Chapter 7-Memory System Design

C
S Chapter 7- Memory System Design

D
A
2/e
 Introduction
 RAM structure: Cells and Chips
 Memory boards and modules
 Two-level memory hierarchy
 The cache
 Virtual memory
 The memory as a sub-system of the computer
Computer Systems Design and Architecture Second Edition © 2004 Prentice Hall
C
S Introduction
D
A
2/e
So far, we’ve treated memory as an array of words limited in
size
only by the number of address bits. Life is seldom so easy...
Real world issues arise:

•cost
•speed
•size
•power consumption
•volatility
•etc.
What other issues can you think of that will influence memory
design?
C
S In This Chapter we will cover–
D
A
2/e •Memory components:
•RAM memory cells and cell arrays
•Static RAM–more expensive, but less complex
•Tree and Matrix decoders–needed for large RAM chips
•Dynamic RAM–less expensive, but needs “refreshing”
•Chip organization
•Timing
•Commercial RAM products" SDRAM and DDR RAM
•ROM–Read only memory
•Memory Boards
•Arrays of chips give more addresses and/or wider words
•2-D and 3-D chip arrays
• Memory Modules
•Large systems can benefit by partitioning memory for
•separate access by system components
•fast access to multiple words
–more–
C
S In This Chapter we will also cover–
D
A
2/e
• The memory hierarchy: from fast and expensive to slow and cheap
 Example: Registers->Cache–>Main Memory->Disk
 At first, consider just two adjacent levels in the hierarchy
 The Cache: High speed and expensive
 Kinds: Direct mapped, associative, set associative
 Virtual memory–makes the hierarchy transparent
 Translate the address from CPU’s logical address to the
physical address where the information is actually stored
 Memory management - how to move information back and forth
 Multiprogramming - what to do while we wait
 The “TLB” helps in speeding the address translation process
 Will discuss temporal and spatial locality as basis for success of
cache and virtual memory techniques.
• Overall consideration of the memory as a subsystem.
C
Fig. 7.1 The CPU–Main Memory Interface
S
D
A
2/e
Sequence of events:
Read:
1. CPU loads MAR, issues Read, and REQUEST
2. Main Memory transmits words to MDR
3. Main Memory asserts COMPLETE.
Write:
1. CPU loads MAR and MDR, asserts Write, and REQUEST
2. Value in MDR is written into address in MAR.
3. Main Memory asserts COMPLETE.
C
S
D The CPU–Main Memory Interface - cont'd.
A
2/e
Additional points:
•if b<w, Main Memory must make w/b b-bit transfers.
•some CPUs allow reading and writing of word sizes <w.
Example: Intel 8088: m=20, w=16,s=b=8.
8- and 16-bit values can be read and written
•If memory is sufficiently fast, or if its response is predictable,
then COMPLETE may be omitted.
•Some systems use separate R and W lines, and omit REQUEST.
C
S
D Table 7.1 Some Memory Properties
A
2/e
Symbol Definition Intel Intel IBM/Moto.

8088 8086 601
w CPU Word Size 16bits 16bits 64 bits

m Bits in a logical memory address 20 bits 20 bits 32 bits
s Bits in smallest addressable unit 8 8 8
b Data Bus size 8 16 64
2m Memory wd capacity, s-sized wds 220 220 232
2mxs Memory bit capacity 220x8 220x8 232x8
C
S Big-Endian and Little-Endian
D Storage
A
When data types having a word size larger than the smallest
2/e addressable unit are stored in memory the question arises,
“Is the least significant part of the word stored at the

lowest address (little Endian, little end first) or–
is the most significant part of the word stored at the

lowest address (big Endian, big end first)”?
Example: The hexadecimal 16-bit number ABCDH, stored at address 0:

msb ... lsb
AB CD
Little Endian Big Endian
1 AB 1 CD
0 CD 0 AB
C
S Table 7.2 Memory Performance
D Parameters
A
2/e
Symbol Definition Units Meaning
ta Access time time Time to access a memory word
tc Cycle time time Time from start of access to start of next

access
k Block size words Number of words per block
b Bandwidth words/time Word transmission rate
tl Latency time Time to access first word of a sequence
of words
tbl = Block time Time to access an entire block of words

