+ Chapter 4

Cache Memory
 Table 4.1
Key Characteristics of Computer Memory Systems
+
Characteristics of Memory Systems

 Location
 Refers to whether memory is internal and external to the computer
 Internal memory is often equated with main memory
 Processor requires its own local memory, in the form of registers
 Cache is another form of internal memory
 External memory consists of peripheral storage devices that are accessible to the
processor via I/O controllers

 Capacity
 Memory is typically expressed in terms of bytes

 Unit of transfer
 For internal memory the unit of transfer is equal to the number of electrical lines
into and out of the memory module
Method of Accessing Units of Data
Sequential Random
Direct access Associative
access access

Each addressable location in

A word is retrieved based on
Memory is organized into Involves a shared read-write memory has a unique,
a portion of its contents
units of data called records mechanism physically wired-in
rather than its address
addressing mechanism

Each location has its own

The time to access a given
Individual blocks or records addressing mechanism and
Access must be made in a location is independent of
have a unique address based retrieval time is constant
specific linear sequence the sequence of prior
on physical location independent of location or
accesses and is constant
prior access patterns

Any location can be selected

Cache memories may
Access time is variable Access time is variable at random and directly
employ associative access
addressed and accessed

Main memory and some

cache systems are random
access
Capacity and Performance:

The two most important characteristics of memory

Three performance parameters are used:

Memory cycle time

Access time (latency) • Access time plus any additional time
Transfer rate
• For random-access memory it is the time required before second access can • The rate at which data can be transferred
it takes to perform a read or write commence into or out of a memory unit
operation • Additional time may be required for • For random-access memory it is equal to
• For non-random-access memory it is the transients to die out on signal lines or to 1/(cycle time)
time it takes to position the read-write regenerate data if they are read
mechanism at the desired location destructively
• Concerned with the system bus, not the
processor
+ Memory
 The most common forms are:
 Semiconductor memory
 Magnetic surface memory
 Optical
 Magneto-optical

 Several physical characteristics of data storage are important:

 Volatile memory
 Information decays naturally or is lost when electrical power is switched off
 Nonvolatile memory
 Once recorded, information remains without deterioration until deliberately changed
 No electrical power is needed to retain information
 Magnetic-surface memories
 Are nonvolatile
 Semiconductor memory
 May be either volatile or nonvolatile
 Nonerasable memory
 Cannot be altered, except by destroying the storage unit
 Semiconductor memory of this type is known as read-only memory (ROM)

 For random-access memory the organization is a key design issue

 Organization refers to the physical arrangement of bits to form words
+
Memory Hierarchy

 Design constraints on a computer’s memory can be summed up by

three questions:
 How much, how fast, how expensive

 There is a trade-off among capacity, access time, and cost

 Faster access time, greater cost per bit
 Greater capacity, smaller cost per bit
 Greater capacity, slower access time

 The way out of the memory dilemma is not to rely on a single

memory component or technology, but to employ a memory
hierarchy
+
Memory
 The use of three levels exploits the fact that semiconductor memory
comes in a variety of types which differ in speed and cost

 Data are stored more permanently on external mass storage devices

 External, nonvolatile memory is also referred to as secondary

memory or auxiliary memory

 Disk cache
 A portion of main memory can be used as a buffer to hold data temporarily
that is to be read out to disk
 A few large transfers of data can be used instead of many small transfers of
data
 Data can be retrieved rapidly from the software cache rather than slowly
from the disk
 Table 4.2
Elements of Cache Design
+
Cache Addresses
Virtual Memory
 Virtual memory
 Facility that allows programs to address memory from a logical point of
view, without regard to the amount of main memory physically available
 When used, the address fields of machine instructions contain virtual
addresses
 For reads to and writes from main memory, a hardware memory
management unit (MMU) translates each virtual address into a physical
address in main memory
Mapping Function
 Because there are fewer cache lines than main memory blocks, an
algorithm is needed for mapping main memory blocks into cache
lines

 Three techniques can be used:

Direct Associative Set Associative

• The simplest technique • Permits each main memory • A compromise that exhibits
• Maps each block of main block to be loaded into any the strengths of both the
memory into only one line of the cache direct and associative
possible cache line • The cache control logic approaches while reducing
their disadvantages
interprets a memory address
simply as a Tag and a Word
field
• To determine whether a block
is in the cache, the cache
control logic must
simultaneously examine
every line’s Tag for a match
+
Direct Mapping Summary

