15-447 Computer ArchitectureFall 2007 © November 12th, 2007 Majd F. Sakr CS-447– Computer Architecture.

15-447 Computer ArchitectureFall 2007 © November 12th, 2007 Majd F. Sakr msakr@qatar.cmu.edu www.qatar.cmu.edu/~msakr/15447-f07/ CS-447– Computer Architecture M,W 10-11:20am Lecture 20 Cache Memories

15-447 Computer ArchitectureFall 2007 © Processor-DRAM Memory Gap (latency) µProc 60%/yr. (2X/1.5yr) DRAM 9%/yr. (2X/10 yrs) 1 10 100 1000 19801981198319841985198619871988198919901991199219931994199519961997199819992000 DRAM CPU 1982 Processor-Memory Performance Gap: (grows 50% / year) Performance Time “Moore’s Law”

15-447 Computer ArchitectureFall 2007 © Locality °A principle that makes having a memory hierarchy a good idea °If an item is referenced, temporal locality: it will tend to be referenced again soon spatial locality: nearby items will tend to be referenced soon.

15-447 Computer ArchitectureFall 2007 © A View of the Memory Hierarchy Regs L2 Cache Memory Disk Tape Instr. Operands Blocks Pages Files Upper Level Lower Level Faster Larger Cache Blocks

15-447 Computer ArchitectureFall 2007 © Our initial focus: two levels (upper, lower) block: minimum unit of data hit: data requested is in the upper level miss: data requested is not in the upper level Why Does Code Have Locality?

15-447 Computer ArchitectureFall 2007 © Cache Design °How do we organize cache? °Where does each memory address map to? (Remember that cache is subset of memory, so multiple memory addresses map to the same cache location.) °How do we know which elements are in cache? °How do we quickly locate them?

15-447 Computer ArchitectureFall 2007 © Block Size Tradeoff (1/3) °Benefits of Larger Block Size Spatial Locality: if we access a given word, we’re likely to access other nearby words soon Very applicable with Stored-Program Concept: if we execute a given instruction, it’s likely that we’ll execute the next few as well Works nicely in sequential array accesses too

15-447 Computer ArchitectureFall 2007 © Block Size Tradeoff (2/3) °Drawbacks of Larger Block Size Larger block size means larger miss penalty -on a miss, takes longer time to load a new block from next level If block size is too big relative to cache size, then there are too few blocks -Result: miss rate goes up °In general, minimize Average Access Time = Hit Time x Hit Rate + Miss Penalty x Miss Rate

15-447 Computer ArchitectureFall 2007 © Block Size Tradeoff (3/3) °Hit Time = time to find and retrieve data from current level cache °Miss Penalty = average time to retrieve data on a current level miss (includes the possibility of misses on successive levels of memory hierarchy) °Hit Rate = % of requests that are found in current level cache °Miss Rate = 1 - Hit Rate

15-447 Computer ArchitectureFall 2007 © Block Size Tradeoff Conclusions Miss Penalty Block Size Increased Miss Penalty & Miss Rate Average Access Time Block Size Exploits Spatial Locality Fewer blocks: compromises temporal locality Miss Rate Block Size

15-447 Computer ArchitectureFall 2007 © Direct Mapped Cache °Mapping: address is modulo the number of blocks in the cache

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache (1/2) °In a direct-mapped cache, each memory address is associated with one possible block within the cache Therefore, we only need to look in a single location in the cache for the data if it exists in the cache Block is the unit of transfer between cache and memory

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache (2/2) °Cache Location 0 can be occupied by data from: Memory location 0, 4, 8,... 4 blocks => any memory location that is multiple of 4 Memory Memory Address 0 1 2 3 4 5 6 7 8 9 A B C D E F 4 Byte Direct Mapped Cache Cache Index 0 1 2 3

15-447 Computer ArchitectureFall 2007 © Issues with Direct-Mapped °Since multiple memory addresses map to same cache index, how do we tell which one is in there? °What if we have a block size > 1 byte? °Answer: divide memory address into three fields ttttttttttttttttt iiiiiiiiii oooo tagindexbyte to checkto offset if have selectwithin correct blockblockblock

