![[^]](images/ManUpArrow.gif)
Absolute performance, 141
Accelerated testing, 175
Acceptance test (see Acceptance testing)
Acceptance testing, 570-571, 581
Acceptance voting, 584-585
reliability modeling, 603-604
Accessibility matrix, 49
Accuracy:
model, 281
operational profile, 171
Acquired software, reliability measurement, 235
Action model, 58-60
Activation function, 700-701
Active fault, 28
Adaptive maintenance, 32
Adequacy criterion for test set, 534
Adjudication:
adjudicator, 575
algorithm, 577
by voting, 577-578
Adjusted relative complexity, 503
Adjustment of failure time, 240-241
Administrator, system, 174
Advantage:
implicit profile, 178-180
operational profile, 167-168
Advice to reliability users, 162-163
Akaike information criterion, 297, 458
Algorithm defect, 375
Allocation of development resources, 172
Alternates, 570
Alternatives of model combinations, 295-298
Analysis:
cluster, 326, 348
confirmatory, 450
environment, 448-449
exploratory, 449-450
failure dependency, 324-327
failure modes and effects, 233
fault tree, 233, 619-622, 624
Markov reward, 334
multiple discriminant, 511
multiple regression, 512
multivariate, 326, 776-779
neural network model, 718
performability, 351-352
principal component, 326
principal components, 495-499
sensitivity, 338-339
subadditive property, 406-407
trend, 410
trend test, 410
variance, 716
Application result for reliability measurement, 285-289
Approach:
classification, 719
data domain, 73
early prediction, 162
error seeding and tagging, 73
estimative, 719
experimental, 723
linear combination model, 289-299, 739
software reliability engineering, 220-226
time domain, 73
tree-based, 202
Approximation of operational profile, 168
Approximation of execution time, 240
Architecture discovery process, 211-212
Arithmetical mean test, 411-412
Artifact, software 663
Artifact simulation (see Artifact-based simulation)
Artifact-based simulation, 665, 669-676
execution harness, 674
inputs, 671-672
reliability assessment, 674-675
results, 675-676
self-driven, 672-673
trace-driven, 674
simulator architecture, 670-675
Artificial neural networks (see Neural networks)
Artificial neuron, 700
Assessing reliability capabilities of competitors, 227
Assessment of quality factor, 222
Assignment defect, 374
Assumptions:
independence, 403, 580, 588
reliability growth, 450
interversion failure correlation, 589
stochastic independence, 403, 580, 588
AT&T software reliability engineering toolkit, 732
Atomic effort, 548
Atomic systems, 35-41
Attribute of dependability, 29, 32-33
Attribute-value set, 394
Authorized Program Analysis Report, 268, 362
Autocorrelation coefficient, 777
Automation:
diagnosis environment, 331
diagnosis of recurrences, 331-333
failure data collection, 245
selection of best predictor, 162
test, 238
verification of numerical results, 573-574
Availability, 479-486
average, 62, 481
consideration in setting reliability objectives, 231
definition of, 14, 29, 35
instantaneous, 480-481
n-step approximation, 484-485
pointwise, 61
prediction, 485-486
steady-state, 483
approximation, 484
system, 483
uptime, 481, 483
Availability modeling, 60-64, 483-486
average availability, 62
Laplace transform, 62
pointwise availability, 61
unavailability, 62-64
Average bias, 714
Average error, 714-715
Average failure rate, 481
Average recovery rate, 481
Back-propagation learning algorithm, 705-707
Back-to-back testing, 606
Back-to-back voting, 582
Background:
event, 207
model, 72-73
Backward compatibility trigger, 388
Backward elimination, 519
Backward error recovery, 570
Basic assumption, 78
Basic block, 555
Basic event in fault tree, 622
Basis for characterizing system mode, 174-175
Bathtub curve, 753-754
Bayesian model, 104-109, 524-526
Littlewood-Verrall (see Model: Littlewood-Verrall)
Behavior (see Service)
Belady's bandwidth metric, 513
Bell-shaped response curve, 702
Benefit of software reliability engineering, 210, 220-226
Bernoulli distribution, 749
Best current practice, 219
Best estimate of reliability, 259
Best predictor, automatic selection, 162
Beta test, 241
Bias:
average, 714
definition of, 139, 144-145
detection of, 145
trend (see y-plot)
unit, 701, 703
Bimodal failure rate curve, 268-270
Binomial distribution, 749
Binomial model, 75, 77
Birth rate, 762
Birth-death model, calendar-time, 468
Birth-death process, 762-765
Block, 555
Book overview, 8-12
Book scope, 21-23
Boundary value, 775
Boundary version reliability, 592
Braun statistic, 130
Brooks and Motley model, 92-93
Bug, 13, 362 (see also Fault)
Busy hour, 201
Butterfly model, 360
c2 statistic (see Chi-square statistic)
C-use (see Computational use)
Calendar time:
birth-death model, 468
data requirements, 88
definition of, 14
reliability analysis, 461, 463-469
Call trees, 178-179, 202
Cascade architecture, 707
Cascade-correlation learning algorithm, 707-709
Catastrophic hazard, 624
Categorization of fault, 314-315
data definition fault, 314-315
data fault, 314-315
incorrect computation, 314-315
missing operation, 314-315
side effect of code update, 314-315
unexpected situation, 314-315
Causal analysis, 364-366
Cause:
identifier, 395
system outages, 439-440
Certification of reliability objectives, 240-241
Change:
control, 235
definition of, 363
during testing, 269-271
in defect type distribution with phase, 370
in trend, 407-408
in evolving software, 276
local trend, 406
Chapman-Kolmogorov dynamics, 763
Characteristics:
code structure, 670
error model, 343
project, 281-284
recovery model, 343
software environment, 669
Characterization:
project, 278-279
system modes, 174-175
Checkpointing and restart, 328
Chi-square statistic:
definition of, 130
distribution, 750, 770
test, 714, 774-775
Classification:
accuracy of classifiers, 724
approach, 719
data, 280, 312, 517-518
error type, 516
failure, 31, 227
fault, 30-31, 314-317
error, 314-317
model, 59
opinion-based, 373
severity, 12, 227-228, 477-478
semantic, 373
Classification scheme, model, 73-76
Classifier:
classification accuracy, 724
general, 442-443
minimal network growth using cross-validation, 721-722
minimum distance, 722
multilayer network, 721
perceptron, 721
software-specific, 442-443
Clean-room, 222
Clock time, 14
Cluster analysis, 326, 348, 778-779
Code churns, 275
Code identifiers for software systems, 448
Code structural characteristics, 670
Coincident failure, 575, 596, 628, 635
k-fold, 575
Collection of data, 159-160, 284-285
Combination techniques for reliability measurement, 161-162
Combinatorial model, 758
Command-driven system, 191
Commercial tool, 730
Common-cause fault, 576, 594
Comparison:
model, 120-127
sequences of predictions, 131
model performance, 286-288, 292-295
Complexity:
control flow, 497, 522
coupling, 500
data structure, 521
dynamic, 504-510
functional, 505-506
metric, relative, 501
operational, 509-510
program size, 496, 522
relative complexity, 501-503
metric, 501
adjusted, 503
scaled, 501
relative subgraph complexity, 509
space, 681
system, 500
time, 679, 681, 683
Component configuration change in evolving software, 276
Component failure rate, 46
interface failure probability, 46
internal component failure rate, 46
Component model, 289
Compression ratio, 548
selection, 551-553
Computation complexity (see Time complexity)
Computation error model, 646
Computation diagnostics, 457-458
Computational use, 533
Computer-Aided Software Reliability Estimation tool (CASRE), 739-740
Concurrent engineering, 667
Conditional probability, 749-750
Conditional probability density function, 749-750
Conditional reliability, 55, 57-58
Confidence:
bound, 260
coefficient, 769
interval, 260, 769
for means, 769-770
for proportions, 771
for variances, 770-771
Confidentiality, 29
Configuration:
change in evolving software, 276
management system, 503
profile, 202
vectors, 503
Confirmatory analysis, 450
Consensus recovery block, 583-584
reliability modeling, 601-603
Consensus voting, 578
reliability, 592-593, 599-601
versus majority voting, 600
Consistency check, 320
external, 571-573
data, 448
Constant:
hazard, 754
failure-rate model, 474-475
subadditivity factor, 408-409
Continuous distribution, 750
Continuous random variable, 748-749
Continuous-space Markov model, 758
Continuous-time:
Markov model, 758-759
system reliability, 36-37, 40
Control:
charts, 423
error, 315-317
testing progress, 422
Control flow:
complexity, 497, 522
testing, 532-533
Convolution operation, 54
Corrective maintenance, 32
Corrective service, 329
Correlated elements in functional and operational profiles, 183
Correlated failure, 324-325, 576, 598-599
Correlated fault, 596
Correlation analysis, 777
Correlation coefficient, 325-326, 777
Corruption, data, 197, 332, 449
Cosine arch moving average, 376
Cost:
fault-tolerant software, 607-609
testing, 265
life-cycle, 229-230
operational failure, 264-265
operational profile, 168, 171, 222
orthogonal defect classification, 397
processing and analyzing failure data, 223
reduction, 168
root-cause analysis, 397
software reliability engineering consultant, 223
Coupling complexity, 500
Covariance matrix, 497, 516
Covariance, 449, 708
Coverage:
definition of, 34
failure, 647
incorporation in reliability estimation, 546-547
information for failure data filtering, 547-551
model, 646-647
parameter, 337, 647
statistics, 544
CP statistic, 520
CPU-bound workload, 349-350
Criteria for selecting software reliability tool, 731-732
Critical hazard, 624
Critical operation, 175
Criticality:
category, 176
measurement, 175
Cross-referencing between logs, 311
