CSC470                                                 Fall, 2000
             Computer System Modelling And Analysis
                          Laboratory #1
                  September 19/21 & 26/28, 2000

Verification of Authorship:
      By the signatures below, we certify that we are thoroughly
familiar with the contents of this laboratory report: it is our
own work, we have referenced all our sources of information, and
we are the sole authors.

Team Members:
         Name                Signature                 Date

   (1).
                             ________________________  _________

   (2).
                             ________________________  _________

   (3).
                             ________________________  _________

       
Lab #1 Assignment

I.   Effect of the Think Time Distribution

     (a). With customers using only a single service centre  (the
          cpu)  where  service times are exponential, investigate
          the difference  in  performance  for  each  of  Erlang,
          Exponential  and  Hyperexponential  think  times [three
          cases].

                ***** enter summary results for three cases *****

     (b). Briefly describe how the degree of variance (or coeffi-
          cient  of  variation)  of  the think times affects mean
          response time.

                ***** enter answer *****

II.  Interaction of Scheduling Discipline and Service  Time  Dis-
     tribution

     (a). With customers using only a single service centre  (the
          cpu)  and  with  exponential  think  times, compare the
          relative effectiveness of FCFS  scheduling  and  Round-
          Robin  scheduling (with quantum 0.2) for each of Deter-
          ministic and Hyperexponential service times at the  cpu
          [four cases].

                ***** enter summary results for four cases *****

     (b). Describe briefly how mean response time is affected  by
          the  choice  of service discipline for the two types of
          service time distribution.

                ***** enter answer *****

III. Effect of Form of Service Time Distributions

     (a). For exponential think times with customers using  three
          service  centres,  change the forms of the service time
          distributions at the two service centres to be each  of
          both  Erlang,  both Exponential, and both Hyperexponen-
          tial [three cases].

                ***** enter summary results for three cases *****

     (b). Describe briefly how mean response time is affected  by
          the  various choices for the service time distributions
          at the centres.

                ***** enter answer *****

IV.  Effect of Balanced I/O

     (a). Using a three service  centre  model  with  exponential
          think  times, investigate the effect of various degrees
          of balance between the  number  of  I/O  operations  at
          disk1 and at disk2 [three cases].

                ***** enter summary results for three cases *****

     (b). Describe briefly the effect on response times as I/O at
          the two disks becomes more and more imbalanced.

                ***** enter answer *****

V.   Effect of Many Short Visits versus Few Long Visits

     (a). With Exponential service times everywhere, compare  the
          response  times that result when mean service times and
          routing frequencies are altered  to  reflect  different
          frequencies  and  durations  of  service  centre visits
          while keeping the total service demand constant at  all
          three centres [two cases].

                ***** enter summary results for two cases *****

     (b). Comment briefly on the  impact  on  response  times  of
          changing  the frequencies and durations of service time
          visits.

                ***** enter answer *****

VI.  Effect of Round-Robin Quantum Length

     (a). For  Exponential  think  times  and  a  single  service
          center,  using hyperexponental service times and round-
          robin scheduling, compare the performance with the fol-
          lowing  three  choices  of  round-robin quantum length:
          20.0, 2.0, 0.2 [three cases].

                ***** enter summary results for three cases *****

     (b)  Briefly discuss the impact of the choice of the  round-
          robin  quantum length on the performance of the system,
          and on the efficiency of the simulation runs.

                ***** enter answer *****

Csim Program:

                ***** insert Csim program (for one of the 18 cases) *****

Csim Output:

                ***** insert full output of Csim program for one case *****

***************************************************************************