• Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

Measurement of Processors Performance report. In the experiment, a testing code was developed in C Programming Language. The design involved filling up each memory hierarchy level with data, then by noting down the time taken to access each level, the ban

Extracts from this document...

Introduction

Measurement of Processors Performance        image07.jpg

Measurement of Processors Performance        

Commissioned by THE UNIVERSITY OF AUCKLAND

Written by

Electrical and Electronic Department  (EED)

        COMPSYS 304 – Computer Architecture    

10. 05


Table of Contents

Introduction

1.Experiment Overview

1.1Theoretical Background

1.2Testing Platform

2.Theoretical Prediction

2.1Hypothesis

2.2Estimation Calculation

3.Design Methodology

3.1Experimental Achievement

3.2Experimental Implementation

3.3Experimental Errors

4.Result

5.Discussion

6.Conclusion and Recommendation

7.List of References

Table of Figures

Figure 1 Memory Hierarchy

Figure 2 Design Strategy

Figure 3 Graph of Data Transfer Rate versus Array Size for Sequential Access for Core i7 975 EE

Figure 4 Graph of Data Transfer Rate versus Array Size for Sequential Access for Core i5 750

Figure 5 Average Data Transfer Rate Comparison between Core i7 EE and Core i5 750

Figure 6 Comparison of Bandwidth between Core i5 750 and Core i7 975 EE

Introduction

In modern computers, the memory is always divided into multiple levels, which are structured into a memory hierarchy (Figure 1). In the memory hierarchy, each level is distinguished by the response time with the top-most level being the fastest. In most computers, this level is the processor registers which are often accessed in one CPU clock cycle. The next level is the Level 1 (L1) Cache, which is temporally and physically closest to the main processor. The third level is usually the Level 2 (L2) Cache and the forth level is the main memory [1].

...read more.

Middle

133

64

* Values are estimated based on research, no official documentation

  • Motherboard:               Asustek RAMPAGE II EXTREME Intel X58
  • RAM:Cosair 6GB Triple channel DDR3-1333
  • HD:Hitachi 1TB SATA2 3Gb/s
  • Graphic:ATI Radeon HD4870 x 2 2Gb DDR5
  • Case:Antec Nine Hundred Two
  • Power:Hyena True-Power 500W, 12V DUAL RAIL
  • OS:Windows Vista Ultimate
  1. Theoretical Prediction

  1. Hypothesis

For the sequential access to the memory, we expect to observe four distinct levels on the graph, corresponding to accessing the L1 cache, the L2 cache, the L3 cache and the main memory. The reason for this expectation is that the accessing time will increase significantly when there is a transition between different memory hierarchies. [6] For the random access to the memory, a graph of an exponentially decrease in the data transfer rate should be seen, as the current level of memory hierarchy is saturating. Then it should level off at a particular value since the memory size of the current level is exceeded.

  1. Estimation Calculation

  • Intel Core i7 975 Extreme Edition:

Bandwidth for L1 Cache = processor speed x bus width

                                                                = 3300 MHz x 32 bytes = 106,000* MB/s ± 10MB

                        Bandwidth for L2 Cache = processor speed x bus width

                                                                = 3300 MHz x 16 bytes = 52,800* MB/s ± 10MB

             Bandwidth for Main Memory = memory speed x bus width x 2

                                                  = 133 MHz x 8 bytes x 2 = 2,130* MB/s

...read more.

Conclusion

Alternatively this could have been done on Linux using gcc O4 which would run with the maximum optimization [9].Latency of the processor should be found by using 3rd party software and should be considered for the testing stage [10].Enough repetitions of the memory access should be performed to obtain more reliable results.
  1. List of References

  1. Comer, D. E. (2005) Essentials of Computer Architecture. Pearson/Prentice Hall, USA (New Jersey)
  2. Wikipedia (2009) from http://en.wikipedia.org/wiki/Memory_hierarchy

Retrieved 29 September 2009.

  1. Corei7ee (2009) from www.intel.com/design/corei7ee/documentation.htm

Retrieved 29 September 2009.

  1. Corei750 (2009) fromwww.intel.com/design/corei5_750/documentation.htm

   Retrieved 29 September 2009

  1. A.Swan(2008), MEMORY HIERARCHDESIGN ,from

http://www.cs.iastate.edu/~prabhu/Tutorial/CACHE/mem_title.htmlRetrieved 29 September 2009.