tl + k/b access time
(Information is often stored and moved in blocks at the cache and disk level.)
C
S Table 7.3 The Memory Hierarchy, Cost, and
D Performance
A
2/e Compo-
nent
CPU Tape
Some Cache Main Memory Disk Memory Memory
Typical
Values:† Access Random Random Random Direct Sequential
Capa- 64-1024+ 8KB-8MB 64MB-2GB 8GB 1TB

city,
bytes
Latency .4-10ns .4-20ns 10-50ns 10ms 10ms-10s
Block 1 word 16 words 16 words 4KB 4KB

size
Band- System System 10-4000 50MB/s 1MB/s

width clock Clock MB/s
Rate rate-80MB/s
Cost/MB High $10 $.25 $0.002 $0.01

†As of 2003-4. They go out of date immediately.
C
S
D Fig. 7.3 Memory Cells - a conceptual view
A
2/e Regardless of the technology, all RAM memory cells must provide
these four functions: Select, DataIn, DataOut, and R/W.
Select Select
DataIn
DataOut ≡ DataIn DataOut
R/W
R/W
This “static” RAM cell is unrealistic.

We will discuss more practical designs later.
C
S
D Fig. 7.4 An 8-bit register as a 1D RAM array
A
2/e The entire register is selected with one select line, and uses one R/W line
Data bus is bi-directional, and buffered. (Why?)
C
S Fig. 7.5 A 4x8 2D Memory Cell Array
D 2-4 line decoder selects one of the four 8-bit arrays
A
2/e
2-bit
address
R/W is common
to all.
Bi-directional 8-bit buffered data bus
C
S
D Fig. 7.6 A 64Kx1 bit static RAM (SRAM) chip
A
~square array fits IC design
2/eparadigm
Selecting rows separately

from columns means only
256x2=512 circuit elements
instead of 65536 circuit
elements!
CS, Chip Select, allows chips in arrays to

be selected individually
This chip requires 21 pins including power and ground, and so
will fit in a 22 pin package.
C
S
D Fig 7.7 A 16Kx4 SRAM Chip
A
2/e
There is little difference

between this chip and
the previous one, except
that there are 4, 64-1
Multiplexers instead of 1,
256-1 Multiplexer.
This chip requires 24 pins including power and ground, and so will require a 24
pin pkg. Package size and pin count can dominate chip cost.
C
S Fig 7.8 Matrix and Tree Decoders
D
A
2/e•2-level decoders are limited in size because of gate fanin.
Most technologies limit fanin to ~8.
•When decoders must be built with fanin >8, then additional levels
of gates are required.
•Tree and Matrix decoders are two ways to design decoders with large fanin:
3-to-8 line tree decoder constructed 4-to-16 line matrix decoder

from 2-input gates. constructed from 2-input gates.
C
S
D
Fig 7.9 A 6 Transistor static RAM cell
A
2/e
This is a more practical

design than the 8-gate
design shown earlier.
A value is read by
precharging the bit
lines to a value 1/2
way between a 0 and
a 1, while asserting the
word line. This allows the
latch to drive the bit lines
to the value stored in
the latch.
C
S Figs 7.10 Static RAM Read Timing
D
A
2/e
Access time from Address– the time required of the RAM array to decode the
address and provide value to the data bus.
C
S Figs 7.11 Static RAM Write Timing
D
A
2/e
Write time–the time the data must be held valid in order to decode address and
store value in memory cells.
C
S Fig 7.12 A Dynamic
D RAM (DRAM) Cell
A
2/e
Capacitor will
discharge in 4-15ms.
Refresh capacitor by reading
(sensing) value on bit line,
amplifyingacitor.
Write: place value on bit line

and assert word line.
Read: precharge bit line,
assert word line, sense value
on bit line with sense/amp.
This need to refresh the
storage cells of dynamic
RAM chips complicates
DRAM system design.
C
S Fig 7.13 DRAM Chip
D organization
A
2/e
•Addresses are time-

multiplexed on address bus
using RAS and CAS as
strobes of rows and
columns.
•CAS is normally used as
the CS function.
Notice pin counts:

•Without address multiplexing:
27 pins including power and
ground.
•With address multiplexing: 17
pins including power and
ground.
C
S
D
Figs 7.14, 7.15 DRAM Read and Write cycles
A
2/e Typical DRAM Read operation Typical DRAM Write operation
Memory Memory
Row Addr Col Addr Row Addr Col Addr
Address Address
RAS t RAS t Prechg RAS t RAS Prechg
CAS CAS
R/W W
Data Data
tA tDHR
tC tC
Access time Cycle time Data hold from RAS.