Cross-validation:
minimal network growth, 721-722
neural network, 724
set, 722
Cumulative distribution function, 403, 748
empirical, 773
Cumulative failure function, 14
Cumulative failure plot, 452
Cumulative reward, 352
Customer:
definition of, 173, 175
expectations, 206
partnership, 444
perspective for quality measurement, 221
satisfaction tracking, 245
testing, 202-203
type, 173
Cut-off point, 485
Cutset:
definition of, 617
generation, 617-619
minimal, 617, 630-634
Cycle-time reduction, 360
Cyclomatic complexity number (see Cyclomatic number)
Cyclomatic number, 497, 500, 513, 521
Data aging, 454
Data analysis:
detailed analysis, 323-333
preliminary analysis, 312-323
principles, 449-459
Data analysis techniques:
data modeling, 450
data resistance, 450
data transformation, 450, 458-459, 739
plots and graphs, 450
Data and Tool Disk, 12
Data behavior, 454
Data classification, 280, 312, 517-518
Data coalescing, 313
Data collection and analysis environment:
data collection system, 449
manipulation, filtering, and data merging system, 449
software reliability modeling tool, 449
statistical analysis and visualization system, 449
Data collection and analysis process, standardization, 440
Data collection, 159-160, 284-285
automation, 245
form, 285
in field, 244-245
in system test, 239
items, 271
principle, 441-449
process, 266-267
plan, 266
system, 449
trial-run, 267
with multiple functional area, 272
Data consistency checking, 448
Data conversion (see Data transformation)
Data corruption, 197, 332, 449
Data definition, 267-272
Data diagnostics, 454-455
Data Disk (see Data and Tool Disk)
Data domain:
approach, 73
reliability modeling (see Reliability modeling in data domain)
Data-driven system, 193
Data extraction, 313
Data fault, 314-315
Data filtering:
definition of, 449
process, 549-550
filtering using coverage information, 547-551
Data flow:
diagrams, 211
testing and coverage, 532-534
Data granularity, 446-447
Data maintenance, 447-448
Data modeling, 450, 454-455
Data preprocessing, 312-314
Data recalibration with respect to usage, 454
Data recording, 190
Data references, global, 521
Data representation, 712-713, 722-723
Data resistance, 450
Data set:
fitting, 513
validation, 513
Data smoothing, 454, 739
Data-splitting, 513
Data structure complexity, 521
Data transformation, 17, 450, 458-459, 739
Data validation, 447-448
Data variability, 439
Deadlock error, 315-316
Death rate, 762
Debugging phase, 753
Decision coverage, 533-534
Decision-making process, 367
Def-use pair, 533
Defect:
definition of, 13, 361
density, 113
latent, 465
multidimensional analysis, 393-396
reduction, 360
reintroduction rate, 468
tool for tracking, 394
Defect model, 364-366
Defect prevention process, 397
Defect simulation status profile, 691-693
Defect trigger, 384-393
attribute, 384-393
function test triggers, 388-389
combination path coverage trigger, 389
simple path coverage trigger, 389
test coverage trigger, 389
test interaction trigger, 389
test interaction trigger, 389
test sequencing trigger, 389
test variation trigger, 389
review and inspection triggers, 387-388
backward compatibility trigger, 388
concurrency trigger, 388
design conformance trigger, 388
document consistency/completeness trigger, 388
lateral compatibility trigger, 388
operational semantics trigger, 388
rare situation trigger, 388
system test triggers, 387
hardware configuration and software configuration trigger, 387
normal mode trigger, 387
recovery/exception handling trigger, 387
system start-up and restart trigger, 387
workload volume/stress trigger, 387
definition of, 385
difference with symptom, 386
distribution, 391-393
use of, 389-393
Defect type, 368, 373-384
association with process, 376
attribute, 374-376
algorithm defect, 375
assignment defect, 374
build/package/merge defect, 374
checking defect, 374
defect due to something incorrect, 374
defect due to something missing, 374
documentation defect, 374
function defect, 374
interface defect, 374
timing/serialization defect, 374
definition of, 368, 373
distribution, 390-391
change with phase, 370
relative risk assessment, 376-384
Definition-free path, 533
Delayed S-shaped model (see Model: S-shaped model)
Density function (see Probability density function)
Dependability, 28-35
attribute of, 29, 32-33
impairment to, 28-32
means for, 33-35
requirement, 65-66
Dependence of faults/failures, 403, 586-589
Description statistics, 423
Design and implementation phase:
software reliability engineering, 232-235
trend analysis on inspection data, 431-434
Design conformance trigger, 388
Design constraints, 261
Design fault, 19
Design methodology versus operational profile, 173
Determination:
functional profile, 226-227
new feature introduction, 245-246
operational profile, 236-237
Development:
developer-oriented measures, 221
improvements of developer and customer communication, 199
methodology, 234
operational profile, 206-208, 211-214
practices for fault management, 234-235
resources allocation, 172
test environment, 208
Diagnosis, symptom-based, 332
Diagnostics for model determination, 455-458
computational, 457-458
graphical, 455-457
Differential-difference equation, 764
Discovery reviews, 231
Discrete distribution, 749
Discrete random variable, 748-749
Discrete-event simulation, 665
Discrete-space Markov model, 758
Discrete-time Markov model, 759
Discrete-time system reliability, 36, 39-40
Discriminant model, 515-516
Discrimination technique, 515
Disk (see Data and Tool Disk)
Distributed recovery block:
definition of, 581, 627, 629
fault tree model, 629-630
reliability model, 630
system reliability model, 647-648
system safety model, 647-648
Distributed systems, operational profile, 198-199
Distribution:
characterization, 772-776
function (see Cumulative distribution function)
fitting, 773-774
sample, 135
Bernoulli, 749
binomial, 749
chi-square, 750, 770
continuous, 750
defect trigger, 391-393
defect type, 390-391
discrete, 749
empirical, 772-773
exponential, 750
Gamma, 750
normal, 750
Poisson, 749
Rayleigh, 94, 755
Documentation:
defect, 374
document completeness trigger, 388
document consistency trigger, 388
operational profile, 203-204
Domain coverage and predictive model, 520-523
Domain loading, 498
Domain metric, 498
Domain pattern, 498, 514, 522-523
Domain testing, 554
Dormant fault, 28
Dot-product function, 701
Doubly stochastic process, 346
Duane model (see Model: Duane)
Duration of outage, 13
Dynamic aspect of functional complexity, 507-509
Dynamic consensus voting, 584
Dynamic majority voting, 584
Dynamic program complexity, 504-510
Dynamic simulation:
approaches, 664-665
need, 663-664
Dynamically weighted linear combination model:
basic model, 291
model with fixed window, 296
model with sliding window, 296
h point, 378-379, 382
Early prediction model, 17, 111-114, 162
Phase-based model, 111-112
Predicting defects from Ada designs, 112-113
Rome Laboratory model, 113-114
Economic Stop Testing Model (ESTM) tool, 740-741
Effect identifier, 395
Effort:
atomic, 548
estimated, 496
filtered, 548
testing, 540
Element, 168
11-point symmetrical moving-average, 481
Elman network (see Recurrent network: simple recurrent network)
End-point prediction, 711, 714
End-to-end reliability process, 691
Entity relationship diagram, 211
Environment:
automatic diagnosis, 331
characteristics, 669
consideration, 688-689
data collection and analysis, 448-449
profile, 181
variable, 175-176, 180-181, 190, 192
Equally weighted linear combination model, 290
Error:
definition of, 13, 28-29, 362-363
detected, 28
latent, 28
average, 714
relative, 458
Error bursts (see Multiple error)
Error compensation, 570
Error confinement area, 627
hardware, 627
software, 627
Error distributions, 320-323
Error entry, sample, 308-309
Error frequency, 320-321
Error log, preprocessing, 313
Error model, 319, 337-344, 349-351
Error propagation, 317-320
Error record, 308-309
Error recovery, 340, 570
distributions, 320-323
Error report:
Authorized Program Analysis Report (APAR), 268, 362
Modification Request (MR), 235, 268, 284
Problem/Failure Report (P/FR), 267
Tandem project report (TPR), 311
Error seeding and tagging, 73
Error states, 340
mean waiting time, 340, 342
Error statistics, 320
Error types, 315-317
control error, 315-317
deadlock error, 315-316
I/0 and data management error, 316-317
memory error, 317
multiple error, 316, 322
programming exception error, 316, 344
storage exception error, 316, 344
storage management error, 316
Establishment of software reliability requirements, 259-266
Estimated effort, 496
Estimation:
least-squares, 520, 768-769
number of sites, 475
software reliability, 17
Eestimative approach, 719
Estimator, 139, 766
consistent, 766
least-squares, 768-769
maximum likelihood, 767
unbiased minimum-variance, 766
Euclidean distance, 778
Evaluation:
fault tolerance, 328, 570
model, 342
operational phase, 306
reliability, 64-68
Events, reportable, 473, 483
Event process:
simulation, 665, 676
statistics, 677
Evolving software:
complexity metrics, 502
component configuration change, 240, 276
failure time adjustment, 240, 276-277
ignoring changes, 240, 276
Exception handling, 328, 571
trigger, 387
operating system exceptions, 573
exclusion:
remaining set of run types, 188-189
transient behavior data, 454
Execution of programs, 209-210
Execution profile, 505
relationship to relative complexity, 506-507
Execution time:
approximation, 240
component data requirements, 88
definition of, 14
measurement, 239-240