  1. V.Guistin(2004), Fast Data Dependence Analysis in a Multimedia Vectorizing Compiler,
    Proceedings of the 12th Euromicro Symposium on Parallel and Distributed Computing 2004,
    PDP 2004, February, 11–13, La Coruna, Spain, pp. 176–183. 2004.
  1. Assoc. Prof. John Morris and Dr. Morteza Biglari-Abhari(2009) “Computer architecture”, Lecture/class university of Auckland, New Zealand. Unpublished.
  1. David A. Patterson, John L. Hennessy(2009) Computer organization and design : the hardware/software interface ,4th ed.Amsterdam ; Boston : Elsevier Morgan Kaufmann, c2009
  1. Nils J. Nilsson(2007), computer architecture - A New Synthesis, Morgan Kaufmann Publishers, 2000.
  1. H.joshon(2006) Software Architecture and Design, from
    http://vorlon.case.edu/~flm/bsms/Research/ResearchDescriptions.pdf

Retrieved 29 September 2009

image08.jpg

...read more.

This student written piece of work is one of many that can be found in our University Degree Computer Science section.

Found what you're looking for?

  • Start learning 29% faster today
  • 150,000+ documents available
  • Just £6.99 a month

Not the one? Search for your essay title...
  • Join over 1.2 million students every month
  • Accelerate your learning by 29%
  • Unlimited access from just £6.99 per month

See related essaysSee related essays

Related University Degree Computer Science essays

  1. The purpose of this assignment is to research about Electronic data interchange and needs ...

    "The Internet is the future of EDI," says Geri Speiler, a research analyst at Gartner Group. Gartner believes that as the demand for Internet commerce increases so will the demand for EDI. (33) 21.0 Conclusion After completing this assignment I have found out what EDI is for, what EDI is

  2. Traffic Light Simulation and Machine Code Program.

    Description of the Simulator In the shareware version the following instructions are not included. CALL, RET, INT and IRET. The hardware timer interrupt does not function because IRET cannot be used either. The registered version includes these features. The simulator emulates an eight-bit CPU that is similar to the low eight bits of the 80x86 family of chips.

  1. NETWORK DESIGN PROPOSAL

    South District Network Sub-Diagram Administration LAN South District Network Sub-Diagram Finance/Accounting & Computer Services LAN South District Network Sub-Diagram Human Resources LAN DETAILED DESIGN DOCUMENTATION Key for Lists = Product * Cost * Availability * Performance * Maintainability HP File/Print Server LH 3000 * $4,089.00 * Available/In Stock * P3

  2. Network Design

    "Layer 3 switches" as their whitepaper it describes that layer 3 switches are capable to do switching and routing on every port in high speeds and can handle high bandwidth of traffic than layer 2 switches To the backbone switch servers, internet connectivity and other switches are connected, so through

  1. OMC-R and CBSC ATP Overview

    SCSI SCSI CONFIGURED CNFG_OK 124254 A06-08 D0KJHC devDsk4gbCru 1.3.302 SCSI SCSI CONFIGURED CNFG_OK 124254 A06-08 D0KJFD devDsk4gbCru 1.3.305 SCSI SCSI CONFIGURED CNFG_OK 112390 B11-19 D059VV devDdsCru 1.3.307 SCSI SCSI CONFIGURED CNFG_OK 124254 A04-06 D0AA0D devDsk4gbCru 1.3.308 SCSI SCSI CONFIGURED CNFG_OK 124254 A04-06 D0JXE8 devDsk4gbCru 1.3.311 SCSI SCSI CONFIGURED CNFG_OK 112462

  2. The project explains various algorithms that are exercised to recognize the characters present on ...

    Hence, the power function from the math library routine was used to optimize the resizing_quad_gen function. Similarly, another technique was used to optimize the 'for' loop in Template_matching function. Let us see the basic requirement to perform Template Matching before diving into the optimization technique.

  1. Unix Shell Programming

    5 Factorial of 5 is 120 -:PROGRAM TO CHECK A NUMBER WHEATHER IT IS A PRIME NUMBER OR NOT:- SOURCE CODE echo Enter an integer number: read num o= ' expr $num/2 ' flag = 0 i=2 while [ $i -le $o] do m = 'expr $num % $i '

  2. Methods and technology used in Computer Forensics

    Because of the popularity of the Windows operating system, one of the chief concerns in computer forensics is recontructing the platform's registry, the files required for the system to function properly. Windows stores its registry files in a binary format that is propriatory to Microsoft.

  • Over 160,000 pieces
    of student written work
  • Annotated by
    experienced teachers
  • Ideas and feedback to
    improve your own work