR working

• The technology in DDR plays with the way data is transferred.

Data transfer in SDRAM Data transfer in DDR
happens on the rising SDRAM happens on the
edge of the clocks pulse rising and falling edges of
the clock pulse

• Clock frequency can be represented as a square wave.
• This has a rising edge, a high-plane, a falling edge, and a low-
plane.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DDR working (contd.)

• In conventional SDRAM, one bit of data is transferred during

the rising edge of the clock cycle.

• Since the rising edge gets all the data, the falling edge

performs nothing in SDRAM.

• In DDR RAM the falling edge performs a ‘bit’ of data transfer.

• Resulted in two bits of data being transferred per clock cycle,

essentially doubling the transfer rate.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DDR vs SDRAM

• DDR memory also fits into DIMM (Dual In-line Memory Module)
slots, although the pin count is different.

• SDRAM has 168 pins

• DDR consists of 184 pins.

• You can buy DDR memory and fit it in your existing
motherboard with SDRAM.

• DDR reduced power consumption.

• SDRAM consumes 3 volts per signal

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

DDR vs SDRAM (contd.)

• DDR takes just 2.5 volts.

• Lower power requirements can help increase the
battery backup time in notebooks.

• Motherboard chipsets have to be designed to
support DDR as well as SDRAM.

• Many manufacturers provide this support for eg:-
Micron Samurai and AMD 760 chipsets.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Cont….
DDR vs SDRAM (contd.)

• DDR memory mostly used into high-end graphics workstations

or high-end server systems with multiple CPUs.

• This lets several users share the same system, and at the

same time giving them dedicated devices and memory space.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

RD RAM (Rambus Inline Memory Module)

• Rambus DRAM generally designed for AMD's CPU's, Intel's
copper mine, with speed up to 800MHz of teams

• Data transferring rates reaching 1.66B/sec.

• RDRAM offers high performance because of increased
operational frequency

• Three versions of it are intended: PC 600 (clock speed:
300MHz), PC700 (actually 711 or 356MHz), and PC 800
(400MHz).

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

RD RAM (contd.)

• RIMMs can develop hot spots apparently related to their speed

of operation, each RIMM has a heat spreader cover plate to try

to diffuse the heat.

• RDRAM RIMMs comes in 2 sizes: 184 pin for desktops and 160

pins SO RIMM for laptops.

• RIMMs can't be used on motherboards not designed with

Rambus sockets in place.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

SGRAM

• It is streamlined to work with graphics cards.

• Enables fast read and write operation for the graphics
processor when working with the information in the Video
frame buffer.

VRAM (Video RAM)

• Memory that is optimized for Video Cards where each memory
cell is dual ported.

• Video data can be written to the RAM while the graphics
adapter simultaneously reads from it to refresh the display.
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

SRAM

• Its the fastest type of RAM.

• It is expensive to fabricate.

• Storing of each bit requires several transistor.

• No refreshing required

• Classified as

• Core RAM

• Cache RAM

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Cache Memory

• Cache memory is a relatively small amount (normally less than

1MB) of high speed memory that resides very close to the

CPU.

• Cache memory is designed to supply the CPU with the most

frequently requested data and instructions.

• Retrieving data from cache takes a fraction of the time that it

takes to access it from main memory.

• Having cache memory can save a lot of time.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Cache Memory (contd.)

• Caches are organized into layers.

• The highest layer is closest to the device (such as the CPU)

using it.

• There are two levels of cache built right into the CPU.

• Any cache memory component is assigned a "level" according

to its proximity to the processor.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Cache Memory (contd.)

• The cache that is closest to the processor is called Level 1 (L1)
Cache.
• The next level of cache is numbered L2, then L3, and so on.
• Hit Rate
• Whenever the CPU finds the data it needs in the cache then it
is called a cache hit.
• When the CPU fails to find the data it needs in the cache that is
called a cache miss.
• The ratio of cache hits to cache misses is called that is called
a cache hit ratio.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Layers Of Cache

• Each layer of cache is closer to the processor and faster than
the layer below it.
• Each layer also caches the layers below it, due to its increased
speed relative to the lower levels.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Level 1 (Primary) Cache

• Level 1 or primary cache is the fastest memory on the PC.

• It is built directly into the processor itself.

• It is very small, generally from 8 KB to 64 KB, but it is

extremely fast and runs at the same speed as the processor.

© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

Level 2 (Secondary) Cache

• Level 2 cache is a secondary cache to the level 1 cache, and is

larger and slightly slower.

• Used to catch recent accesses that are not caught by the level

1 cache, and is usually 64 KB to 2 MB in size.

• Usually found either on the same package as the processor

itself (though it isn't in the same circuit where the processor

and level 1 cache are) or on the motherboard or as a

daughterboard that inserts into the motherboard.

Level 2 (Secondary) Cache (contd.)

• Write Through and Write Back :
• When the CPU writes new data to the cache, the cache
controller must update main memory with the new data.
• By making sure that the information in the cache is the same
as that in main memory the cache controller is said to maintain
cache coherency.
• If the cache controller allows the data in the cache to differ
from data in main memory, the data is said to be stale.
• Every time the CPU updates the cache, the data is
automatically written through to the main memory. Which is
called as write through cache.