Expected life of system, 756
Expected value (see Mean)
Experimental approach, 723
Explanatory variables, 162
Explicit versus implicit functional profile, 177-180
Exploratory analysis, 449-450
Exponential distribution, 750
Exponential failure time class model, 77-93
Brooks and Motley (see Model: Brooks and Motley)
extended execution time (see Model: extended execution time)
hyperexponential (see Model: hyperexponential)
Inflection S-shaped (see Model: Inflection S-shaped)
Jelinski-Moranda (see Model: Jelinski-Moranda)
Laprie hyperexponential (see Model: Laprie hyperexponential)
Musa basic execution time (see Model: Musa basic execution time)
nonhomogeneous Poisson process (see Model: nonhomogeneous Poisson process)
Schneidewind (see Model: Schneidewind)
Yamada modified exponential growth (see Model: Yamada modified exponential growth)
Exponential order statistic model class, 111
Exponential random variables, 140
Exposure period, 540
Expressing software reliability, 259-261
Extended execution time model, 92
Extensions of model combinations, 295-298
Eyeball regression, 694
Factor analysis, 326, 777-778
common factor, 778
communality, 778
factor loading, 778
unique factor, 778
unique variance, 778
Factors affecting software reliability, 339
Factors in reliability objectives, 230-231
Fail-safe state, 647
Failure:
coincident (see Coincident failure)
correlated (see Correlated failure)
definition of, 12, 28-29, 227, 362-363
examples of, 4-5
fatigue, 754
interweaving, 53
load-induced, 348
median time to next, 128
multiple component, 325
single point, 619
wear-out, 754
workload-related, 345-352
Failure classification, 31, 227
Failure data:
automated collection, 245
definition of, 15, 442-444
failure-count, 15, 268, 271
failures per time period, 15, 268, 271
filtering process, 549-550
format, 410
interfailure times, 16, 268, 271
time-between-failures, 16, 268, 271
Failure density function, 751
Failure dependence, 586-589
Failure dependency analysis, 324-327
multiway, 326-327
two-way, 325-326
Failure domain, 36, 38-39
Failure frequency, 320-321
Failure functions, 14
cumulative failure function, 14
failure intensity function, 14
failure rate function, 14
mean time between failures (see Mean time to failure)
mean time to failure, 14
mean value function, 14
MTBF (see Mean time between failures)
MTTF (see Mean time to failure)
rate of occurrence of failures, 14, 55 (see also Failure rate function)
ROCOF (see Rate of occurrence of failures)
Failure impact, 333-335, 386
Failure independence, 580
Failure intensity:
decay parameter, 470
definition of, 757-758
function, 14
improvement factor, 258, 263
influence of usage, 462
Laplace test, 413
model, 59 (see also Failure-count model)
optimum objective value, 229-230
plots, 452-453
Failure model, 337-344, 349-351, 385-386
Failure mode, 31
Failure mode and effect analysis, 233
Failure rate:
average, 481
bimodal curve, 268-270
definition of, 752-756
field, 482
function, 14
instantaneous, 480
model, 59 (see also Time-between-failures model)
per-office, 475
system, 49
Failure rediscovery, 53
Failure severity classification, 12, 227-228, 477-478
Failure source, 65-66
Failure statistics, 320
Failure symptom, 385
Failure time adjustment, 240-241
in evolving software, 276-277
Failure-based measure, 221
Fan-in and fan-out connections, 700-701
Fatigue failure, 754
Fault:
active, 28
common-cause, 576, 594
definition of, 13, 28-29, 362-363
design, 19
dormant, 28
first occurrence of, 315
physical, 18
similar, 576, 594, 596
regression, 403, 423
related, 594, 596, 628, 636
spawned, 196
Fault avoidance, 34
Fault-based measures, 221
Fault classification, 30-31, 314-317
Fault categorization, 314-315
data definition fault, 314-315
data fault, 314-315
incorrect computation, 314-315
missing operation, 314-315
side effect of code update, 314-315
unexpected situation, 314-315
Fault data, 442-444
Fault density, 112-114
Fault dependence, 403
Fault discovery rate, 688-689
Fault-exposure ratio, 673
Fault forecasting, 19, 21, 29
Fault injection, 673
Fault management, 234-235
Fault masking, 570
Fault model, Gompertz curve, 467
Fault pathology, 28-29
Fault prediction model, 518-520
Fault prevention, 19-20, 29
Fault profile, 314
Fault-prone module identification, 710, 718-725
Fault rate, 112
Fault recurrence, 315
Fault removal, 19-20, 29
Fault sensitization, 402
Fault simulation, 691-693
Fault tolerance, 19-20, 29 (see also Fault-tolerant software)
Fault-tolerant software:
advanced techniques, 583-585
acceptance voting, 584-585
consensus recovery block, 583-584
dynamic consensus voting, 584
dynamic majority voting, 584
N self-checking programming, 583-586
applications, 568-569
basic techniques, 581-583
distributed recovery block, 581, 627, 629
N-version programming, 582-583, 627
recovery blocks, 581-582
comparative system analysis, 642-645
cost, 607-609
definition of, 567-568
design, 605-606
development, 604-609
evaluation, 328, 570
experiment, 596-598
fault tree analysis, 627
hybrid techniques, 583-585
parameter estimation, 639
performability analysis, 628
present status, 568-569
quantitative analysis, 635-645
reliability in presence of interversion failure correlation, 596-604
reliability growth modeling, 65, 585-596
requirement, 605-606
testing, 604-609
validation and verification, 606-607
Fault tree:
analysis of fault-tolerant software, 627
analysis, 233, 619-622, 624
as design aid, 622-623
assignment statement, 624
basic event, 622
construction, 616
if-then-else statement, 624
modeling, 615-617, 629-634
distributed recovery block, 629-630
N self-checking programming, 632-634
N-version programming, 630-632
recovery block, 629-630
safety validation, 623-627
Feasibility and requirement phase of software reliability engineering, 226-231
Feature:
coarse, 538
definition of, 536-537
granularity, 537
interacting, 538
testing, 238
tasks, 175
Feed-forward network, 703-704, 712
Field data, 244-245, 440
Field failure rate, 482
Field failure ratio, 467
Field reliability:
compared with objectives, 243-245
data variability, 439
reliability growth, 475-476
projected, 259
Field repair rate, 482
Field trial, 241
Field-defect model, Weibull-type, 464-465
Field required for reliability analysis, 443
Filtering of failure data, 547-551
Finite failure model, 73
First occurrence:
error, 318
propagation characteristics (see Propagation characteristics of first error)
failure, 244-245
fault, 315
First office application, 241
Fitting data set, 513
Fluctuation, 404, 433
Forecasting:
failure, 19, 21, 29
fault, 19, 21, 29
long-range, 447
Formal design review, 622
Formal inspection, 20
Formal methods, 19-20
Forward error recovery, 570
Forward inclusion, 519
F-statistics, 717
Function, 168-170, 176
mapping to operation, 184
number of, 177, 181
Function defect, 374
Function list, 180-181
Function test trigger, 388-389
Functional complexity, 505-506
dynamic aspect, 507-509
variability, 509
Functional profile, 169, 176-183, 211-212
determination, 226-227
explicit versus implicit, 177-180
number of functions, 177
occurrence probability, 181-183
Functional testing, 268-270, 427, 534
Gain factor, 702
Gamma distribution, 750
Gaussian function, 702
General rate-based simulation, 684-686
General-purpose tool, 730
Generalization training, 712
General classifiers, 442-443
Generalized exponential model class, 109-110
Generalized squared distance, 515
Generic users, 201
Geometric model (see Model: geometric)
Geometric Poisson model (see Model: geometric Poisson)
Global data reference, 521
Gnedenko test, 411
Golden answer (see Correct answer)
Gompertz curve fault model, 467
Goodness-of-fit, 714
Gradient descent procedure, 705-706
Graphical diagnostics, 455-457
Graphical interpretation of subadditive property, 404-406
Graphical trend test, 410
Graph, 450-454
Group run type, 169
Halstead number, 496, 513
Handicap value, 292
Hard-limiting threshold function, 701
Hardware:
configuration trigger, 387
recovery, 350
reliability, relationship to software reliability, 18-19
transient fault, 628
Hardware-related software failures, 327-328
Hazard, 624
catastrophic, 624
constant, 754
critical, 624
definition of, 752
function, 677
linearly increasing, 754-755
per-fault rate, 689
Heisenbug, 574
Hidden layer, 703
Hierarchy of interpreters, 43
High-risk area identification, 233
Histogram, 772
Hollander test, 411, 414
Homogeneous Markov process, 759
Homogeneous Poisson process, 74
Homogeneous run types, 194
Human triggers, 388
Hyperexponential model (see Model: hyperexponential)
Hypothesis test, 410-411, 450, 774
Identical and wrong answer, 575-576, 598
Identification:
customer reliability needs, 227-228
fault-prone software modules, 710, 718-725
high-risk area, 233
Identifier:
cause, 395
effect, 395
subpopulation, 395
Imminent-failure state, 450
Impact of failures on performance, 333-335
Impact of system activity, 345-352
Impairment to dependability, 28-32
Implicit profile advantage, 178-180
Improvement:
developer and customer communication, 199
method, 160-163
product and process, 246
quality, 222
Inclusion-exclusion method, 619
Incorrect computation, 314-315
Incremental build problem, 526-527
Independence:
assumption, 403, 580, 588
failure, 580
fault, 628
input sequence, 40
Index of stability, 231
Infant mortality, 753
Infinite failure class model, 73, 98-104
Duane (see Model: Duane)
geometric (see Model: geometric)
Lipow (see Model: Lipow)
Moranda geometric Poisson (see Model: Moranda geometric Poisson)
Musa-Okumoto logarithmic Poisson (see Model: Musa-Okumoto logarithmic Poisson)
