1 / 20

Memory (RAM) Organization

Memory (RAM) Organization. Each location is addressable Addresses are binary numbers Addresses used at different granularities each bit is possible, but not very likely each byte is possible, but seldom used (today) typical usage is a “word” of memory

loretta
Télécharger la présentation

Memory (RAM) Organization

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Memory (RAM) Organization • Each location is addressable • Addresses are binary numbers • Addresses used at different granularities • each bit is possible, but not very likely • each byte is possible, but seldom used (today) • typical usage is a “word” of memory • Each CPU designed with a particular word size (e.g. 8-bit, 16-bit, 32-bit, 64-bit)

  2. 96 bits = How Many Words?

  3. Memory • Latches: 1-bit memory circuits • SR latch (S=Set, R=Reset) • Set/Reset control signals • Memory value and its complement as outputs • D latch (D = Data) • like SR, but avoids problem of S = R = 1 • S comes from D; R comes from D’ • Clocked latch: clock input as enable signal

  4. An SR Latch

  5. A Clocked D Latch

  6. Flip-Flops • Flip-flops have same function as latches • Flip-flops are edge-triggered • Latches are level-triggered • Flip-flops typically used in industry

  7. D Flip-Flop

  8. Registers • Essentially a group of flip-flops • Common control signals • Different data signals • Distinct from main memory and cache • Typically found at the heart of the CPU

  9. 2-bit, 8-bit Memory Chips

  10. A 4 x 3 Memory Circuit

  11. Memory Circuit Notes • Data input lines (typically word size) • Data output lines (also word size) • Address lines • Read/write indicator line • Output enable buffers (disconnects bus) • Too many address lines? • Send 'em in a piece at a time

  12. Split Address Handling

  13. Selecting Row and Address Same two address lines that choose a row (RAS) sent in later to choose one of four columns (CAS)

  14. Real Memory Chips

  15. Chips and Clips • Multiple chips in a series • Chip select/enable choose among them • highest order address bit(s) do the select • e.g. memory address is 32 bits • one memory chip takes 28 bits (14x14) • remaining 4 bits chooses which chip • Control signals: • WE: reading/writing • OE: output enable (for tri-state buffer)

  16. RAM Chips • SRAM: flip-flops • S = static (i.e. “power’s always on”) • DRAM: capacitors • less circuitry/space/power, but slower • D = dynamic (i.e. “refreshed periodically”) • EDO: two-stage pipeline • FPM: fast page mode (cache nearby words) • SDRAM: S=synchronized with bus, CPU • DDR: double data rate (access on rise, fall)

  17. RAM and Cache

  18. Cache Memory • Basic idea • very fast, but more expensive memory • closer and smaller than main memory • typically multiple levels, some inside CPU package • used to save frequently used data/instructions • Problem • how to decide what to put in cache? • locality principle: reuse data, use nearby instructions • cache typically split for data and instructions

  19. Blocks, Tags, and Lines

  20. Cache Design and Performance • Mean access time = c + (1 - h) * m • Where c is cache access time (e.g. 20ns) • And h is the hit ratio (e.g. 32% or 0.32) • And m is main memory access time (e.g. 50ns) • Cache lines: fixed-size blocks and tags • Direct-mapped versus (set-)associative

More Related