1 / 24

Memory

Memory. Computer Architecture CS 215. Increasing performance & cost. Memory Hierarchy. Registers Cache Main memory Secondary storage Off-line storage. Random Access Memory. SRAM (Static RAM) Based on flip-flops Content persists as long as current is present DRAM (Dynamic RAM)

sebille
Télécharger la présentation

Memory

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 Computer ArchitectureCS 215

  2. Increasing performance & cost Memory Hierarchy Registers Cache Main memory Secondary storage Off-line storage

  3. Random Access Memory • SRAM (Static RAM) • Based on flip-flops • Content persists as long as current is present • DRAM (Dynamic RAM) • Employs a capacitor • Stores minute charge for each bit • Charge dissipates requiring periodic refresh • Susceptible to premature discharging

  4. Chip Organization

  5. New RAM from Old • Two 4-word by 4-bit RAMs make up an 8-Word by 4-Bit RAM

  6. CommercialMemory Modules • Address Inputs A0-A9 • Column-Address Strobe CAS • Data In/Data Out DQ1-DQ8 • No Connection NC • Row-Address Strobe RAS • 5-V Supply Vcc • Ground Vss • Write Enable W

  7. Read-Only Memory • Note how each decoder output is connected to specific OR gates • What is value of Q when A is 10?

  8. Cache Memory • Associative Mapped • Direct Mapped • Set Associative Mapped • Performance • Hit Ratios • Effective Access Times • Multilevel • Management

  9. AssociativeMapped • Assume … • 232 word memory • 227 blocks of 25 = 32 words per block • Cache consists of 214 slots • If the addressed word is in the cache, it will be found in a slot that has tag made up of the 27 most significant bits of the address • If the addressed word is not in the cache, then the block brought into an available slot in the cache from the main memory, and the memory reference is then satisfied from the cache.

  10. AssociativeMapped • Replacement policy(Associative and set-associative mapping schemes, and also for virtual memory) • Least recently used (LRU) • First-in/first-out (FIFO) • Least frequently used (LFU) • Random • Optimal (used for analysis only – look backward in time and re-verse engineer the best possible strategy for a particular sequence of memory references.)

  11. DirectMapped • Each Block can be mapped to only one slot, but … • Each slot can be maped to more than one block

  12. Example Cache Main Slot 0 Block 0 3 0-255 - Slot 1 Block 1 256-511 - Block 2 512-767 Slot 2 Block 3 768-1023 • Access M(776) • Choose Slot 0 • Copy Block 3 into Slot 0 Block 4 1024-1279 Block 5 1280-1535 Block 6 1536-1791 Block 7 1792-2047

  13. Example Cache Main Slot 0 Block 0 3 0-255 5 Slot 1 Block 1 256-511 - Block 2 512-767 Slot 2 Block 3 768-1023 • Access M(1296) • Choose Slot 1 • Copy Block 5 into Slot 1 Block 4 1024-1279 Block 5 1280-1535 Block 6 1536-1791 Block 7 1792-2047

  14. Example Cache Main Slot 0 Block 0 3 0-255 5 Slot 1 Block 1 256-511 0 Block 2 512-767 Slot 2 Block 3 768-1023 • Access M(64) • Choose Slot 2 • Copy Block 0 into Slot 2 Block 4 1024-1279 Block 5 1280-1535 Block 6 1536-1791 Block 7 1792-2047

  15. Example Cache Main Slot 0 Block 0 3 0-255 5 Slot 1 Block 1 256-511 0 Block 2 512-767 Slot 2 ? Block 3 768-1023 • Access M(1536) • Which slot should be used? Block 4 1024-1279 Block 5 1280-1535 Block 6 1536-1791 Block 7 1792-2047

  16. Hit Ratios • Hit - Referenced word is in cache memory (Block previously loaded) • Miss - Referenced word is not in cache (Block needs to be loaded)

  17. Effective Access Times

  18. Multilevel

  19. Virtual Memory • Overlays • Paging • Fragmentation • Virtual vs. Cache

  20. Overlays

  21. Overlays Main Routine Subroutine A

  22. Overlays Main Routine Subroutine B Subroutine C

  23. Virtual Memory • Virtual memory is stored in a hard disk image • The physical memory holds a small number of virtual pages in physical page frames

  24. Paging • Page table maps between virtual memory and physical memory • Column 1 – Is page occupied? • Column 2 – Hard disk starting address • Column 3 – Physical address prefix

More Related