Inflection point in reliability growth curve, 378
Inflection S-shaped model, 92, 381
In-process measurement, 367
Input domain partitioning, 551
Input layer, 703
Input space, 170
Input state, 169
Input trajectory, 38-39, 402 (see also Failure: domain)
Input variable, 169
indirect, 197
key, 177
number reduction, 186
In-service time, 445, 460
Inspections and reviews, 234, 431-432
Instantaneous availability, 480-481
Instantaneous failure rate, 480
Instantaneous recovery rate, 480
Instantaneous reward rate, 760
Integrity, 29, 32
Interaction matrix, 191
Interdependencies among project elements, 668-669
Interface:
defect, 374
failure probabilities, 46
interpretive, 43
Interfailure times:
data, 16
Laplace test, 412
Interpreters:
hierarchy, 43
system, 43
Interpretive interface, 43
Interrupt, externally initiated, 169
Interval estimation, 769-772
Interval reliability, 55, 58
Interval reward rate, 760
Interversion failure correlation assumption, 589
Interweaving of failure, 53
I/O and data management error, 316-317
I/O workload, 349
Iterative development process, 363
Jelinski-Moranda model (see Model: Jelinski-Moranda)
Jensen's estimator, 513
Jordan network (see Recurrent network: semirecurrent network)
modified, 712
Jump-start consulting, 223
Kendall test, 411, 737
Key input variables, 177
indirect, 197
interaction matrix, 191
Key operational input selection, 196
Key quality objectives, 257
K-fold coincident failure, 575
K-means algorithm, 349, 779
Knee of reliability growth curve, 378
Knowledge model, 58-60
Kolmogorov-Smirnov distance, 136, 738
Kolmogorov-Smirnov test, 714, 775-776
KS distance (see Kolmogorov-Smirnov distance)
Laplace factor, 412-413
Laplace test, 411, 737
failure intensity, 413
interfailure times, 412
relation to subadditive property, 414-419
Laplace transform:
availability, 62
failure intensity, 56
Laprie hyperexponential model, 92, 468
Latent defect, 465
Latent discovery effect, 348
Latent error, 28
Lateral compatibility trigger, 388
Law of large numbers, 693
Layer, 41-42, 703
Learning algorithm, 700, 705, 707
back-propagation, 705-707
cascade-correlation, 707-709
Learning rate, 706
Least significant difference, 717
Least-squares estimation, 520, 734, 739, 768-769
License information, 445
Life-cycle cost, 229-230
optimization for reliability specification, 263-266
Likelihood function, 767
Limitation:
present prediction techniques, 159-160
models, 76-77
testing methods, 542-543
Limited growth simulation, 683
Linear combination model, 289-299, 739
Linear discriminant model, 515
Linear relationships in reliability growth models, 459
Linear threshold function, 701
Linearly increasing hazard, 754-755
Lines of code, 495
Lipow model, 102
List:
customer type, 173
system mode, 174-176
user type, 173-174
Listing operations, 184-186
Littlewood-Verrall model (see Model: Littlewood-Verrall)
Load hazard model, 346-348
Load test program execution, 209-210
Load-induced effects, 348
Load-induced software failures, 348
Local fluctuations, 404
Local trend changes, 406
Log, cross-referencing, 311
Logging:
challenges in on-line evaluation, 310
on-line event, 308-309
on-line machine, 308-310
operator, 311
Logical time, 445
Logistic function, 702
Long-term prediction, 123-126, 159, 298-299, 447, 456-457, 711
Loss in performance measure, 761
Maintainability, 29, 32
Maintenance:
adaptive, 32
corrective, 32
perfective, 32
personnel, 174
policy, 51-53
Majority voting:
comparison with consensus voting, 600
definition of, 577
reliability, 592-593
Management of fault, 234-235
Manipulation, filtering, and data merging system, 449
Manual reporting, 308, 310-312
Mapping from functions to operations, 184
Markov chain, 43-44, 47, 63
Markov model, 646, 758-759
continuous-space, 758
continuous-time, 758-759
discrete-space, 758
discrete-time, 759
Markov process:
homogeneous, 759
nonhomogeneous, 759
Markov reward analysis, 334, 759-762
Matrix,
accessibility 49
covariance, 497, 516
rediscovery, 329-330
service utilization, 48
standardized transformation, 498-499, 514-515
Maximum likelihood estimation, 734, 736, 738-739, 766-767
Mazzuchi's modification of Littlewood-Verrall model, 108
McCabe's cyclomatic number, 497, 500, 513, 521
Mean, 748
Mean recurrence time, 342
Mean square error, 299
Mean time between failures (see Mean time to failure)
Mean time to failure, 14, 756-757
Mean time to repair, 14
Mean value function, 14
Mean value voting, 578
Mean waiting time in error states, 340, 342
Means for dependability, 33-35
Measure:
developer-oriented, 221
failure-based, 221
fault-based, 221
predictability evaluation, 713-715
static, 502
validation of, 521
Measurement:
connection with semantics, 372
criticality, 175
customer perspective of quality, 221
definition of, 360-361
execution time, 239-240
in-process, 367
process-based, 372
reliability for acquired software, 235
software reliability, 17 (see also Software reliability measurement)
Measurement-based analysis:
overall framework, 304
steps, 304-305
studies, 307
Measurement plan quality, 471
Measurement program, operational-profile, 198
Measurement system, multidimensional, 360
Measurement techniques, 307-312
Median time to next failure, 128
Median voting, 577, 583
Median-weighted linear combination model, 291
Meeting reliability objectives, 232-233
Memory error, 317
Merge defect, 374
Method:
analyzing predictive accuracy, 128-145
improvement, 160-163
inclusion-exclusion, 619
least square, 768-769
maximum likelihood, 766-767
moments, 767-768
Metrics:
Belady's bandwidth metric, 513
definition of, 495-496
domain, 498
Jensen's estimator, 513
path, 449
usage, 499-500
Middleware software, 390
Minimal cutset, 617, 630-634
Minimal network growth using cross-validation, 721-722
Minimizing number of recurrences, 329-331
Minimum distance classifier, 722
Missing operation, 314-315
Mistake, 13 (see also Error)
Mode of failure, 31
Model:
action, 58-60
Bayesian, 104-109, 524-526
binomial type, 75, 77
Brooks and Motley, 92-93
Butterfly, 360
computation, 646
constant-failure-rate, 474-475
coverage, 646-647
Duane, 98-99
assumptions, 98
data requirements, 98
model estimation, 99
model form, 99
overview, 98
reliability prediction, 99
dynamically weighted linear combination, 291, 296
equally weighted linear combination, 290
error, 337-344, 349-351
multiple occurrence, 344-345
propagation, 319
exponential failure time class, 77-93
extended execution time, 92
failure, 337-344, 349-351
finite, 73
infinite, 73
intensity, 59
no-failure, 474
rate, 59
geometric, 99-102
assumptions, 100
data requirements, 100
model estimation, 101
model form, 100-101
overview, 99-100
reliability prediction, 101
hyperexponential, 90-92
assumptions, 90-91
data requirements, 91
model estimation, 91
model form, 91
overview, 90
reliability prediction, 91
infinite failure class, 98-104
Inflection S-shaped, 92, 381
interconnection structure, 700
Jelinski-Moranda, 77-80
assumptions, 78
data requirements, 78
model estimation, 79
model form, 79
overview, 77-78
reliability prediction, 79
knowledge 58-60
Laprie hyperexponential, 92, 468
linear combination, 289-299, 739
Lipow, 102
Littlewood-Verrall, 105-109, 524
assumptions, 106
data requirements, 106
model estimation, 107-108
model form, 106-107
linear form, 107
quadratic form, 107
overview, 105
reliability prediction, 107-108
load hazard, 346-348
median-weighted linear combination, 291
Moranda geometric Poisson, 102
Musa basic execution time, 87-90
assumptions, 87-88
data requirements, 88
calendar time component, 88
execution time component, 88
model estimation, 89-90
model form, 88-89
overview, 87
reliability prediction, 89-90
Musa-Okumoto logarithmic Poisson, 102-104
assumptions, 102
data requirements, 102
model estimation, 103
model form, 102-103
overview, 102
reliability prediction, 103
nonhomogeneous Poisson process (NHPP), 80-82
assumptions, 80
data requirements, 80
model estimation, 81
model form, 81
overview, 80
reliability prediction, 81-82
nonparametric, 699
neurons, 700
parametric, 709
Poisson type, 74-75, 77
Raleigh, 94
rare-event, 473-476
recovery, 337-344
resource-usage, 349-351
Schick-Wolverton, 94
Schneidewind, 82-87
assumptions, 83
data requirements, 83
model estimation, 84-86
model form, 84
overview, 82-83
reliability prediction, 84-86
simulation, 662
software quality, 511
software reliability, 17-18
S-shaped, 95-98
assumptions, 95-96
data requirements, 96
model estimation, 97-98
model form, 96-97
overview, 95
reliability prediction, 97-98
system, 41-43, 646
ultrareliability, 277-278
unequally weighted linear combination, 291
weight determination based on ranking model results, 290-291
Weibull and gamma failure time class, 93-98, 756
Weibull, 93-95, 756
assumptions, 93-94
data requirements, 94
model estimation, 95
model form, 94-95
overview, 93
reliability prediction, 95
workload, 348-349
Yamada modified exponential growth, 91
Model accuracy, 126, 281
Model application issues, 273-276
Model assumptions:
basic, 78
Duane model, 98
general 274-275
geometric model, 100
hyperexponential model, 90-91
Jelinski-Moranda model, 78
Littlewood-Verrall model, 106
Musa basic execution time model, 87-88
Musa-Okumoto logarithmic Poisson model, 102
nonhomogeneous Poisson process model, 80
S-shaped model, 95-96
Schneidewind model, 83
standard, 78
Weibull model, 93-94
Model background, 72-73
Model bias (see u-plot)
Model classification, 59, 73-76
Model combinations, extensions and alternatives, 295-298
Model comparison, 120-127