 Address length = (s + w) bits

 Number of addressable units = 2s+w words or bytes

 Block size = line size = 2w words or bytes

 Number of blocks in main memory = 2s+ w/2w = 2s

 Number of lines in cache = m = 2r

 Size of tag = (s – r) bits

+
Victim Cache

 Originally proposed as an approach to reduce the conflict misses of

direct mapped caches without affecting its fast access time

 Fully associative cache

 Typical size is 4 to 16 cache lines

 Residing between direct mapped L1 cache and the next level of

memory
+
Associative Mapping Summary

 Address length = (s + w) bits

 Number of addressable units = 2s+w words or bytes

 Block size = line size = 2w words or bytes

 Number of blocks in main memory = 2s+ w/2w = 2s

 Number of lines in cache = undetermined

 Size of tag = s bits

+
Set Associative Mapping

 Compromise that exhibits the strengths of both the direct and

associative approaches while reducing their disadvantages

 Cache consists of a number of sets

 Each set contains a number of lines

 A given block maps to any line in a given set

 e.g. 2 lines per set

 2 way associative mapping
 A given block can be in one of 2 lines in only one set
+
Set Associative Mapping Summary
 Address length = (s + w) bits

 Number of addressable units = 2 s+w words or bytes

 Block size = line size = 2w words or bytes

 Number of blocks in main memory = 2 s+w/2w=2s

 Number of lines in set = k

 Number of sets = v = 2d

 Number of lines in cache = m=kv = k * 2 d

 Size of cache = k * 2d+w words or bytes

 Size of tag = (s – d) bits

+
Replacement Algorithms

 Once the cache has been filled, when a new block is brought into the
cache, one of the existing blocks must be replaced

 For direct mapping there is only one possible line for any particular
block and no choice is possible

 For the associative and set-associative techniques a replacement

algorithm is needed

 To achieve high speed, an algorithm must be implemented in

hardware
+ The most common replacement
algorithms are:
 Least recently used (LRU)
 Most effective
 Replace that block in the set that has been in the cache longest with no
reference to it
 Because of its simplicity of implementation, LRU is the most popular
replacement algorithm

 First-in-first-out (FIFO)
 Replace that block in the set that has been in the cache longest
 Easily implemented as a round-robin or circular buffer technique

 Least frequently used (LFU)

 Replace that block in the set that has experienced the fewest references
 Could be implemented by associating a counter with each line
 Write Policy

When a block that is resident in the

There are two problems to contend
cache is to be replaced there are two
with:
cases to consider:

If the old block in the cache has not been altered

More than one device may have access to main
then it may be overwritten with a new block
memory
without first writing out the old block

If at least one write operation has been A more complex problem occurs when
performed on a word in that line of the cache multiple processors are attached to the same
then main memory must be updated by writing bus and each processor has its own local cache
the line of cache out to the block of memory - if a word is altered in one cache it could
before bringing in the new block conceivably invalidate a word in other caches
+
Write Through
and Write Back
 Write through
 Simplest technique
 All write operations are made to main memory as well as to the cache
 The main disadvantage of this technique is that it generates substantial memory
traffic and may create a bottleneck

 Write back
 Minimizes memory writes
 Updates are made only in the cache
 Portions of main memory are invalid and hence accesses by I/O modules can be
allowed only through the cache
 This makes for complex circuitry and a potential bottleneck
 Line Size
When a block of data Two specific effects
is retrieved and come into play:
placed in the cache • Larger blocks reduce the
not only the desired As the block size number of blocks that fit into a
word but also some increases more useful cache
• As a block becomes larger each
number of adjacent data are brought into additional word is farther from
words are retrieved the cache the requested word

As the block size The hit ratio will

increases the hit ratio begin to decrease as
will at first increase the block becomes
because of the bigger and the
principle of locality probability of using
the newly fetched
information becomes
less than the
probability of reusing
the information that
has to be replaced
+
Multilevel Caches
 As logic density has increased it has become possible to have a cache on the
same chip as the processor

 The on-chip cache reduces the processor’s external bus activity and speeds up
execution time and increases overall system performance
 When the requested instruction or data is found in the on-chip cache, the bus access is
eliminated
 On-chip cache accesses will complete appreciably faster than would even zero-wait
state bus cycles
 During this period the bus is free to support other transfers

 Two-level cache:
 Internal cache designated as level 1 (L1)
 External cache designated as level 2 (L2)

 Potential savings due to the use of an L2 cache depends on the hit rates in both
the L1 and L2 caches

 The use of multilevel caches complicates all of the design issues related to
caches, including size, replacement algorithm, and write policy
+
Unified Versus Split Caches
 Has become common to split cache:
 One dedicated to instructions
 One dedicated to data
 Both exist at the same level, typically as two L1 caches

 Advantages of unified cache:

 Higher hit rate
 Balances load of instruction and data fetches automatically
 Only one cache needs to be designed and implemented

 Trend is toward split caches at the L1 and unified caches for higher levels

 Advantages of split cache:

 Eliminates cache contention between instruction fetch/decode unit and execution
unit
 Important in pipelining
Table 4.4

Intel
Cache
Evolution

(Table is on page 150

in the textbook.)