Advanced Transaction Models and Architectures

Pirmais vāks
Sushil Jajodia, Larry Kerschberg
Springer Science & Business Media, 1997. gada 31. aug. - 381 lappuses
Motivation Modem enterprises rely on database management systems (DBMS) to collect, store and manage corporate data, which is considered a strategic corporate re source. Recently, with the proliferation of personal computers and departmen tal computing, the trend has been towards the decentralization and distribution of the computing infrastructure, with autonomy and responsibility for data now residing at the departmental and workgroup level of the organization. Users want their data delivered to their desktops, allowing them to incor porate data into their personal databases, spreadsheets, word processing doc uments, and most importantly, into their daily tasks and activities. They want to be able to share their information while retaining control over its access and distribution. There are also pressures from corporate leaders who wish to use information technology as a strategic resource in offering specialized value-added services to customers. Database technology is being used to manage the data associated with corporate processes and activities. Increasingly, the data being managed are not simply formatted tables in relational databases, but all types of ob jects, including unstructured text, images, audio, and video. Thus, the database management providers are being asked to extend the capabilities of DBMS to include object-relational models as well as full object-oriented database man agement systems.

No grāmatas satura

Saturs

Transactions in Transactional Workflows
3
12 Advanced Transaction Models
6
121 Nested Transactions
7
124 MultiLevel Transactions
8
13 Transactional Workflows
9
131 Previous Research on using Transactions for Workflows
10
14 Workflow Recovery
13
141 Transaction Concepts in Modeling Workflow Recovery
14
Concurrency Control and Recovery
181
Customizable Concurrency Control for Persistent Java
183
711 Overview of Persistent Java
184
712 Customizable Concurrency Control
186
721 Transactions as Java objects
187
723 Implicit transaction semantics
190
725 Outline of the modified JVM
191
74 Transaction Shell
195