Model determination diagnostics (see Diagnostics for model determination)
Model disagreement, 121
Model estimation:
Duane model, 99
geometric model, 101
hyperexponential model, 91
Littlewood-Verrall model, 107-108
Musa basic execution time model, 89-90
Musa-Okumoto logarithmic Poisson model, 103
Nonhomogeneous Poisson process model, 81
S-shaped model, 97-98
Schneidewind model, 84-86
Weibull model, 95
Model evaluation, 342
Model form:
Duane model, 99
geometric model, 100-101
hyperexponential model, 91
Jelinski-Moranda model, 79
Littlewood-Verrall model, 106-107
Musa basic execution time model, 88-89
Musa-Okumoto logarithmic Poisson model, 102-103
Nonhomogeneous Poisson process model, 81
S-shaped model, 96-97
Schneidewind model, 84
Weibull model, 94-95
Model limitations, 76-77
Model performance comparisons, 286-288, 292-295
Model relationship, 109-111
Model selection, a priori, 126-127
Model trade-off study, 230
Modification Request (MR), 235, 268, 284
Modified renewal process, 56-57
Monitoring field reliability versus objectives, 243-245
Monotone reliability growth, 404
Moving average:
11-point symmetrical, 481
cosine arch, 376
three-point symmetric, 465-466
MR (see Modification Request)
MTBF (see Mean time between failures)
MTTF (see Mean time to failure)
MTTR (see Mean time to repair)
Multi-interpreter system, 47-49
Multicollinearity, 499
Multidimensional defect analysis, 393-396
Multidimensional measurement system, 360
Multilayer network, 703
classifier, 721
Multiple checks, 332-333
Multiple component failure, 325
Multiple correct output, 577
Multiple discriminant analysis, 511
Multiple error, 316, 322, 344-345
Multiple operational profile, 173, 237
Multiple regression analysis, 512
Multivariate analysis, 326, 776-779
software quality control, 514-518
Musa basic execution time model (see Model: Musa basic execution time)
Musa's modification of Littlewood-Verrall model, 108-109
Musa-Okumoto logarithmic Poisson model (see Model: Musa-Okumoto logarithmic Poisson)
Mutant, 533
Mutation testing, 532-533, 554
N self-checking programming, 583-586, 627
fault tree model, 632-634
reliability model, 633
system reliability model, 650
system safety model, 650
Nature:
prediction error, 136, 151
reliability process, 666-667
software faults, 385
Near homogeneous run category, 194-195
Necessary change, 363
Need:
simulation, 663-664
software reliability, 3-5
Network (see Neural network)
Neural network, 699-718
application for software reliability, 709-710
architecture, 702-704, 706-707
cross-validation, 724
feed-forward, 703-704, 712
initialization, 708
mapping, 711
model analysis, 718
multilayer, 703
processing unit, 700-702
recurrent (see Recurrent network)
single-layer, 703
software reliability growth model, 711
software reliability growth modeling, 710-718
New feature introduction, 245-246
Next-release effect, 465, 478-479
Next-step prediction, 714 (see also Short-term prediction)
NHPP model (see Model: non-homogeneous Poisson)
N-modular-redundancy, 582
Noise, 128, 144
Nonhomogeneous Markov process, 759
Nonhomogeneous Poisson process, 59, 74
Nonhomogeneous Poisson process model (see Model: nonhomogeneous Poisson process)
Nonparametric model, 699
Nonstationary Markov process (see Nonhomogeneous Markov process)
Normal distribution, 750
Normal mode trigger, 387
Notion of prediction error, 145
N-step approximation to availability, 484-485
N-step-ahead prediction (see Long-term prediction)
Null hypothesis, 410
Number of remaining defects, 463
Number of sites estimation, 475
N-version programming, 582-583, 627
fault tree model, 630-632
reliability model, 631
system reliability model, 648-650
system safety model, 648-650
Objectives:
project application, 278, 281
quality, 257
reliability, 257-258
software reliability engineering, 219
Occurrence probabilities:
functional profile, 181-183
operational profile, 189-194
ODC (see Orthogonal defect classification)
One-step-ahead prediction, 122
On-line machine logging, 308-310
Operating phase, normal, 754
Operation, 15, 168-170, 184
critical, 175, 189
listing, 184-186
number reduction, 186-189
selection, 195-196
key input selection, 196
with an explicit operational profile, 195
with an implicit operational profile, 195
verification of, 185-186
Operational architecture, 174, 184
Operational complexity, 509-510
Operational development, prioritization, 199
Operational failures cost, 264-265
Operational phase:
evaluation, 306
reliability modeling, 335-339
Operational profile, 15, 183-194, 257-258, 534-535
accuracy, 171
advantage, 167-168
application:
DEFINITY, 200-204
FASTAR, 204-210
PQRS, 210-214
approximation, 168
as educational aid, 200
concepts, 168-170
determination, 236-237
development, 206-208, 211-214
cost, 168, 171
procedure, 170-194
documentation, 203-204
distributed systems, 198-199
elimination of steps, 172
estimation difficulty, 535-537
feature definition, 536-537
feature granularity, 537
multiple and unknown user groups, 537
operational profile for new features, 536
operational profile for new software, 536
granularity, 171
introduction, 167-168
multiple, 173, 237
new software, 536
occurrence probabilities, 189-194
range of operation, 168
reliability estimation, 537-539
coarse features, 538
inadequate test set, 538
interacting features, 538
relationship to design methodology, 173
selection of operation, 195-196
selection of test, 194-197
simple case, 168
updating, 197-198, 213
uses, 199-200
Operational-profile measurement program, 198
Operational-profile-based testing, 202-203, 222
Operational semantics trigger, 388
Operator log, 311
Opinion-based classification, 373
Optimal release time, 82
Optimistic prediction, 122-123, 137, 151
Optimum network architecture, 706-707
Optimum value of failure intensity objective, 229-230
Oracle capability, 569-570, 573
Orthogonal defect classification:
collapsing of defect categories, 379
combination with growth modeling, 379-384
cost, 397
defect trigger (see Defect trigger attribute)
defect type (see Defect type attribute)
definition of, 367
deployment, 396-398
design, 370-371
intuition, 367-370
necessary condition, 371-373
pilot, 396-397
principles, 367-374
production, 396-397
staged production, 396-397
sufficient conditions, 373-374
Orthogonality, 370
Outage, 12-13
duration, 13
reportable, 473
system, 13
Outlier, 779
Output layer, 703
Overall project results for reliability modeling, 292-295
Overview of book, 8-12
P-use (see Predicate use)
Package defect, 374
Parameter estimation, 766-772
Jelinski-Moranda model, 79
methodology, 635-639
Parameter:
coverage, 337, 647
failure intensity decay, 470
reliability profile, 669
Parametric model, 709
Partnering with customers, 444
Path metric, 449
Pathology, fault, 28-29
PBX (see Private branch exchange)
People involved in software reliability engineering, 247-248
Per-office failure rate, 475
Perceptron, 703, 721
Percolation process, 340
Perfective maintenance, 32
Performability analysis, 35, 351-352
fault-tolerant software, 628
Performance:
absolute, 141
degradation by recording function, 198
load testing, 239
loss measure, 761
relative, 141
stress testing, 214
Personnel, maintenance, 174
Pessimistic prediction, 122-123, 137, 151
Phase-based model, 111-112
Physical fault, 18
Plots, 450-454
cumulative failure, 452
failure intensity, 452-453
PLR (see Prequential likelihood ratio)
Plurality decision, 600
Problem reports, preprocessing, 313-314
Point estimation, 766-769
Pointwise availability, 61
Poisson distribution, 749, 765
Poisson process, 74, 765
homogeneous Poisson process, 74
mean, 765
nonhomogeneous Poisson process, 74
variance, 765
Poisson type model, 74-75, 77
Population, 762
moment, 768
Postdelivery phase of software reliability engineering, 242-246
Postprocessing method, 710
Postrelease staff needs, 242-243
Predictability:
next-step, 714
variable-term, 714
Prediction:
availability, 485-486
based on structural decomposition, 162
defects from Ada designs, 112-113
end-point, 711, 714
error:
nature, 136, 151
notion, 145
stationary, 140
fault model, 518-520
long-term, 123-126, 159, 298-299, 456-457, 711
next-step, 714
noise, 128
one-step-ahead, 122
optimistic, 122-123, 137, 151
pessimistic, 122-123, 137, 151
recursive comparison, 128-130
sequential, 711
short-term, 120-123, 711
software reliability, 17
techniques, limitation, 159-160
time to next failure, 447
training, 712
use, 182
Predictive accuracy, methods of analyzing, 128-145
Predictive model and domain coverage, 520-523
Preliminary analysis of data, 312-323
Preliminary selection of software reliability models, 272-273
Premature convergence, 706
Preprocessing:
automatically generated error logs, 313
data, 312-314
human-generated problem reports, 313-314
Prequential likelihood, 281, 291
ratio, 131-134
statistic, 734, 736, 738-739
Prevention of fault, 19-20, 29
Preventive service, 329
Principal component analysis, 326, 495-499
Prioritized operational development, 199
Private branch exchange, 171, 200
Probability:
conditional, 749-750
occupancy, 342
software failure, 628
system failure, 620
theory, 751
transition, 342
unacceptable result, 646
Probability density function, 54, 748
conditional, 749-750
empirical, 773
joint, 767
Probability integral transform, 135
Problem, 362
Problem/Failure Report (P/FR), 267
Problem with root-cause phase analysis, 372-373
Process:
data recording, 190
decision-making, 367
defect prevention, 397
iterative development model, 363
reliability (see Reliability process)
semantic extraction, 366
