UGC NET CSE | June 2012 | Part 2 | Question: 32

Question

Cached and interleaved memories are ways of speeding up memory access between CPU’s and slower RAM. Which memory models are best suited (i.e. improves the performance most) for which programs ?

1. Cached memory is best suited for small loops.
2. Interleaved memory is best suited for small loops
3. Interleaved memory is best suited for large sequential code.
4. Cached memory is best suited for large sequential code.
   • A. I and II are true
   • B. I and III are true
   • C. IV and II are true
   • D. IV and III are true

Answer 1

Answer is B) (i) and (iii) are true

Compared to the processor speed, the speed of the primary memory is slow. Cache memory is a small memory which sits in between the processor and primary memory and fetches information to the processor at a much higher speed or it makes it appear so. Caching can be effective based on a property of computer programs called locality of reference. Analysis of program show that the majority of the execution time is spent around a small part of the program may be a simple loop,nested loop or a few functions. The rest of the program is accessed infrequently. There is something called temporal locality and spatial locality also which we need to know when we talk about cache. But cache memory is ideally suited for small loops. 

Interleaved memory is a technique for increasing the speed of RAM. Here multiple memory chips are grouped together to form what are known as banks.Each of them take turns for supplying data. An interleaved memory with "n" banks is said to be n-way interleaved. Macintosh systems are considered to be one using memory interleaving.
So the answer for this question is option B.

Answer 2

Size of Secondary Memory > Size of Main Memory > Size of Cache.

So, Cache would be suitable for holding relatively smaller pieces of code, while Interleaved Memory (a special logical architecture of Main Memory) is able to hold relatively larger pieces of code.

I and III are correct.

Option B