581 views
0 0 votes

Consider a demand-paging system with a paging disk that has an average access and transfer time of 20 milliseconds. Addresses are translated through a page table in main memory, with an access time of 1 microsecond per memory access. Thus, each memory reference through the page table takes two accesses. To improve this time, we have added an associative memory that reduces access time to one memory reference if the page-table entry is in the associative memory. Assume that 80 percent of the accesses are in the associative memory and that, of those remaining, 10 percent (or 2 percent of the total) cause page faults. The effective memory access time is _____ micro-seconds (Round off to 1 decimal place)

1 Answer

1 1 vote

Average access and transfer time = 20 ms = 20000 µs

Address Translation time = 1 µs

% of access in the associated memory = 80% = 0.8

% of page faults = 2% = 0.02

[We have to find the effective access time.]

EMAT(Effective Memory Access Time) = Time taken by 80% accesses in the memory + Time taken by 18% accesses not in the memory + Time taken by 2% accesses not in the memory causing page fault 

 = 0.8(1µs) + 0.18(2µs) + 0.02(20000µs + 2µs)

 = 0.8µs + 0.36µs + 400.02µs = 401.2µs

Hence Answer : 401.2µs

Position:
Show:

Related questions

3 3 votes
1 1 answer
111
111 views
GO Classes asked Aug 26
111 views
A demand-paging system has $3$ physical page frames.Consider the reference string:$\text{A, B, C, D, B, A, B, A, D, C}$Starting with empty memory, determine the total num...
1 1 vote
1 1 answer
86
86 views
GO Classes asked Aug 13
86 views
Suppose a $32K \times 8K$ matrix $A$ with $1$-byte elements is stored in row-major order in virtual memory.Assume:Only this program occupies physical memory. The matrix b...
2 2 votes
2 2 answers
126
126 views
GO Classes asked Aug 11
126 views
An operating system can use page tables and exceptions to perform allocation on demand, where it does not allocate physical memory for a program until the program tries t...