Notice that it is the bit line precharge
operation that causes the difference
between access time and cycle time.
C
S DRAM Refresh and row access
D
A
2/e
•Refresh is usually accomplished by a “RAS-only” cycle. The row address
is placed on the address lines and RAS asserted. This refreshed the entire row.
CAS is not asserted. The absence of a CAS phase signals the chip that a
row refresh is requested, and thus no data is placed on the external data lines.
•Many chips use “CAS before RAS” to signal a refresh. The chip has an internal
counter, and whenever CAS is asserted before RAS, it is a signal to refresh the row
pointed to by the counter, and to increment the counter.
•Most DRAM vendors also supply one-chip DRAM controllers that encapsulate
the refresh and other functions.
•Page mode, nibble mode, and static column mode allow rapid access to
the entire row that has been read into the column latches.
•Video RAMS, VRAMS, clock an entire row into a shift register where it can
be rapidly read out, bit by bit, for display.
C
S
D Fig 7.16 A CMOS ROM Chip
A
2/e
2-D CMOS ROM Chip
+V
Row
Decoder
00 Address
CS
1 0 1 0
C
S
D Tbl 7.4 Kinds of ROM
A
2/e
ROM Type Cost Programmability Time to program Time to erase
Mask pro- Very At the factory Weeks (turn around) N/A

grammed inexpensive
PROM Inexpensive Once, by end Seconds N/A

user
EPROM Moderate Many times Seconds 20 minutes
Flash Expensive Many times 100 us. 1s, large

EPROM block
EEPROM Very Many times 100 us. 10 ms,

expensive byte
C
S
D Memory boards and modules
A
2/e
•There is a need for memories that are larger and wider than a single chip
•Chips can be organized into “boards.”
•Boards may not be actual, physical boards, but may consist of
structured chip arrays present on the motherboard.
•A board or collection of boards make up a memory module.
•Memory modules:
•Satisfy the processor–main memory interface requirements
•May have DRAM refresh capability
•May expand the total main memory capacity
•May be interleaved to provide faster access to blocks of words.
C
S Fig 7.17 General structure of memory chip
D
A
2/e
This is a slightly different view of the memory chip than previous.
Multiple chip selects ease the assembly of

chips into chip arrays. Usually provided
Chip Selects by an external AND gate.
Address
Address
m Decoder
CS
Memory R/W R/W
m
Cell Address
Array
Data
s
I/O s s s
Multiplexer
Data
Bi-directional data bus.
C
S
D Fig 7.18 Word Assembly from Narrow Chips
A
2/e All chips have common CS, R/W, and Address lines.
Select
Address
R/W
CS CS CS
R/W R/W R/W
Address Address Address
Data Data Data
s s s
p×s
P chips expand word size from s bits to p x s bits.
C
S Fig 7.19 Increasing the Number of Words by a
D Factor of 2k
A
2/e k
The additional k address bits are used to select one of 2 chips,
each one of which has 2m words:
Word size remains at s bits.
C
Fig 7.20 Chip
S Matrix Using Two
D Chip Selects
A
2/e
Multiple chip
select lines
are used to
replace the
This scheme last level of
simplifies the gates in this
decoding from matrix
use of a (q+k)- decoder
bit scheme.
decoder
to using one
q-bit and one
k-bit decoder.
C
S Fig 7.21 A 3-D
D DRAM Array
A
2/e
•CAS is used to enable