software reliability engineering, 257
spiral model, 667
usage prediction, 182
waterfall model, 363
Process-based measurement, 372
Process pairs, 328
Process progress for product, 369
Process simulation:
discrete-event, 665
single-event, 678
Process transfer, 396
Processing time, temporal distribution, 505
Processor halt data, 326
Product and process improvement, 246
Product development plan, 228
Product requirements specification, 228
Product, progress through process, 369
Product-in-a-process, 68
Profile, 168
configuration, 202
environmental, 181
execution, 505
fault and error, 314
functional, 169, 176-183, 211-212
supersystem, 173
user, 201-202
Program complexity, relative, 495-504
Program evaluation and review technique, 291
Program length, 496, 513
Program module functionality, 505
Program risk classification, 514
Program size complexity, 496, 522
Program subgraph, 509
Program volume, 496
Programming exception error, 316, 344
Progress of product through process, 369
Project:
CDS, 695-696
DEFINITY, 200-204
FASTAR, 204-210
MIS, 512-515, 719-720
PQRS, 210-214
Space Shuttle flight software, 476-479
Project application objectives, 278, 281
Project characteristics, 281-284
Project characterization, 278-279
Project elements, interdependencies, 668-669
Project investigation, software reliability, 278-299
Project resource schedule, 667
Projection:
field reliability, 259
staffing needs, 242-243
Proliferation of environmental variables, 199
Propagation characteristics of first error, 318-320
further corruption mode, 318-320
no propagation mode, 318-320
quick detection mode, 318-320
Pure birth process, 765
Qualitative causal analysis, 364-366
Quality and performance tests, 427
Quality aspect in discriminant model, 516
Quality attribute, 221
Quality factor assessment, 222
Quality function deployment, 227
Quality improvement, 222,
results, 167-168
Quality measurement from customer perspective, 221
Quality objectives, key, 257
Quality of measurement plan, 471
Quality versus reliability, 221
Quantitative analysis of fault-tolerant software, 635-645
Quantitative reliability requirements, 256
Quantitative requirements, 256
Quantitative statistical defect model, 364-366
R2 statistic, 458, 519
Random sample, 766
Random selection of run types, 194
Random variable:
continuous, 748-749
definition of, 748
discrete, 748-749
exponential, 140
Range of operations in operational profile, 168
Rapid prototyping, 667
Rare events handling, 553-554
Rare situation trigger, 388
Rare-event model, 473-476
Rate function of software reliability model, 686-687
Rate of occurrence of failures, 14, 55, 128 (see also Failure rate function)
Rate-based simulation:
event simulation, 665
reliability simulation, 686-690
reliability simulator architecture, 687-689
simulation algorithms, 676-686
Ratio, prequential likelihood, 131-134
Rayleigh density function, 755
Rayleigh distribution function, 94, 755
Realization (see Sample function)
Recalibrated median, 147
Recalibration of data with respect to usage, 454
Recalibration of software reliability model, 145-149
power of, 147-149
technique, 146-147
Record of error, 308-309
Record of intent, 363
Recording function, performance degradation, 198
Recovery:
error, 340
hardware, 350
software-controlled, 350
Recovery blocks, 581-582
fault tree model, 629-630
Recovery distribution, 320-323
Recovery model, 337-344
Recovery point, 570
Recovery rate:
average, 481
instantaneous, 480
software, 482
Recovery routine, 328
Recovery strategy, 232-233
Recovery trigger, 387
Recurrence, 329-333
automatic diagnosis, 331-333
fault, 315
minimizing, 329-331
time, 342
Recurrent event simulation, 681-682
Recurrent event statistics, 679-681
Recurrent network, 703
fully recurrent network, 704
semirecurrent network, 704
simple recurrent network, 704
Recursive comparison of prediction, 128-130
Rediscovery (see Recurrence)
Rediscovery matrix, 329-330
Rediscovery of failure, 53
Reduced-operation software, 186-187
Reducing interaction between successive runs, 187
Reducing number of input variables, 186
Reducing number of operations, 186-189
Reducing number of run types, 186
Reducing unnecessary interaction, 187
Reduction:
cost, 168
defect, 360
time, 168, 360
Redundancy, 574-575
Redundant software elements, 233
Regression coefficient, 347
Regression faults, 403, 423
Regression model selection, 518-520
Regression test, 196-197, 238
Regression:
eyeball, 694
stepwise, 519
Related faults, 594, 596, 628, 636
Related types of testing, 238
Relative complexity metric, 501
relationship to execution profile, 506-507
Relative error, 128, 458
long-term prediction, 456-457
Relative performance, 141
Relative program complexity, 495-504
Relative subgraph complexity, 509
Release criterion for reliability, 222
Release date approach for reliability specification, 262-263
Release time, 82
Reliability allocation among components, 232
Reliability:
advice to users, 162-163
as release criterion, 222
best estimate of, 259
boundary version, 592
definition of, 29, 35
in presence of interversion failure correlation, 596-604
majority voting, 592-593
module, 589
stable, 50, 57-58, 409, 423-424
system, 36-37, 39-40
trade-off with functionality, 229
ultrahigh, 159, 472
versus quality, 221
Reliability analysis:
calendar-time, 461, 463-469
required fields, 443
usage-based, 461, 469-472
Reliability assessment in artifact simulation, 674-675
Reliability decrease, 50
Reliability demonstration:
chart, 236
testing, 235
Reliability discontinuity, 427
Reliability effect on architecture, 231
Reliability estimates, variation, 550-551
Reliability estimation with inaccurate operational profiles, 537-539
coarse features, 538
inadequate test set, 538
interacting features, 538
Reliability estimation incorporating coverage, 546-547
Reliability evaluation, 64-68
Reliability fluctuation, 433
Reliability function, definition of, 752
Reliability growth, 50, 404, 423
definition of, 403
fault-tolerant software, 65
monotone, 404
strict, 404
Reliability growth assumption, 450
Reliability growth curve:
inflection point, 378
knee, 378
subjective aspects, 377-379
Reliability growth factor, 286
Reliability growth in field, 475-476
Reliability growth modeling, 709-710
linear relationships, 459
Reliability growth testing in system test stage, 237-240
Reliability growth modeling, combination with orthogonal defect classification, 379-384
Reliability measurement for evolving software (see Evolving software)
Reliability model:
acceptance voting, 603-604
comparison with simulation, 694-696
consensus recovery block, 601-603
consensus voting, 592-593, 599-601
N self-checking programming, 633
N-version programming, 631
distributed recovery block, 630
fault-tolerant software, 585-596
trend analysis, 424
Reliability modeling, 53-60, 257-258
conditional reliability, 55, 57-58
data domain, 588-594
N-version programming, 591-593
comparison, 593-594
recovery block, 590-591
interval reliability, 55, 58
operational phase, 335-339
time domain, 588, 594-596
reliability decrease, 50
reliability growth, 50
results, 291-292
stable reliability, 50, 57-58
tools, 273
with software complexity metrics, 524-526
Reliability objectives, 230-231, 257-258
achieving, 232-233
availability consideration, 231
certification, 240-241
influential factors, 230-231
reliability effect on architecture, 231
Reliability overestimation due to saturation, 545-546
Reliability prediction, 81-82, 713-718
definition of, 17
Duane model, 99
experiment, 713-718
geometric model, 101
hyperexponential model, 91
Jelinski-Moranda model, 79
Littlewood-Verrall model, 107-108
Musa basic execution time model, 89-90
Musa-Okumoto logarithmic Poisson model, 103
Nonhomogeneous Poisson process model, 81-82
Schneidewind model, 84-86
S-shaped model, 97-98
Weibull model, 95
Reliability process, 661, 665-669
definition of, 661
end-to-end, 691
fault identification, 667
fault repair, 667
nature, 666-667
primary event, 689
repair validation, 667
retest, 667
schedule items, 668
secondary event, 689-690
simulation for project application, 690-696
software construction, 666
software correction, 666
software inspection, 666
software integration, 666
structures and flows, 667-668
test preparation, 667
testing, 667
Reliability profile parameters, 669
Reliability requirements specification, 256, 259-266
definition of, 256
life-cycle cost optimization, 263-266
product, 228
release date approach, 262-263
system balance method, 261-262
testable requirements, 260
untestable requirements, 261
Reliability simulation, 662-665, 686-690
rate-based, 686-690
simulator architecture, 687-689
Reliability theory, 751-758
Reliability trend test (see Trend test)
Reliability validation, 257, 259
Remaining set of run types, exclusion, 188-189
Removal of fault, 19-20, 29
Renewal process, 56
Renewal theory, 54
Repair rate, field, 482
Repetitions of failures, 244
Reportable events, 473, 483
outages, 473
Reporting, manual, 308, 310-312
Request for proposal, 226
Requirement of dependability, 65-66
Residual risk, 561
Residuals of fit, 455
Residue, 738
Resource, usage function, 346
Resource-usage model, 349-351
Restart and checkpointing, 328
Restoration, 34, 49-50
Result verification, 570-574
Results:
application, 285-289
artifact simulation, 675-676
neural network for error-prone module identification, 723-725
neural network for reliability prediction, 715-718
overall project for reliability modeling, 292-295
quality-improvement, 167-168
reliability modeling, 291-292
simulation experiments, 691-694
