2. Chapter Two Memory organisation

3. Examples of operating system n Windows 95/98/2000, Windows NT n Unix, Linux, n VAX/VMS IBM MVS n Novell Netware and Windows NT: Client/Server model

4. Objective n how Virtual Memory works; n the limitation and inaccuracy of data n the internal structure between DRAM and Video RAM; and n the function and location of Cache memory.

5. Overlay n Divide the program into several manageable smaller overlays to be stored on the disk n Load the first overlay into the memory and run for a while. n When it finishes, it reads the next overlay into the memory to perform program execution.

6. Idea of virtual memory n Allow programmers to write programs without paying attention to the size of the program n Programs could be as large as necessary n run even though the whole program would not fit into the main memory.

7. Relationship between virtual and physical page

8. Hardware and software n Break program (code + data) into pages (page means fixed length data block.) n Keep all code/data pages on disk n Brings pages from disk into memory only as needed by the programme only.

9. To implement virtual memory, n we must be able to: n break code/data into pages n recognize when a given page is not present in memory n be able to bring the missing page into various places in main memory

10. Difference Between Virtual Address and Physical Address n Compiled addresses are called virtual addresses for this reason. n Physical address refers to the address in the main memory.

11. Mapping n Mapping is about the virtual addresses to physical addresses. n It is composed of two parts, page # and offset n Example in a 16 bit virtual address has: n 4 bit page # (can be changed for different physical frame) n 12 bit offset

12. Addressing n For example 12310h, n The page number is 0011, the first four digits, n The offset is 12310 - 12288 = 22.

13. Address Mapping

14. Virtual to Real addresses n to map one virtual address to several different locations in physical memory.

15. For example 12310 n The page number is 0011 in binary, which is 3 in decimal n The offset is 0000 0001 0110 in binary, which is 22 in decimal n If PAGE 3 was stored in page frame 0, the page is 0 offset = 22

16. Loading data from memory

17. Windows NT n All early PC using 8086 and 80286 used a segmented model. n A segmented addressing system divides physical memory (RAM) into units of continguious address, called segment. n the page size for Windows NT is chosen as 4K bytes

18. Windows NT disk address

19. Memory error Detection /Correction n deals with basic error detection and protection to ensure the data in the memory is accurate n invisible to software; and n set/checked by hardware only

20. 3 bit word with parity checking

21. Hamming Codes n Suppose there is an m-bit word, how many extra bits are required to detect and correct the single bit error. n 2^n >= 2m(n + 1) n Since n = m + r n 2^(m+r) >= 2^m (m+r+1) n 2^r >= m + r + 1 n it is more efficient to apply this error checking for large word size but not for small word size.

22. 6 bit word with one incorrect

23. Memory Chip Architectures n Main memory n Video memory n Main memory

24. DRAM and SRAM n DRAM uses capacities to hold the data and needs refreshment to store the data as a capacitor

25. Video memory n Video DRAM is a conventional DRAM with an additional 256 bit shift register on board. n This shift register is connected to help download a complete row of memory cells. n The shift register has its own output which can be controlled with separate input lines to the video DRAM.

26. Video RAM

27. Cache Memory n Cache is a French word which means “to hide”. n Cache memory refers to the memory that is invisible. n This helps to speed up the operation.

28. Cache Memory

29. characteristics for cache memory n It is usually built within a CPU/board chip to facilitate the frequently used commands. n Most frequently used code/data in memory is also kept in cache but invisible to software. n When program accesses this code/data, it comes from high speed cache rather than from slower main memory. n Cache memory is small, high speed memory usually on-chip or on same board as CPU.

30. Cache Design n size n how much will fit on a CPU/board? n at what point does increased size result in small improvements in hit rate? Issues

31. Cache Hit

32. Cache Miss

33. Software characteristic n Locality of addressing n Cache contents

34. Hit against cache size

35. Cache Performance n Mean access time = [cache memory access time] + (1-h)*[main memory access time] n h refers to the hit rate n Note that as h -> 0 the cache memory is a little worse than main memory alone.

36. Example n What is the effective memory speed of a cache-main memory system with the following characteristics? n main memory access time is 100 nsec n cache access time is 20 nsec n the hit rate is 95% u Using the formula, the effective memory speed is 20 + (1- 0.95)x100 = 25 nsec

37. Summary n Virtual Memory and Overlay n The memory protection using Hamming code n Cache memory was designed to enhance performance. n To measure how efficient the system is running, hit ratio is used.