top decoder in decoder
tree.
•Use one 2-D array for

each bit. Each 2-D array
on separate board.
C
S
Fig 7.22 A Memory Module interface
D
A Must provide–
2/e •Read and Write signals.
•Ready: memory is ready to accept commands.
•Address–to be sent with Read/Write command.
•Data–sent with Write or available upon Read when Ready is asserted.
•Module Select–needed when there is more than one module.
Address
Bus Interface: k+m
Address register
k
m
Chip/board
selection
Module Memory boards
Control signal generator: select and/or
for SRAM, just strobes Read Control
chips
data on Read, Provides signal
Write generator
Ready on Read/Write
Ready w
For DRAM–also provides
Data register
CAS, RAS, R/W, multiplexes
address, generates refresh Data
w
signals, and provides Ready.
C
S
D Fig 7.23 DRAM module with refresh control
A .
Address
2/e k+m
Address Register
k
Chip/board
selection m/2 m/2 m/2
Refresh counter Address

Refresh
clock and Multiplexer
control 2 m/2
Address lines
Request
Refresh
Board and
Grant
chip selects
Module
select
RAS
Dynamic
Read Memory RAM Array
timing CAS
Write generator
R/W
Data lines
Ready w
Data register
w
Data
C
S Fig 7.24 Two Kinds of Memory Module Organization.
D
A
2/e
Memory Modules
are used to allow
access to more
than one word
simultaneously.
•Scheme (a)
supports filling a
cache line.
•Scheme (b) allows
multiple processes
or processors to
access memory at
once.
C
S Fig 7.25 Timing of Multiple Modules on a
D Bus
A
2/e If time to transmit information over bus, tb, is < module cycle time, tc,
it is possible to time multiplex information transmission to several
modules;
Example: store one word of each cache line in a separate module.
Main Memory Address: Word Module No.
This provides successive words in successive modules.
Timing: Bus Read module 0 Write module 3 Module 0

Address Address & data Data return
Module 0 Module 0 read
Module 3 Module 3 write
tb tc tb
With interleaving of 2k modules, and tb < tb/2k, it is possible to get a 2k-fold

increase in memory bandwidth, provided memory requests are pipelined.
DMA satisfies this requirement.
C
S
Memory system performance
D
A
2/e Breaking the memory access process into steps:
For all accesses:

•transmission of address to memory
•transmission of control information to memory (R/W, Request, etc.)
•decoding of address by memory
For a read:
•return of data from memory
•transmission of completion signal
For a write:
•Transmission of data to memory (usually simultaneous with address)
•storage of data into memory cells
•transmission of completion signal
The next slide shows the access process in more detail --

C
S
D Fig 7.26 Static and dynamic RAM timing
A
2/e
“Hidden refresh” cycle. A normal cycle would exclude the

pending refresh step.
-more-
C
S Example SRAM timings (using unrealistically
D long timing)
A
2/e
Approximate values for static RAM Read timing:
•Address bus drivers turn-on time: 40 ns.

•Bus propagation and bus skew: 10 ns.
•Board select decode time: 20 ns.
•Time to propagate select to another board: 30 ns.
•Chip select: 20ns.
PROPAGATION TIME FOR ADDRESS AND COMMAND TO REACH CHIP: 120 ns.
•On-chip memory read access time: 80 ns

•Delay from chip to memory board data bus: 30 ns.
•Bus driver and propagation delay (as before): 50 ns.
TOTAL MEMORY READ ACCESS TIME: 280 ns.
Moral: 70ns chips to not necessarily provide 70ns access time!
C
S Considering any two adjacent
D levels of the memory hierarchy
A
Some definitions:
2/e
Temporal locality: the property of most programs that if a given memory
location is referenced, it is likely to be referenced again, “soon.”
Spatial locality: if a given memory location is referenced, those locations

near it numerically are likely to be referenced “soon.”
Working set: The set of memory locations referenced over a fixed period of
time, or in a time window.
Notice that temporal and spatial locality both work to assure that the contents
of the working set change only slowly over execution time.
Defining the Primary and Secondary levels:
Faster, Slower,
smaller larger
Primary Secondary
CPU • • • level • • •
level
two adjacent levels in the hierarchy

C
S Figure 7.28 Temporal and Spatial Locality Example
D
A
2/e
Consider the C for loop:
for ((I=0); I<n; I++)

A[I] = 0;
C
S Primary and secondary levels
D of the memory hierarchy
A
2/e Speed between levels defined by latency: time to access first word, and
bandwidth, the number of words per second transmitted between levels.
Primary Secondary
level level Typical latencies:
cache latency: a few clocks
Disk latency: 100,000 clocks
•The item of commerce between any two levels is the block.
•Blocks may/will differ in size at different levels in the hierarchy.