Reuse, software, 20, 234
Review and inspection, 234
Review and inspection trigger, 387-388
Review of discovery, 231
Reward functions, 334, 351
Reward, cumulative, 352
Reward rate, 351-352
instantaneous, 760
interval, 760
steady-state, 335, 760-762
Reward-loss rate, steady-state, 335
Risk, 472
residual, 561
testing-based model, 554-555
testing-based notation, 555
untested code, 554
Risk assessment, 376-384, 472-473, 555-558
using defect types, 376-384
Risk browser, 560-561
Risk classification, 514-515
Risk computation example, 558-559
Risk confidence growth model, 562
Risk model, 472, 554-563
software reliability, 561-563
Risk prediction result, 715-718
ROCOF (see Rate of occurrence of failures)
Rome Laboratory model, 113-114
Root-cause analysis cost, 372-373, 397
R-squared statistic, 458, 519
Run, 169, 185
Run category, 194
near homogeneous, 194-195
Run-time overhead, 563
Run type, 169
group, 169
homogeneous, 194
number reduction, 186
Safety, 29, 35
Safety analysis, 624
Safety validation using fault trees, 623-627
Sample distribution function, 135
Sample error entry, 308-309
Sample function, 677
Sample moment, 768
Sample space, 747
Sampling users, 190
Saturation effect, 541-542, 562
empirical basis, 543-544
reliability overestimation, 545-546
Scaled relative complexity, 501
Schick-Wolverton model, 94
Schneidewind model (see Model: Schneidewind)
Scope of book, 21-23
Secondary event simulation, 682-683
Security, 29
SEI software process, 477
Selection:
operation, 195-196
regression model, 518-520
with replacement, 195
without replacement, 195
Self-assessment in software reliability engineering, 250-252
Self-checking capability, 569-570
Self-checking software component, 633
Self-driven simulation, 664, 672-673
Semantic classification, 373
Semantic extraction of process, 366, 372, 375
Semantic information, 368
Semantics, connection with measurement, 372
Sensitivity analysis, 338-339
Sequential prediction, 131, 711
Serialization defect, 374
Service, 12, 34
corrective, 329
preventive, 329
Service utilization matrix, 48
Setting up data collection process, 266-267
Severity assignment, 244
Severity classification (see Severity classification of failure)
Severity classification of failure, 12, 227-228, 477-478
Severity of failures, 477
Short-term prediction, 120-123, 711, 714
Side effect of code update, 314-315
Sigmoidal unit, 702
Significance level, 775
Significant environmental conditions, 174
Similar faults, 576, 594, 596
Simple operational profile, 168
Simple path coverage trigger, 389
Simple short-term prediction, 120-123
Simulated code generation, 672-674
Simulated defect status profile, 691-693
Simulated fault status profile, 691-693
Simulation, 664
artifact-based, 665, 669-676
compared with reliability model, 694-696
discrete-event, 665
event process, 676
limited growth, 683
rate-based, 665
recurrent event, 681-682
reliability, 662-665
secondary event, 682-683
self-driven, 664, 672-673
trace-driven, 664
Simulation algorithms, rate-based, 676-686
Simulation approaches, dynamic, 664-665
Simulation experiments and results, 691-694
display of results, 689-690
Simulation model, 206, 662
Simulation program structure for single event occurrence, 678-679
Simulation time choice, 678-679
Simulator architecture:
artifact-based, 670-675
rate-based, 687-689
Simulator program, 685
Single-event occurrence simulation, 678-679
Single-interpreter system, 43-47
Single-layer network, 703
Single point of failure, 619
Smallest divisions of work, 169
Smoothing of software behavior, 409
Smoothing technique, 161
Soak test, 197
Soak time, 197
Software:
data recording, 190
middleware, 390
reduced-operation, 186-187
Software artifact, 663
Software behavior, smoothing, 409
Software complexity and software quality, 510-523
Software complexity metrics for reliability modeling, 524-526
Software configuration trigger, 387
Software defects, 361-364
Software diversity, 567
Software elements, redundant, 233
Software environment characteristics, 669
Software error model, 341
Software evolution (see Evolving software)
Software failure correlation, 324-325, 598-599
Software failure data collection and analysis, 440
Software failure mode, 754
Software failure: (see also Failure)
hardware-related, 327-328
load-induced, 348
probability, 628
with workload parameters, 345-352
Software fault:
model, 318
nature, 385
Software fault tolerance (see Fault-tolerant software)
Software field data, 440
Software life cycle, 306
Software metrics, 495-496
Software process, SEI, 477
Software quality:
models, 511
multivariate analysis, 514-518
relationship to software complexity, 510-523
relationship to software reliability, 6
Software recovery, 350
model, 341
recovery rate, 482
Software reliability:
affecting factors, 339
early prediction, 17
expression, 259-261
definition of, 5, 540
need for, 3-5
prediction, 17
relationship to software quality, 6
Software reliability and neural networks, 709-718
Software reliability and risk model, 561-563
Software reliability estimation, 17
with time and structure, 539-554
Software reliability engineering:
activities, 223-225
approaches, 220-226
benefits, 210, 220-226
cost, 222-223
operational profile cost, 222
processing and analyzing failure data cost, 223
software reliability engineering consultant cost, 223
definition of, 8
design and implementation phase, 232-235
design stage, 232-233
feasibility and requirement phase, 226-231
feasibility stage, 226-228
field trial stage, 241-242
getting started, 246-252
implementation on existing projects, 224
implementation on short-cycle projects, 226
implementation stage, 233-235
incremental implementation, 223, 225
information and support, 250
involved people, 247-248
objective, 219
organization preparation, 247-250
postdelivery and maintenance phase, 242-246
practice, 219-220
process, 247, 257
requirement stage, 228-231
self-assessment, 250-252
success story, 221-222
system test and field trial phase, 235-242
system test stage, 237-241
technical areas, 19-22
fault prevention, 19-20, 29
fault removal, 19-20, 29
fault tolerance, 19-20, 29
fault/failure forecasting, 19, 21, 29
Software reliability measurement, 17
early prediction, 17
estimation, 17
framework, 256-278
prediction, 17
reliability prediction, 17
Software reliability modeling tool, 449 (see also Software reliability tool)
Software reliability model, 17-18
preliminary choice, 272
preliminary selection, 272-273
rate function, 686-687
relationship to hardware reliability, 18-19
tool, 449 (see also Software reliability tool)
Software reliability prediction, 17
Software Reliability Program (SoRel) tool, 737-739
Software reliability project investigation, 278-299
Software reliability requirement, 67
Software reliability requirements, establishment, 259-266
Software Reliability Simulator (SoftRel) tool, 688
Software reliability study goals, 441-442
Software reliability testing, 68
Software reliability tool:
AT&T toolkit (see AT&T software reliability engineering toolkit)
CASRE (see Computer-Aided Software Reliability Estimation tool)
ESTM (see Economic Stop Testing Model tool)
SMERFS (see Statistical Modeling and Estimation of Reliability Functions for Software tool)
SoftRel (see Software Reliability Simulator tool)
SoRel (see Software Reliability Program tool)
SRMP (see Statistical Modeling and Reliability Program tool)
tool comparisons, 741-744
Software reuse, 20, 234
Software risk (see Risk)
Software science metrics (see Halstead number)
Software system, 12
Software testing, 20
Software-specific classifiers, 442-443
Source of failure, 65-66
Space complexity, 681
Space Shuttle flight software, 476-479
Spawned faults, 196
Spearman test, 411, 737
Special events, 472-479
Specification: (see also Requirements specification)
product requirements, 228
reliability requirements, 261-266
Spectrum of defect analysis, 364-367
Spiral model, 667
Squared distance, generalized, 515
SRE (see Software reliability engineering)
S-shaped model, (see Model: S-shaped model)
Stability index, 231
Stability test program execution, 209-210
Stable reliability, 50, 57-58, 409, 423-424
Stack trace, 332
Standard assumptions of software reliability models, 78
Standardization of data collection and analysis process, 440
Standardized transformation matrix, 498-499, 514-515
State space, 751
State transition equation, 759
State-transition-rate diagram, 764
Statement coverage testing, 533
State of error, 340
Static analysis and trend analysis, 431-434
Static analysis conduct, 431-433
Static measure, 502
Stationary prediction error, 140
Stationary process, 56
Statistic:
Braun, 130
chi-square (see Chi-square statistic)
c2 (see Chi-square statistic)
CP, 520
F-statistics, 717
prequential likelihood (see Prequential likelihood: statistic)
R2 (r-squared), 458, 519
Statistical analysis and visualization system, 449
Statistical cluster analysis, 348
Statistical defect model, 364-366
Statistical Modeling and Estimation of Reliability Functions for Software (SMERFS) tool, 733-735
Statistical Modeling and Reliability Program (SRMP) tool, 736
Steady-state availability, 483-484
Steady-state reward rate, 335, 760-762
Stepwise regression, 519
Stochastic independence assumption, 403, 580, 588
Stochastic process, 750-751
doubly 346
Stochastically smaller than, 403
Stop testing criteria, 467, 740
Stopping rule in selecting principal components, 498
Storage exception error, 316, 344
Storage management error, 316
Stress test program execution, 209-210
Strict reliability growth, 404
