Theme: CPU and Memory

Theme: CPU and Memory

1

PART III

COMPUTER HARDWARE

Theme: CPU and Memory

  1. Functional Unit: Central Processing Unit (CPU) = ALU + Control
  2. Physical Unit: Processor
  3. Evaluation Criteria: Machine cycle time and Processor Architecture
  1. What do we mean by “Machine Cycle Time”?

A computer executes one instruction at a time when a program is executed under the control of a control unit. A control unit runs in cycles. Within each cycle, it completes four activities:fetch, decode, execute, and store the result. The time duration for a control unit to complete a cycle is called a CPU cycle time (control cycle time, machine cycle time, or processor cycle time).

  1. What is Hz then?

By completing millions (and eventually billions, trillions, and so on) of these machine cycles per second, computers execute their instructions very rapidly and work their magic. The number of instructions executed is normally used to measure how fast a CPU (control unit in this case) works. The unit is Hertz (Hz) which is equal to 1/second. Normally we would see Giga Hertz (GHz) which is 109 instructions per second.

  1. Download MarieSim Computer Machine Simulator from and clicking “Student Resources”and then the “Download MarieSim” button. After the “.zip” file is stored at “My Documents”, right click on this file and choose “Extract All’.
  2. In the folder “MarieSim” extracted by the “extract all” tool, you will find the MarieGuide and the QuickGuide for more information about how to use the machine simulator. Answer the following questions:

(1)What is MarieSim used for? ______

(2)What functional unit does Accumulator (AC) Register belong to? ______

(3)What do we call the “statement” such as “Load X”? ______

(4)What kind of language is the about “statement” written? ______

(5)What software is used to convert “statements” mentioned above to machine code? ______

  1. What are the two major processor design principles of a computer?

The two computer processor design principles are CISC and RISC, i.e., Complex Instruction Set Computer (CISC) and Reduced Instruction Set Computer (RISC).

  1. What is CISC Architecture versus RISC Architecture? Please refer to CISC vs. RISC an article posted on the web site at

details. Every computer processor can only accept a finite set of instructions known as an Instruction Set. There are two design philosophies about the design of a processor. One uses a large set of instructions, while another uses a small set of instructions. The difference can be explained with an example. For executing a simple multiplication of two operands in memory two spaces A(2:3) and B(5:2) and store the result into another memory space A, a CISC machine would require a programmer to write only “MULT 2:3, 5:2. It is easy to write. However, the compiler would have to generate a long instruction which requires say several cycles to complete one instruction. For RISC machines, a programmer would have to code:

LOAD A, 2:3
LOAD B, 5:2
PROD A, B
STORE 2:3, A

But each instruction can be executed in one cycle.

  1. Which one is better?

CISC machines need less number of instructions to complete a task. But the execution of an instruction takes several cycles. RISC machines are just the opposite—they take more instructions to complete a task withone cycle for executing each instruction. The final CPU performance really depends on the type of applications. Most computers started with RISC architecture and some started to switch to CISC architecture. For example, Apple uses Motorola chips for its computers until 2006. (Apple computers switched to Intel-based CISC processors by 2007.)For more information about the comparison of CISC and RISC, refer to It has become more difficult to classify today’s processors as either CISC or RISC since many use hybrid technologies.

  1. How is computer performance measured?

Following information was taken from the “CISC vs. RISC” article.

“The following equation is commonly used for expressing a computer's performance ability:

The CISC approach attempts to minimize the number of instructions per program, sacrificing the number of cycles per instruction. RISC does the opposite, reducing the cycles per instruction at the cost of the number of instructions per program.”

4. Main Memory: Random Access Memory (RAM)

a. DRAM (Dynamic RAM) – SDRAM (Synchronous DRAM), DDR SDRAM

(Dual Data-Rate Synchronous DRAM)

b. SRAM (Static RAM) – faster than DRAM but more expensive

5. Dual Inline Memory Modules (DIMM) – A memory chips installed with matched pairs of memory modules. DIMM is used to replace SIMM (Single Inline Memory Modules) to support 64-bit buses.

6. Buses: A bus is an electrical wire in the computer’s circuitry. You can consider a bus to be the major data highway inside a computer for connecting various devices: a processor, memory, video card, or sound cards. Some examples are ISA, EISA, or PCI.

Top View