Example: Cache block size ~ 16-64 bytes.
Disk block size: ~ 1-4 Kbytes.
•As working set changes, blocks are moved back/forth through the
hierarchy to satisfy memory access requests.
•A complication: Addresses will differ depending on the level.

Primary address: the address of a value in the primary level.
Secondary address: the address of a value in the secondary level.
C
S
D
Primary and secondary address examples
A
2/e
•Main memory address: unsigned integer
•Disk address: track number, sector number, offset of word in sector.
C
SFig 7.29 Addressing and Accessing a 2-Level Hierarchy
DThe
Acomputer
2/e
system, HW
or SW,
must perform
any address
translation
that is
required:
Two ways of forming the address: Segmentation and Paging.

Paging is more common. Sometimes the two are used together,
one “on top of” the other. More about address translation and paging later...
C
S
D
Fig 7.30 Primary Address Formation
A
2/e
C
S Hits and misses; paging;
D block placement
A Hit: the word was found at the level from which it was requested.
2/e
Miss: the word was not found at the level from which it was requested.
(A miss will result in a request for the block containing the word from
the next higher level in the hierarchy.)
Hit ratio (or hit rate) = h = number of hits

total number of references
Miss ratio: 1 - hit ratio
tp = primary memory access time. ts = secondary memory access time
Access time, ta = h • tp + (1-h) • ts.
Page: commonly, a disk block. Page fault: synonymous with a miss.
Demand paging: pages are moved from disk to main memory only when
a word in the page is requested by the processor.
Block placement and replacement decisions must be made each time a

block is moved.
C
S
D Virtual memory
A
2/e
a Virtual Memory is a memory hierarchy, usually consisting of at least
main memory and disk, in which the processor issues all memory
references as effective addresses in a flat address space. All translations
to primary and secondary addresses are handled transparently to the
process making the address reference, thus providing the illusion of a flat
address space.
Recall that disk accesses may require 100,000 clock cycles to complete,
due to the slow access time of the disk subsystem. Once the processor
has, through mediation of the operating system, made the proper request
to the disk subsystem, it is available for other tasks.
Multiprogramming shares the processor among independent programs

that are resident in main memory and thus available for execution.
C
S Decisions in designing a
D 2-level hierarchy
A
2/e
•Translation procedure to translate from system address to primary address.
•Block size–block transfer efficiency and miss ratio will be affected.
•Processor dispatch on miss–processor wait or processor multiprogrammed.
•Primary level placement–direct, associative, or a combination. Discussed later.
•Replacement policy–which block is to be replaced upon a miss.
•Direct access to secondary level–in the cache regime, can the processor
directly access main memory upon a cache miss?
•Write through–can the processor write directly to main memory upon a cache
miss?
•Read through–can the processor read directly from main memory upon a
cache miss as the cache is being updated?
•Read or write bypass–can certain infrequent read or write misses be satisfied

by a direct access of main memory without any block movement?
C
S Fig 7.31 The Cache Mapping Function
D
A Example: 256KB 16words 32MB
2/e
The cache mapping function is responsible for all cache operations:

•Placement strategy: where to place an incoming block in the cache
•Replacement strategy: which block to replace upon a miss
•Read and write policy: how to handle reads and writes upon cache misses.
Mapping function must be implemented in hardware. (Why?)
Three different types of mapping functions:

•Associative
•Direct mapped
•Block-set associative
C
S Memory fields and address translation
D
A
2/e Example of processor-issued 32-bit virtual address:
0
31
32 bits
That same 32-bit address partitioned into two fields, a block field,
and a word field. The word field represents the offset into the block
specified in the block field:
Block Number Word
26 6
226 64 word blocks
Example of a specific memory reference: word 11 in block 9.
00 ••• 001001 001011

C
S Fig 7.32 Associative mapped caches
D
A
Associative mapped
2/e
cache model: any
block from main
memory can be put
anywhere in the
cache.
Assume a 16-bit main
memory.*
*16 bits, while unrealistically small, simplifies the examples

