TEST BANK FOR The Art of Computer Systems Performance By Raj Jain
- GradeMaster1
- Rating : 1
- Grade : C+
- Questions : 0
- Solutions : 1124
- Blog : 0
- Earned : $278.60
Compare the ratio with system A as the base
System Workload Workload Average
A
B
Considering the ratio of performance with system A as base we con
clude that system B is better
Compare the ratio with system B as the base
System Workload Workload Average
A
B
Considering the ratio of performance with system B as base we con
clude that system A is better
incomplete
Can be done
a Measurements Run your favourite programs and pick the one that
runs them faster
b Use measurements and simulations of various network con guirations
c Measurement
d
a Analytical modelling
b Analytical modelling and simulations
c Extensive simulations and modelling
a Response time for commonly used programs
Failure rate
rate of crashing
Storage capacity
Userfriendliness
b Query response time
Failure rate
Storage capacity
Usability
c Capacity
Response time
Failure rate
d Response time
The following information is from SPEC
Standard Performance Evaluation
Corporation home page
CPU benchmarks
CINT current release Rel Integer benchmarks contains
Name Application
espresso Logic Design
li Interpreter
eqntott Logic Design
compress Data Compression
sc Spreadsheet
gcc Compiler
CFP current release Rel
Floating point benchmark suite contains
Name Application
spiceg Circuit Design
doduc Simulation
mdljdp Quantum Chemistry
wave Electromagnetism
tomcatv Geometric Translation
ora Optics
alvinn Robotics
ear Medical Simulation
mdljsp Quantum Chemistry
swm Simulation
sucor Quantum Physics
hydrod Astrophysics
nasa NASA Kernels
fpppp Quantum Chemistry
More information about these benchmarks can be found in httpperformancenetliborgperforma
web page
A C program to implement sieve workload
seivec Program to implement sieve workload
include stdioh
define MaxNum List all primes upto MaxNum
define NumIterations
Repeats procedure NumIterations times
define TRUE
define FALSE
void mainvoid
int IsPrimeMaxNum
int ikIteration Loop indexes
int NumPrimes Number of primes found
printfUsing Eratosthenes Sieve to find primes up to dn MaxNum
printfRepeating it d timesnNumIterations
for Iteration Iteration NumIterations Iteration
Initialize all numbers to be prime
for i i MaxNum i
IsPrimei TRUE
i
while ii MaxNum
if IsPrimei
Mark all multiples of i to be nonprime
k i i
while k MaxNum
IsPrimek FALSE
k k i
of while k
of if IsPrime
i i
of WHILE ii
NumPrimes
for i i MaxNum i
Count the number of primes
if IsPrimei
NumPrimes NumPrimes
printfd primesnNumPrimes
of for Iterations
The following can be added during debugging to list primes
for i
i MaxNum i
if IsPrimei printfdni
The result of running the program
Using Eratosthenes Sieve to find primes up to
Repeating it
times
primes
primes
primes
primes
primes
primes
primes
primes
primes
primes
a Cannot compare systems oering dierent services
b Metric response time
Workload Favourite programs Word processor spreadsheet
c Metric response time functionality Workload A synthetic program
which tests the versions using various operating system commands
operating system services
d Metric Response time reliability time between failures
Workload A synthetic program generating representative oppy
drive IO requests
e Metric size of code structure of code execution time
Workload A representative set of programs in C and Pascal
a
tCPU
n
n
Xi
tCPU
nIO
n
n
Xi
nIO
s
xs
n
n
Xi
xsi xs
n n
Xi
x
si nx
s
Similarly
s
xr
b Normalize the variables to zero mean and unit standard deviation The
normalized values xs and xr are given by
xs
xs xs
sxs
xs
xr
xr xr
sxr
xr
The normalized values are shown in the fourth and fth columns of
Table b
The other steps are similar to example
Observation Variables Normalized Variables Principal Factors
No xs xr xs xr y y
Px
Px
Mean
Standard
Deviation
The correlation between CPU time
tCPU and number of IOs
nIO
is The principal factors y and y are
y
y p
p
p p
tCPU
nIO
The rst factor explains
or of total variation
There is no unique solution to this exercise Depending upon the choice of
outliers scaling technique or distance metric dierent results are possible
all of which could be considered correct One solution using no outliers range
normalization to
and Euclidean distance starts with the the normalized
values shown in the following
Program CPU time IOs
TKB
MAC
COBOL
BASIC
Pascal
EDT
SOS
BASIC Pascal EDT COBOL SOS MAC and TKB join the dendro
gram at distances of and respectively
[Solved] TEST BANK FOR The Art of Computer Systems Performance By Raj Jain
- This solution is not purchased yet.
- Submitted On 16 Nov, 2021 06:32:35
- GradeMaster1
- Rating : 1
- Grade : C+
- Questions : 0
- Solutions : 1124
- Blog : 0
- Earned : $278.60