KENYATTA UNIVERSITY

SIT 105 Computer Hardware

CAT 2

1. One challenge for architects is that the design created today will require several years of implementation,
verification, and testing before appearing on the market. This means that the architect must project what
the technology will be like several years in advance. Sometimes, this is difficult to do.

a. According to the trend in device scaling observed by Moore’s law,

the number of transistors on a chip in 2025 should be how many times the
number in 2015? Moore law states that the number of transistors on a microchip doubles about every two years
that mean it will be 8 billion transistors
b. The increase in clock rates once mirrored this trend. Had clock
rates continued to climb at the same rate as in the 2000s, approximately how
fast would clock rates be in 2025? Co mputers back in 2000 would double after every year back in the 2000 the
1ghz mark was archived meaning if the trend continued to double every year that would be 50 ghz clock speed

c. At the current rate of increase, what are the clock rates now projected to be in 2025?

d. What has limited the rate of growth of the clock rate, and what are
architects doing with the extra transistors now to increase performance?

Heat and transistors size has limited the growth of the clock speed the bigger the transistors ( the size of the
wires ) so architects reduces the size of the transistors and adding more cores to CPU chip since increasing clock
speed consumes more electricity and produces a lot of heat adding more cores to a CPU helps to manage that and
hence makes a CPU more efficient

e. rate of growth for DRAM capacity has also slowed down. For
20 years, DRAM capacity improved by 60% each year. That rate dropped to
40% each year and now improvement is 25 to 40% per year. If this trend continues, what will be the approximate
rate of growth for DRAM capacity by
2030?

2. Each processor has a number of address, data and control lines to connect to memory and
peripheral devices. If you design a board with all these components, what is your
measurement point?

3. If more performance is needed, faster processors can be used or you can enhance bus
bandwidth by going from 8 to 16, 32 and even 64 bit data transfers. The use of cache may
enhance overall system performance. However, there is an end to what technology can do
for you today and if one processor is not enough, why not use several?
 You cant have several processors on one board it will consume more electric and the latency or data transfer
between the CPU and the other components will be and heat will be also a major issue

4. A designer wants to use hard disk as main memory instead of semiconductor memory.
This idea can be easily rejected since the hard disk is a sequential access memory,
however if we accept a large access time, is it possible to build a RAM around the hard
disk with, necessary hardware and/or software so that it can be used atleast for
experimental purposes. How Can this be implemented?

This can be implemented with the help of the operating system with virtual memory whrere it makes a page file in
the primary partion and use the hdd as ram the downside being it will be slower compared to ram