C
S Fig 7.33 Associative cache mechanism
D
A
2/e Because any block can reside anywhere in the cache, an associative, or content
addressable memory is used. All locations are searched simultaneously.
C
S Advantages and disadvantages
D of the associative mapped cache.
A
2/e
Advantage
•Most flexible of all–any MM block can go anywhere in the cache.
Disadvantages
•Large tag memory.
•Need to search entire tag memory simultaneously means lots of
hardware.
Replacement Policy is an issue when the cache is full. –more later–
Q.: How is an associative memory implemented? Hint: Think XNOR gates.
–next–
Direct mapped caches simplify the hardware by allowing each MM block
to go into only one place in the cache.
C
S Fig 7.34 The direct mapped cache
D
A
2/e
Key Idea: all the MM

blocks from a given
group can go into only
one location in the
cache, corresponding to
the group number.
Now the cache needs only examine the single group

that its reference specifies.
C
S Fig 7.35 Direct Mapped Cache Operation
D
A
2/e 1. Decode the
group number of
the incoming
MM address to
select the group
2. If Match
AND Valid
3. Then gate out

the tag field
4. Compare
cache tag with
incoming tag
5. If a hit, then
gate out the
cache line, 6. and use the word field to
select the desired word.
C
S
D Direct mapped caches
A
2/e
•The direct mapped cache uses less hardware, but is
much more restrictive in block placement.
•If two blocks from the same group are frequently

referenced, then the cache will “thrash.” That is,
repeatedly bring the two competing blocks into and out of
the cache. This will cause a performance degradation.
•Block replacement strategy is trivial.
•Compromise - allow several cache blocks in each

group–the Block Set Associative Cache. –next–
C
S
D Fig 7.36 2-Way Set Associative Cache
A
2/e
Example shows 256 groups, a set of two per group.
Sometimes referred to as a 2-way set associative cache.
C
S Getting Specific:
D The Intel Pentium Cache
A
2/e
•The Pentium actually has two separate caches–one for instructions and
one for data. Pentium issues 32-bit MM addresses.
•Each cache is 2-way set associative

•Each cache is 8K=213 bytes in size
•32 = 25 bytes per line.
•Thus there are 64 or 26 bytes per set, and therefore 213/26 or 27=128 groups
•This leaves 32-5-7 = 20 bits for the tag field:
Tag Set (group) Word

20 7 5
31 0
This “cache arithmetic” is important, and deserves your mastery.
C
S
D
Cache Read and Write policies
A
2/e
•Read and Write cache hit policies
•Write-through–updates both cache and MM upon each write.
•Write back–updates only cache. Updates MM only upon block removal.
•“Dirty bit” is set upon first write to indicate block must be written back.
•Read and Write cache miss policies

•Read miss - bring block in from MM
•Either forward desired word as it is brought in, or
•Wait until entire line is filled, then repeat the cache request.
•Write miss
•Write allocate - bring block into cache, then update
•Write - no allocate - write word to MM without bringing block into cache.
C
S Block replacement strategies
D
A
2/e
•Not needed with direct mapped cache
•Least Recently Used (LRU)

•Track usage with a counter. Each time a block is accessed:
•Clear counter of accessed block
•Increment counters with values less than the one accessed
•All others remain unchanged
•When set is full, remove line with highest count.
•Random replacement - replace block at random.

•Even random replacement is a fairly effective strategy.
C
S Cache performance
D
A
2/e
Recall Access time, ta = h • tp + (1-h) • ts for Primary and Secondary levels.
For tp = cache and ts = MM,
ta = h • tC + (1-h) • tM

We define S, the speedup, as S= Twithout/Twith for a given process,
where Twithout is the time taken without the improvement, cache in
this case, and Twith is the time the process takes with the improvement.
Having a model for cache and MM access times, and cache line fill time,
the speedup can be calculated once the hit ratio is known.
C
SFig 7.37 Getting Specific: The PowerPC 601 Cache
D
A
2/e
•The PPC 601 has a unified cache - that is, a single cache for both instructions and
data.
•It is 32KB in size, organized as 64x8block set associative, with blocks being 8 8-
byte words organized as 2 independent 4 word sectors for convenience in the
updating process
•A cache line can be updated in two single-cycle operations of 4 words each.
•Normal operation is write back, but write through can be selected on a per line
basis via software. The cache can also be disabled via software.
C
S Virtual memory
D The Memory Management Unit, MMU is responsible for mapping logical
A addresses issued by the CPU to physical addresses that are presented to
2/e the Cache and Main Memory. CPU Chip
A word about addresses:

