5_15_25_35_45_5Chapter 5Large and Fast: Exploiting Memory HierarchyChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 2Memory Technology Static RAM (SRAM) 0.5ns – 2.5ns, $2000 – $5000 per GB Dynamic RAM (DRAM) 50ns – 70ns, $20 – $75 per GB Magnetic disk 5ms – 20ms, $0.20 – $2 per GB Ideal memory Access time of SRAM Capacity and cost/GB of disk§5.2 Memory TechnologiesDRAM Technology Data stored as a charge in a capacitor Single transistor used to access the charge Must periodically be refreshed Read contents and write back Performed on a DRAM “row”Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 3Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 4DRAM Generations050100150200250300'80 '83 '85 '89 '92 '96 '98 '00 '04 '07TracTcacYear Capacity $/GB1980 64Kbit $15000001983 256Kbit $5000001985 1Mbit $2000001989 4Mbit $500001992 16Mbit $150001996 64Mbit $100001998 128Mbit $40002000 256Mbit $10002004 512Mbit $2502007 1Gbit $50Chapter 6 — Storage and Other I/O Topics — 5Flash Storage Nonvolatile semiconductor storage 100× – 1000× faster than disk Smaller, lower power, more robust But more $/GB (between disk and DRAM)§6.4 Flash StorageChapter 6 — Storage and Other I/O Topics — 6Disk Storage Nonvolatile, rotating magnetic storage§6.3 Disk StorageChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 7Principle of Locality Programs access a small proportion of their address space at any time Temporal locality Items accessed recently are likely to be accessed again soon e.g., instructions in a loop, induction variables Spatial locality Items near those accessed recently are likely to be accessed soon E.g., sequential instruction access, array data§5.1 IntroductionChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 8Taking Advantage of Locality Memory hierarchy Store everything on disk Copy recently accessed (and nearby) items from disk to smaller DRAM memory Main memory Copy more recently accessed (and nearby) items from DRAM to smaller SRAM memory Cache memory attached to CPUChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 9Memory Hierarchy Levels Block (aka line): unit of copying May be multiple words If accessed data is present in upper level Hit: access satisfied by upper level Hit ratio: hits/accesses If accessed data is absent Miss: block copied from lower level Time taken: miss penalty Miss ratio: misses/accesses= 1 – hit ratio Then accessed data supplied from upper levelChapter 5Large and Fast: Exploiting Memory HierarchyChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 2Cache Memory Cache memory The level of the memory hierarchy closest to the CPU Given accesses X1, …, Xn–1, Xn§5.3 The Basics of Caches How do we know if the data is present? Where do we look?Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 3Direct Mapped Cache Location determined by address Direct mapped: only one choice (Block address) modulo (#Blocks in cache) #Blocks is a power of 2 Use low-order address bitsChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 4Tags and Valid Bits How do we know which particular block is stored in a cache location? Store block address as well as the data Actually, only need the high-order bits Called the tag What if there is no data in a location? Valid bit: 1 = present, 0 = not present Initially 0Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 5Cache Example 8-blocks, 1 word/block, direct mapped Initial stateIndex V Tag Data000 N001 N010 N011 N100 N101 N110 N111 NChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 6Cache ExampleIndex V Tag Data000 N001 N010 N011 N100 N101 N110 Y 10 Mem[10110]111 NWord addr Binary addr Hit/miss Cache block22 10 110 Miss 110Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 7Cache ExampleIndex V Tag Data000 N001 N010 Y 11 Mem[11010]011 N100 N101 N110 Y 10 Mem[10110]111 NWord addr Binary addr Hit/miss Cache block26 11 010 Miss 010Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 8Cache ExampleIndex V Tag Data000 N001 N010 Y 11 Mem[11010]011 N100 N101 N110 Y 10 Mem[10110]111 NWord addr Binary addr Hit/miss Cache block22 10 110 Hit 11026 11 010 Hit 010Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 9Cache ExampleIndex V Tag Data000 Y 10 Mem[10000]001 N010 Y 11 Mem[11010]011 Y 00 Mem[00011]100 N101 N110 Y 10 Mem[10110]111 NWord addr Binary addr Hit/miss Cache block16 10 000 Miss 0003 00 011 Miss 01116 10 000 Hit 000Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 10Cache ExampleIndex V Tag Data000 Y 10 Mem[10000]001 N010 Y 10 Mem[10010]011 Y 00 Mem[00011]100 N101 N110 Y 10 Mem[10110]111 NWord addr Binary addr Hit/miss Cache block18 10 010 Miss 010Chapter 5 — Large and Fast: Exploiting Memory Hierarchy — 11Address SubdivisionChapter 5Large and Fast: Exploiting Memory HierarchyChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 2Example: Larger Block Size 64 blocks, 16 bytes/block To what block number does address 1200 map? Block address = 1200/16 = 75 Block number = 75 modulo 64 = 11Tag Index Offset034910314 bits6 bits22 bitsChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 3Block Size Considerations Larger blocks should reduce miss rate Due to spatial locality But in a fixed-sized cache Larger blocks fewer of them More competition increased miss rate Larger blocks pollution Larger miss penalty Can override benefit of reduced miss rate Early restart and critical-word-first can helpChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 4Cache Misses On cache hit, CPU proceeds normally On cache miss Stall the CPU pipeline Fetch block from next level of hierarchy Instruction cache miss Restart instruction fetch Data cache miss Complete data accessChapter 5 — Large and Fast: Exploiting Memory Hierarchy — 5Write-Through On data-write hit, could just update the block in cache But then cache and memory would be inconsistent Write through: also update memory But makes writes take longer e.g., if base CPI = 1, 10% of instructions are stores, write to memory takes 100 cycles Effective CPI = 1 + 0.1×100 = 11 Solution: write buffer
View Full Document