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache Terminology °All fields are read as unsigned integers. °Index: specifies the cache index (which “row” of the cache we should look in) °Offset: once we’ve found correct block, specifies which byte within the block we want °Tag: the remaining bits after offset and index are determined; these are used to distinguish between all the memory addresses that map to the same location

15-447 Computer ArchitectureFall 2007 © Caching Terminology °When we try to read memory, 3 things can happen: 1.cache hit: cache block is valid and contains proper address, so read desired word 2.cache miss: nothing in cache in appropriate block, so fetch from memory 3.cache miss, block replacement: wrong data is in cache at appropriate block, so discard it and fetch desired data from memory (cache always copy)

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache Example (1/3) °Suppose we have a 16KB of data in a direct-mapped cache with 4 word blocks °Determine the size of the tag, index and offset fields if we’re using a 32-bit architecture °Offset need to specify correct byte within a block block contains 4 words = 16 bytes = 2 4 bytes need 4 bits to specify correct byte

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache Example (2/3) °Index: (~index into an “array of blocks”) need to specify correct row in cache cache contains 16 KB = 2 14 bytes block contains 2 4 bytes (4 words) # blocks/cache =bytes/cache bytes/block = 2 14 bytes/cache 2 4 bytes/block =2 10 blocks/cache need 10 bits to specify this many rows

15-447 Computer ArchitectureFall 2007 © Direct-Mapped Cache Example (3/3) °Tag: use remaining bits as tag tag length = addr length – offset - index = 32 - 4 - 10 bits = 18 bits so tag is leftmost 18 bits of memory address °Why not full 32 bit address as tag? All bytes within block need same address (4b) Index must be same for every address within a block, so it’s redundant in tag check, thus can leave off to save memory (here 10 bits)

15-447 Computer ArchitectureFall 2007 © Direct Mapped Cache (for MIPS)

15-447 Computer ArchitectureFall 2007 © °Read hits this is what we want! °Read misses stall the CPU, fetch block from memory, deliver to cache, restart Hits vs. Misses

15-447 Computer ArchitectureFall 2007 © °Write hits: can replace data in cache and memory (write-through) write the data only into the cache (write- back the cache later) °Write misses: read the entire block into the cache, then write the word Hits vs. Misses

15-447 Computer ArchitectureFall 2007 © °Make reading multiple words easier by using banks of memory Hardware Issues

15-447 Computer ArchitectureFall 2007 © °Increasing the block size tends to decrease miss rate: Performance

15-447 Computer ArchitectureFall 2007 © Performance Use split caches because there is more spatial locality in code:

15-447 Computer ArchitectureFall 2007 © Performance °Simplified model: execution time = (execution cycles + stall cycles)  cycle time stall cycles = # of instructions  miss ratio  miss penalty °Two ways of improving performance: decreasing the miss ratio decreasing the miss penalty What happens if we increase block size?

15-447 Computer ArchitectureFall 2007 © Cache Complexities °Not always easy to understand implications of caches: Theoretical behavior of Radix sort vs. Quicksort Observed behavior of Radix sort vs. Quicksort

15-447 Computer ArchitectureFall 2007 © Cache Complexities °Here is why:

15-447 Computer ArchitectureFall 2007 © Cache Complexities °Memory system performance is often critical factor multilevel caches, pipelined processors, make it harder to predict outcomes Compiler optimizations to increase locality sometimes hurt ILP °Difficult to predict best algorithm: need experimental data

15-447 Computer ArchitectureFall 2007 © Summary °The :

15-447 Computer ArchitectureFall 2007 © Our initial focus: two levels (upper, lower) block: minimum unit of data hit: data requested is in the upper level miss: data requested is not in the upper level Why Does Code Have Locality?

15-447 Computer ArchitectureFall 2007 © °Two issues: How do we know if a data item is in the cache? If it is, how do we find it? °Our first example: block size is one word of data "direct mapped" For each item of data at the lower level, there is exactly one location in the cache where it might be. e.g., lots of items at the lower level share locations in the upper level Cache