•Effective Address - an address computed by by the processor while
executing a program. Synonymous with Logical Address
•The term Effective Address is often used when referring to activity
inside the CPU. Logical Address is most often used when referring to
addresses when viewed from outside the CPU.
•Virtual Address - the address generated from the logical address by the
Memory Management Unit, MMU.
•Physical address - the address presented to the memory unit.

(Note: Every address reference must be translated.)
C
S Virtual addresses - why
D
A
2/e The logical address provided by the CPU is translated to
a virtual address by the MMU. Often the virtual address
space is larger than the logical address, allowing program
units to be mapped to a much larger virtual address
space.
Getting Specific: The PowerPC 601

•The PowerPC 601 CPU generates 32-bit logical addresses.
•The MMU translates these to 52-bit virtual addresses, before
the final translation to physical addresses.
•Thus while each process is limited to 32 bits, the main memory

•can contain many of these processes.
•Other members of the PPC family will have different logical

and virtual address spaces, to fit the needs of various members
of the processor family.
C
S Virtual addressing - advantages
D
A •Simplified. Each program unit can be compiled into its own memory space,
2/ebeginning at address 0 and potentially extending far beyond the amount of
physical memory present in the system.
•No address relocation required at load time.
•No need to fragment the program to accommodate
•Cost effective use of physical memory.
•Less expensive secondary (disk) storage can replace primary storage.
(The MMU will bring portions of the program into physical memory as
required)
•Access control. As each memory reference is translated, it can be

simultaneously checked for read, write, and execute privileges.
•This allows access/security control at the most fundamental levels.
•Can be used to prevent buggy programs and intruders from causing
damage to other users or the system.
This is the origin of those “bus error” and “segmentation fault" messages
C
S Fig 7.39 Memory management by Segmentation
D
A
2/e
•Notice that each segment’s virtual address and out of physical memory will result
in gaps between segments. This is called external fragmentation.
C
S Fig 7.40 Segmentation Mechanism
D
A
2/e
•The computation of physical address from virtual address requires an integer

addition for each memory reference, and a comparison if segment limits are
checked.
•Q: How does the MMU switch references from one segment to another?
C
S Fig 7.41 The Intel 8086 Segmentation Scheme
D
A
2/e
The first popular 16-bit

processor, the Intel 8086
had a primitive segmentation
scheme to “stretch” its 16-bit
logical address to a 20-bit
physical address:
The CPU allows 4 simultaneously active segments,

CODE, DATA, STACK, and EXTRA. There are 4 16-bit segment base
registers.
C
S Fig 7.42 Memory management by paging
D
A
2/e
•This figure shows the mapping between virtual memory pages, physical memory
pages, and pages in secondary memory. Page n-1 is not present in physical
memory, but only in secondary memory.
•The MMU that manages this mapping. -more-
C
S Fig 7.43 The Virtual to Physical Address Translation
D Process
A
2/e
•1 table per
user per
program unit
•One
translation per
memory
access
•Potentially
large page
table
A page fault will result in 100,000 or more cycles passing before the page
has been brought from secondary storage to MM.
C
S
D Page Placement and Replacement
A
2/e
Page tables are direct mapped, since the physical page is computed
directly from the virtual page number.
But physical pages can reside anywhere in physical memory.
Page tables such as those on the previous slide result in large page
tables, since there must be a page table entry for every page in the
program unit.
Some implementations resort to hash tables instead, which need have

entries only for those pages actually present in physical memory.
Replacement strategies are generally LRU, or at least employ a “use bit”

