Intro Software Metrics

8/3/2019 Intro Software Metrics

1/22

Software Metrics

Software Engineering


2/22

Definitions

Measure- quantitative indication of extent,amount, dimension, capacity, or size ofsome attribute of a product or process.

E.g., Number of errors

Metric- quantitative measure of degree towhich a system, component or process

possesses a given attribute. A handle orguess about a given attribute.E.g., Number of errors found per person hoursexpended


3/22

Why Measure Software?

Determine the quality of the currentproduct or process

Predict qualities of a product/process

Improve quality of a product/process


4/22

Motivation for Metrics

Estimate the cost & schedule of future projects

Evaluate the productivity impacts of new tools andtechniques

Establish productivity trends over time

Improve software quality

Forecast future staffing needs

Anticipate and reduce future maintenance needs


5/22

Example Metrics

Defect rates

Error rates

Measured by:individual

module

during development

Errors should be categorized by origin,type, cost


6/22

Metric Classification

ProductsExplicit results of software developmentactivities

Deliverables, documentation, by products

ProcessesActivities related to production of software

ResourcesInputs into the software development activities

hardware, knowledge, people


7/22

Product vs. Process

Process MetricsInsights of process paradigm, softwareengineering tasks, work product, or milestonesLead to long term process improvement

Product MetricsAssesses the state of the projectTrack potential risks

Uncover problem areasAdjust workflow or tasksEvaluate teams ability to control quality


8/22

Types of Measures

Direct Measures (internal attributes)

Cost, effort, LOC, speed, memory

Indirect Measures (external attributes)

Functionality, quality, complexity,efficiency, reliability, maintainability


9/22

Size-Oriented Metrics

Size of the software produced

LOC -Lines Of Code

KLOC -1000 Lines Of Code

SLOCStatement Lines of Code (ignorewhitespace)

Typical Measures:

Errors/KLOC, Defects/KLOC, Cost/LOC,Documentation Pages/KLOC


10/22

LOC Metrics

Easy to use

Easy to compute

Language & programmer dependent


11/22

Complexity Metrics

LOC - a function of complexity

Language and programmerdependent

Halsteads Software Science (entropy

measures)

n1

- number of distinct operators

n2 - number of distinct operands

N1 - total number of operators

N2 - total number of operands


12/22

Example

if (k < 2)

{

if (k > 3)

x = x*k;

}

Distinct operators: if ( ) { } > < = * ;Distinct operands: k 2 3 x

n1 = 10n2 = 4N1 = 13N2 = 7


13/22

Halsteads Metrics

Amenable to experimental verification [1970s]

Program length: N = N1 + N2Program vocabulary: n = n1 + n2

Estimated length: = n1 log2 n1 + n2 log2 n2Close estimate of length for well structured programs

Purity ratio: PR = /N

N

N


14/22

Program Complexity

Volume: V = N log2 nNumber of bits to provide a unique designator for each ofthe n items in the program vocabulary.

Difficulty

Program effort: E=D*VThis is a good measure of program understandability


15/22

McCabes Complexity Measures

McCabes metrics are based on a

control flow representation of theprogram.

A program graph is used to depictcontrol flow.

Nodes represent processing tasks(one or more code statements)

Edges represent control flow betweennodes


16/22

Flow Graph Notation

Sequence

If-then-else

While

Until


17/22

Cyclomatic Complexity

Set of independent paths through thegraph (basis set)

V(G) = E N + 2E is the number of flow graph edges

N is the number of nodes

V(G) = P + 1P is the number of predicate nodes


18/22

Example

i = 0;

while (i


19/22

Flow Graph

1

3

54

6

7

2


20/22

Computing V(G)

V(G) = 9 7 + 2 = 4

V(G) = 3 + 1 = 4

Basis Set1, 7

1, 2, 6, 1, 7

1, 2, 3, 4, 5, 2, 6, 1, 7

1, 2, 3, 5, 2, 6, 1, 7


21/22

Another Example

1

6

7

4

5

8

3

9

2

What is V(G)?


22/22

Meaning

V(G) is the number of (enclosed) regions/areas ofthe planar graph

Number of regions increases with the number of

decision paths and loops

A quantitative measure of testing difficulty and anindication of ultimate reliability

Experimental data shows value of V(G) should beno more then 10 - testing is very difficulty abovethis value

Intro Software Metrics

Documents

Transcript of Intro Software Metrics