15 Workflow Error Handling
17
16 Transactions ATMs and Recovery in LargeScale WFMs
18
161 Error Handling and Recovery in the METEOR WFMS
20
A Distributed Implementation of the METEOR2 WFMS
22
1613 Modeling Errors in METEOR2
23
1614 Recovery Framework in ORBWork
25
Beyond Database Transactions
28
18 Conclusion
31
A Normative Perspective
34
WFMS the Next Generation of Distributed Processing Tools
35
22 Workflow Management Systems
37
222 Process Representation
38
225 Architecture
42
2232 Runtime Architecture
44
224 Process Execution
45
23 Functionality and Limitations of Workflow Management Systems
47
232 Scalability
49
233 Industrial Strength
51
24 Evolution of Workflow Management Systems
53
242 Process Support Systems
54
243 Programming in Heterogeneous Distributed Environments
55
25 Conclusions
57
ToolKit Approaches
61
The Reflective Transaction Framework
63
31 Introduction
64
32 Extending a Conventional TP Monitor
66
33 The Reflective Transaction Framework
68
331 Extensions Through Transaction Events
69
332 Implementing Reflection and Causal Connection
70
333 A Separation of Programming Interfaces
73
34 Applications of the Reflective Transaction Framework
75
342 Implementing SemanticsBased Concurrency Control
80
35 Conclusion
87
Flexible Commit Protocols for Advanced Transaction Processing
91
41 Introduction
92
42 Overview of Our Approach
94
422 Illustrative Example
98
43 An Example of Transaction Dependencies
99
44 Primitives for Flexible Commit
103
442 New Primitives
104
443 Discussion
108
45 Realizing Various Transaction Dependencies
109
451 ACTA Framework
110
452 Sagas
111
453 Workflows and Long Lived Activities
113
4531 Semiatomicity
114
454 Secure Distributed Transactions
116
455 Contingent Transactions
119
456 Nested Transactions
121
46 Conclusions and Future Work
123
Long Transactions and Semantics
125
Contracts Revisited
127
511 The Motivation For ConTracts
128
512 A Brief Survey of the Model
129
52 Transactions in a Workflow Environment
132
522 Semi Transactiona1 Activities
133
53 Reconsidering Correctness
134
5312 Permeability
135
532 Recovery and Serializability
136
534 Execution Histories and Correctness
138
54 Compensation in Detail
141
542 Scriptbased Compensation
144
543 Comprehensive Compensation
146
544 Partial Compensation
149
55 Summary
150
SemanticBased Decomposition of Transactions
153
62 Related Work
156
63 The Hotel Database
157
64 The Model
159
641 A Naive Decomposition of the Reserve Transaction
160
642 Generalizing the Original Invariants
161
643 Compensating Steps
162
644 Semantic Histories
163
65 Properties of Valid Decomposition
165
653 Consistent Execution Property
166
655 Successful Execution Property
167
6611 Composition Property
168
6613 Consistent Execution Property
169
662 An Invalid Decomposition
171
67 Successor Sets
172
68 Concurrent Execution
175
682 Concurrency Control Mechanism
176
6821 Algorithms
177
6822 Discussion
179
75 Locking Capabilities
197
751 Ignoring Conflicts
198
752 Delegation
201
754 Summary
202
761 Flat Transactions
203
762 Nested Transactions
204
77 Related Work
208
78 CONCLUSION
210
Toward Formalizing Recovery of Advanced Transactions
213
82 The Formal Model
215
821 Modeling Recovery through Histories
217
522 Events Histories States
219
83 Requirements Assurances Rules
221
832 Failure Atomicity
222
834 Assurances for Failure Atomicity
224
836 Recovery Mechanisms Rules
225
837 Logging and Commit AbortProtocols
226
8411 Data Structures
227
841 a Normal Processing
228
8413 Crash Recovery
229
842 Formalizing some properties of ARIES and ARIESRH
230
843 Proof Sketches
232
Transaction Optimization
235
Transaction Optimization Techniques
237
91 Introduction
238
911 What is Wrong with the Current Architecture?
239
912 How Should We Change the Architecture?
241
913 Chapter Organization
242
914 Related Work
243
93 A Novel Transaction Optimization Strategy
245
931 PreAccess Optimization
246
932 PostAccess Optimization
249
94 Query Optimization Issues
253
942 Interim Replication
254
ECA Approach
257
An Extensible Approach to Realizing Advance Transaction Models
259
1011 Goals
261
1012 Related Work
262
102 Our Approach
263
1021 Realizing Transaction Models using ECA rules
265
103 Implementation Details
268
1032 Making Zeitgeist Active at the Systems Level
270
104 Realizing Transaction Models
272
105 Extensibility
274
106 Conclusions
275
OLTPOLAP
277
Inter and Intratransaction Parallelism for Combined OLTPOLAP Workloads
279
112 Background on MultiLevel Transactions
282
113 The Plenty Architecture
283
114 Granularity of Parrelism
284
115 Transaction Management Internals
288
116 Scheduling Strategies
291
117 An Application Study
294
118 Conclusion
297
RealTime Data Management
301
Toward Distributed RealTime Concurrency and Coordination Control
303
122 Responsiveness and Consistency
306
1222 More Elaborate Coordination Control
307
123 Enabling Technologies
308
1232 Characterization Efforts
309
1233 Performance Studies
310
1235 Applicationspecific Approaches
311
1241 Providing RTR at Local Sites
313
12431 Level A
314
125 Synchronization Using Application Semantics
315
1251 Relaxed Atomicity
316
1252 Communication Level Approaches
317
Mobile Computing
319
Transaction Processing in Broadcast Disk Environments
321
132 Motivation for Weakening Serializability
324
133 Formalization of Consistency Requirement
326
1332 Fonnalization of Requirements
327
1333 Comparison with View Serializability
329
134 Weakened Requirements
330
1341 Motivation for Weaker Requirements
331
1342 Weakened Requirements
332
135 Mechanisms to Guarantee Correctness
333
1351 Broadcast Disks
334
13522 Client Functionality
335
1353 Proof of Correctness
337
References
339
Contributing Authors
365
Index
375
Autortiesības

Citi izdevumi - Skatīt visu

Bieži izmantoti vārdi un frāzes

Par autoru (1997)

Dr. Sushil Jajodia is Professor and Chairman of the Dept. of Information and Software Engineering, and Director of the Center for Secure Information Systems at the George Mason University, Fairfax, Virginia, USA

Bibliogrāfiskā informācija