to guide replacement.
C
S Fast address translation:
D regaining lost ground
A
2/e
•The concept of virtual memory is very attractive, but leads to considerable
overhead:
•There must be a translation for every memory reference
•There must be two memory references for every program reference:
•One to retrieve the page table entry,
•one to retrieveMost caches are addressed by physical address, so
there must be a virtual to physical translation before the cache can be
accessed.
The answer: a small cache in the processor that retains the last few virtual
to physical translations: A Translation Lookaside Buffer, TLB.
The TLB contains not only the virtual to physical translations, but also the
valid, dirty, and protection bits, so a TLB hit allows the processor to access
physical memory directly.
The TLB is usually implemented as a fully associative cache. -more-
C
S Fig 7.44 TLB Structure and Operation
D
A
2/e
C
S
D Fig 7.45 Operation of the Memory Hierarchy
A
2/e
C
S Fig 7.46 The PowerPC 601 MMU Operation
D
A
2/e
“Segments” are
actually more
akin to
large (256 MB)
blocks.
C
S Fig 7.47 I/O Connection to a Memory with a Cache
D
A
2/e
•The memory system is quite complex, and affords many possible tradeoffs.
•The only realistic way to chose among these alternatives is to study a
typical workload, using either simulations or prototype systems.
•Instruction and data accesses usually have different patterns.
•It is possible to employ a cache at the disk level, using the disk hardware.
•Traffic between MM and disk is I/O, and Direct Memory Access, DMA can be
used to speed the transfers:

Chapter 7-Memory System Design

Uploaded by

Copyright:

Available Formats

Chapter 7-Memory System Design

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter 7-Memory System Design

Uploaded by

Copyright:

Available Formats

C

S Chapter 7- Memory System Design

Real world issues arise:

Symbol Definition Intel Intel IBM/Moto.

w CPU Word Size 16bits 16bits 64 bits

“Is the least significant part of the word stored at the

is the most significant part of the word stored at the

Example: The hexadecimal 16-bit number ABCDH, stored at address 0:

ta Access time time Time to access a memory word

tc Cycle time time Time from start of access to start of next

tbl = Block time Time to access an entire block of words

Capa- 64-1024+ 8KB-8MB 64MB-2GB 8GB 1TB

Latency .4-10ns .4-20ns 10-50ns 10ms 10ms-10s

Block 1 word 16 words 16 words 4KB 4KB

Band- System System 10-4000 50MB/s 1MB/s

Cost/MB High $10 $.25 $0.002 $0.01

DataOut ≡ DataIn DataOut

This “static” RAM cell is unrealistic.

Data bus is bi-directional, and buffered. (Why?)

Bi-directional 8-bit buffered data bus

Selecting rows separately

CS, Chip Select, allows chips in arrays to

There is little difference

3-to-8 line tree decoder constructed 4-to-16 line matrix decoder

This is a more practical

Write: place value on bit line

•Addresses are time-

Notice pin counts:

RAS t RAS t Prechg RAS t RAS Prechg

Access time Cycle time Data hold from RAS.

Mask pro- Very At the factory Weeks (turn around) N/A

PROM Inexpensive Once, by end Seconds N/A

EPROM Moderate Many times Seconds 20 minutes

Flash Expensive Many times 100 us. 1s, large

EEPROM Very Many times 100 us. 10 ms,

Multiple chip selects ease the assembly of

P chips expand word size from s bits to p x s bits.

Word size remains at s bits.

•CAS is used to enable

•Use one 2-D array for

Refresh counter Address

Main Memory Address: Word Module No.

This provides successive words in successive modules.

Timing: Bus Read module 0 Write module 3 Module 0

Module 3 Module 3 write

With interleaving of 2k modules, and tb < tb/2k, it is possible to get a 2k-fold

For all accesses:

The next slide shows the access process in more detail --

“Hidden refresh” cycle. A normal cycle would exclude the

•Address bus drivers turn-on time: 40 ns.

•On-chip memory read access time: 80 ns

TOTAL MEMORY READ ACCESS TIME: 280 ns.

Moral: 70ns chips to not necessarily provide 70ns access time!

Spatial locality: if a given memory location is referenced, those locations

two adjacent levels in the hierarchy

Consider the C for loop:

for ((I=0); I<n; I++)

•The item of commerce between any two levels is the block.

•Blocks may/will differ in size at different levels in the hierarchy.