Level 2 (Secondary) Cache (contd.)

• If the CPU needs to access the cache or main memory before
the write through is completed, the CPU must wait.
• This will slow the overall performance of the CPU.
• To prevent this problem the cache controller update a small
but fast buffer instead of directly updating the main memory.
• Because the buffer can be faster than the main memory, the
cache controller can make the cache available to the CPU
sooner.
• This method of updating the main memory is called a Buffered
or posted write through.

Level 2 (Secondary) Cache (contd.)

• The cache controller will keep track of which data is stale and

only update the memory when it must, not immediately

required after every memory write. This technique is called

write back or copy back.

• The concept of buffering, or posting, the writes can also be

applied to the write back cache to further increase its

performance as well.

• It results in the fastest cache.

Memory Management in DOS

• Memory Mapping is the allocation of memory for use by the

CPU.

• Original processors developed by Intel were unable to use

more than 1 MB of RAM.

• The original IBM PC allowed only the first 640 KB of memory

for direct use.

Memory Management in DOS (contd.)

• This 1 MB of memory was

divided into two sections to

make optimum use of the

memory space.

Conventional Memory

• Area from 0 to 640K (00000h to 9FFFFh) is called conventional

memory.

• Conventional memory contains all the memory addresses that

are set aside for RAM to run programs.

Reserved Memory

• All the addresses from A0000 to FFFFF are set aside for other

chips that the CPU may need to access, primarily ROMs and

specialized RAMs are called reserved memory.

• Most important device in the reserved area is the System BIOS.

• Video cards all have RAM that is mapped into the reserved

area in three different areas.

Expanded Memory Specification (EMS)

• IBM felt the 640K maximum of memory addresses was a
massive amount of addresses of RAM provided by DOS.
• As the size of programs and data grew, users added more
RAM chips to computers, which grew from 64K to 128K to
256K to 512K to 640K.
• One could no longer simply add memory chips since no more
memory addresses were available.
• An industry group led by Lotus, Intel, and Microsoft got
together in 1984 and came up with a way to put more memory
into a computer while staying within the 1MB limit by
expanded memory specification (EMS).
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute
Revision no.: PPT/2K403/02

EMS (contd.)

• Expanded memory was originally an expansion card full of
RAM chips.

• Chips on the card were electronically divided into 16K chunks
called pages.

• To access this card of RAM chips, a device driver to access
the card and applications that knew how to talk to the device
driver in order to get work done with the card was needed.

• Device driver was called EMM.SYS. EMM.SYS electronically
readdressed the chips on the expanded memory board.

EMS (contd.)

• When EMM.SYS initialized, it took the expanded memory card's first

four 16K pages and addresses and changed them into one, unused,

64K area in the reserved area called as EMS page frame.

• EMM.SYS could swap different pages into and out of this 64K area, so

you could load large amounts of data onto the EMS card while using

only 64K of memory.

• For a program to use EMS memory, it had to be specifically written to

do so, and virtually every commercial DOS program ever made could

utilize EMS.

EMS (contd.)

Shadow RAM

• Shadow RAM improve the performance of a computer.

• It rewrites (or shadows) the contents of the ROM BIOS

and/or video BIOS into extended RAM (between the 640-

KB boundary and 1 MB).

EMS (contd.)

Upper Memory Blocks (UMBs)

• UMBs sit between 640K and 1 MB, in the holes between video

memory areas, the system and video ROMs, and any other

ROMs inserted into the area by other expansion cards.

Upper Memory Blocks (UMBs) (contd.)

• High Memory Area

• Intel allowed MS-DOS to address the first 64 KB of extended

memory on machines with 80286 or faster processors.

• This special area is called the high memory area (HMA).

• A software driver called an A20 handler is run to allow the

processor to access the HMA.

High Memory Access

Extended Memory (XMS)

• Once the processor goes into protected mode, it can access
more than 1MB of memory and all memory above 1MB is called
extended memory (XMS).
• DOS cannot directly use extended memory since it's not really
part of DOS memory management, with the exception of the
HMA.
• Two ways of using this extended memory.
– Convert extended memory into expanded memory.
– Using extended memory directly using specialized softwares.

Extended Memory (XMS) (contd.)

• To control the access to HMA, Microsoft introduced the XMS

standard & implemented this in HIMEM.SYS extended Memory

driver.

• To convert extended memory to expanded memory following

line should added to your CONFIG.SYS file.

DEVICE=HIMEM.SYS

• Windows 9x and operating systems above loads HIMEM.SYS

automatically in the conventional memory at start up.

Exercise

• Exercise 14.1 Memory Identification & checking DOS Memory
Management

Design & Published by:
CMS Institute, Design & Development Centre, CMS House, Plot No. 91, Street No.7,
MIDC, Marol, Andheri (E), Mumbai –400093, Tel: 91-22-28216511, 28329198
Email: courseware.inst@cmail.cms.co.in
www.cmsinstitute.co.in
© CMS INSTITUTE, 2004. All rights reserved. No part of this material may be reproduced, stored or emailed without the prior permission of Programme Director, CMS Institute

Related docs