Structural decomposition for prediction, 162
Structure-based software reliability estimation, 539-554
Structures and flows in reliability process, 667-668
Student's t-distribution, 770
Subadditive property, 404, 407-408
analysis, 406-407
graphical interpretation, 404-406
relationship to Laplace test, 414-419
summary, 409-410
Subadditivity, 404
Subadditivity factor, 406
constant, 408-409
for validation phase, 417, 419
Subgraph of program, 509
Subjective aspects of growth curves, 377-379
Subpopulation identifier, 395
Subprocess representation, 689
Success story, software reliability engineering, 221-222
Sum of squared errors, 706, 769
Superadditive property, 404 (see also Subadditive property)
Superadditivity, 404 (see also Subadditivity)
Supersystem profile, 173
Survival function, 54
Symptom, difference with trigger, 386
Symptom-based diagnosis, 332
System, 174
command-driven, 191
data-driven, 193
System activity impact, 345-352
System administrator, 174
System availability, 483
System balance method for reliability specification, 261-262
System complexity, 500
System dependability, relationship to software reliability, 7
System dynamics, 665
System failure rate, 49
System failure, probability 620
System for manipulation, filtering, and data merging, 449
System for statistical analysis and visualization, 449
System interpreters, 43
System-level analysis:
case study, 651-657
hardware and software system, 645-657
System-level software reliability:
model, 336, 338
parameters, 336
System life, 756
System made up of components, 41-49
System mode, 174-176
System model, 41-43
System outage, 13
causes, 439-440
System reliability, 36-37, 39-40
continuous-time, 36-37, 40
discrete-time, 36, 39-40
System reliability model:
N self-checking programming, 650
N-version programming, 648-650
distributed recovery block, 647-648
System safety model:
N self-checking programming, 650
N-version programming, 648-650
distributed recovery block, 647-648
System start-up and restart trigger, 387
System structure model, 646
System test phase of software reliability engineering, 235-242
reliability growth modeling, 241-242
reliability growth testing, 237-240
test data collection, 239
System test triggers, 387
System unavailability, 483
t-distribution, 770
Tandem project report (TPR), 311
Task computation, 628
Task, feature-oriented, 175
Technical areas for software reliability engineering, 19-22
fault prevention, 19-20, 29
fault removal, 19-20, 29
fault tolerance, 19-20, 29
fault/failure forecasting, 19, 21, 29
Technique:
combination, 161-162
data analysis 450
discrimination, 515
historical validation, 456
smoothing, 161
Technology transfer, 396
Temperature coefficient, 702
Temporal distribution of processing time, 505
Temporary behavior of software, 409
Test,
arithmetical mean, 411-412
chi-square, 714, 774-775
Gnedenko (see Gnedenko test)
Hollander (see Hollander test)
Kendall (see Kendall test)
Kolmogorov-Smirnov, 714, 775-776
Laplace (see Laplace test)
Spearman (see Spearman test)
Test automation, 238
Test compression factor, 259
Test coverage trigger, 389
Test efficiency, 422
Test environment development, 208
Test interaction trigger, 389
Test selection, 194-197
coarse grain selection, 194
fine grain selection, 194
random selection of run types, 194
selection with replacement, 195
selection without replacement, 195
Test sequencing trigger, 389
Test set:
adequacy criterion, 534
inadequate, 538
Test variation trigger, 389
Testable reliability requirement, 260
Testing:
accelerated, 175
acceptance, 570-571, 581
back-to-back, 606
control flow, 532-533
customer-like, 202-203
data flow coverage, 533
data flow, 532-533
decision coverage, 533
domain, 554
functional, 268-270, 427, 534
increased, 175
kinds of, 532
mutation, 532-533, 554
operational-profile-based, 202-203, 222
performance and stress, 214
progress control, 422
quality and performance, 427
related types, 238
reliability demonstration, 235
software 20
software reliability, 68
statement coverage, 533
varying occurrence probabilities, 214
white-box, 532
Testing and operation trade-off, 265
Testing-based notation of risk, 555
Testing cost, 265
Testing effort, 540-541, 547
Testing hypothesis, 410-411, 450, 774
Testing methods, limits, 542-543
Testing overview, 532-534
Testing progress tracking, 240
Testing tool, 213-214
Testing with replacement, 277
Testing without replacement, 277
Three-point symmetric moving average, 465-466
Threshold function:
hard-limiting, 701
linear, 701
Time, 13-14, 444-445
calendar time, 14
clock time, 14
execution time, 14
in-service, 445, 460
logical, 445
translation, 14
usage, 445
Time choice, simulation, 678-679
Time complexity, 679, 681, 683
Time domain approach, 73
Time measurement, 239-240
Time reduction, 168
Time to error, 320-322
Time to failure, 320, 447
Time to recovery, 322-323
Time to repair, 320
Time units, 419
Time-between-failures data, 16, 268, 271
Time-between-failures models vs. failure-count models, 288-289
Time-domain analysis (see Time-domain reliability modeling)
Time-domain reliability modeling, 588, 594-596
Timing defect, 374
Tolerance, 578-581
Tolerance of fault, 19-20, 29
Tool:
commercial, 730
defect-tracking, 394
general-purpose, 730
reliability modeling, 273
testing, 213-214
Tool comparison, 703
software reliability tool, 741-744
Tool for software reliability modeling, 449
Tool selection criteria, 731-732
Total hazard, 677
Trace-driven simulation, 664, 674
Trachtenberg's general theory, 464
Tracking customer satisfaction, 245
Tracking testing progress, 240
Trade-off between testing and operation, 265
Trade-off studies, 229-230
reliability and functionality, 229
Traffic condition, 175
Training data selection, 723
Training output layer, 708
Training regime, 712
Training set, 705, 722
Trajectory of input, 38-39, 402 (see also Failure: domain)
Transfer, process, 396
Transfer, technology, 396
Transformation of data (see Data transformation)
Transformation of reliability-related functions, 753
Transformation between data types, 17
Transient behavior of software, 409
Transient hardware faults, 628
Transient region, 482, 484
Transition probability, 342
Translation, time , 14
Tree-based approach, 202
Trend analysis:
definition of, 410
reliability models, 424
static analysis, 431-434
trouble reports, 433-434
Trend change, 407-408
Trend test, 410-419, 737
analytical, 410
graphical, 410
Trial-run for data collection, 267
Trigger (see Defect: trigger)
Trouble reports, 6, 244, 431
trend analysis, 433-434
Trouble tracking system, 244
Two-class risk classification model, 515
Two-out-of-N voting, 577
Type of classification error, 516
Type of connectivity, 703
u-plot, 135-140, 281, 734, 736, 739
Ultrahigh reliability (see Ultrareliability)
Ultrareliability, 159, 472
modeling, 277-278
operation, 237
Unacceptable result probability, 646
Unavailability, 62-64, 483
empirical, 481-482
uptime, 485
Understanding customer expectations, 206
Unequally weighted linear combination model, 291
Unique factor, 778
Unique variance, 778
Units of time, 419
Unnecessary interaction, reduction of, 187
Unrelated fault (see Independence: fault)
Unreliability calculation, 621
Untestable reliability requirement, 261
Updating operational profile, 197-198, 213
Uptime availability, 481, 483
Uptime unavailability, 485
Usage, 445-446
fields, 446
function of resource, 346
metrics, 499-500
profile of users, 463
time, 445
Usage-based reliability analysis, 461, 469-472
User, 173, 175
experience, 175
generic, 201
profile, 201-202
sampling, 190
telecommunications, 174
type, 173-174
usage profile, 463
Utilization environment, variation, 403
V&V (see Verification and validation)
Validating data set, 513
Validating new measure, 521
Validation phase, subadditivity factor, 417, 419
variability, 128, 281
field reliability data, 439
functional complexity, 509
Variable-term predictability, 714
Variables, environmental, 175-176, 180-181, 190, 192
Variance, 748
analysis of, 716
Poisson process, 765
Variants (see Alternates)
Variation:
between-cluster, 779
model accuracy, 126
reliability estimates, 550-551
utilization environment, 403
within-cluster, 779
Varying occurrence probabilities for testing, 214
Verification:
operation, 185-186
result, 570-574
Verification and validation, 33-34
Version, 52
Version control, 235
Versions (see Alternates)
Voter, 577, 582
Voting, 577
back-to-back, 582
consensus, 578
majority, 577
mean value, 578
median, 577, 583
two-out-of-N, 577
Walk-through, 432
Waterfall process, 363
Wear-out failure, 754
Weibull and gamma failure time class model, 93-98
Raleigh (see Model: Raleigh)
Schick-Wolverton (see Model: Schick-Wolverton)
S-shaped (see Model: S-shaped)
Weibull (see Model: Weibull)
Weibull-type field-defect model, 464-465
Weight determination based on ranking model results, 290-291
White-box testing, 532
Work, smallest division, 169
Workload:
CPU-bound, 349-350
I/O, 349
Workload model, 348-349
Workload parameters and software failures, 345-352
Workload stress trigger, 387
Workload volume trigger, 387
Workload-dependent cyclostationary model, 346
Weight adjustment phase, 705
Weight determination based on changes, 291
X-ware, 27, 35-49
atomic systems, 35-41
system made up of components, 41-49
multi-interpreter system, 47-49
single-interpreter system, 43-47
system models, 41-43
Yamada modified exponential growth model, 91
y-plot, 140-141, 281, 734, 736, 739
![[^]](images/ManRightArrow1b.gif){kind=small}
A
![[^]](images/ManContents.gif){kind=small}