Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

Enterprise Information Systems > Exam (protocols) > Flashcards

Exam (protocols) Flashcards

1
Q

Which are the different kinds of heterogeneity? Explain each one of them.

A
  • Technical Heterogeneity: how to exchange data (communication protocols (HTTP, SOAP), exchange formats (binary, text, XML), APIs (JDBC, ODBC, proprietary), query mechanisms (forms, query languages), query languages.
  • Data Model Heterogeneity (hierarchical, relational, object-relational, XML, etc)
  • Syntactic Heterogeneity (binary representation, encoding, separators, textual representation)
  • Structural Heterogeneity: modeling identical application concepts differently using the same modelling concepts in the same data model)
  • Schematic Heterogeneity: modeling identical application concepts using different modelling concepts in the same data model.
  • Semantic Heterogeneity: different representation of identical application concepts OR identical representation of different application concepts.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Which are the two types of information integration?

A

Materialized Integration: replication, data warehousing.

Virtual Integration: distributed DBMS, federated DBMS, mediator-based systems, multi-database systems.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Which are the architectures for virtual data integration ?

A
  • Distributed DBMS
  • Federated DBMS
  • Mediator-based Systems
    (the above are based on a global schema, supporting location and distribution transparency)
  • Multi-database Systems
    (no global schema, only location transparency)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Which methods wrappers provide to be used by Garlic work requests?

A

plan_access()
plan_join()
plan_bind()

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Which are the two types of replication?

A
  • Eager replication (avoids inconsistencies, update overhead, lack of scalability, cannot be used if nodes are disconnected)
  • Lazy replication (low overhead, works if sites are disconnected, stale data, conflicting updates may cause inconsistencies)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Which are the two types of replication methods?

A
  • Target table refresh

- Change-capture

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

How to detect concurrent update conflicts in change-capture replication?

A
  • timestamp
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Which are the two possibilities for ownership of replicas?

A
  • Group

- Master

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Which are the two approaches for reconciliation of conflicts?

A
  • automatically, based on rules

- manually

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

How to represent OLAP in a relational schema?

A
  • Star structure

- Snowflake structure (normalized)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Which are the components of a Data Warehouse Architecture?

A
  • Data Warehouse Manager
  • Monitor
  • Extractor
  • Transformer
  • Loader
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Which are the four approaches for the monitor of a DW to detect changes?

A
  • log-based
  • trigger-based
  • audit columns (application-based, don´t support deletion)
  • snapshot differentials
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is the main objective of transformers in DW?

A

Schema and data integration logic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Types of DW refresh?

A
  • Full load vs. Incremental load

- Periodic vs. driven by source updates vs. on request

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Which are the two update alternative to avoid conflicts?

A
  • Fragmentation by key

- Fragmentation by time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Alternative to detect conflicts.

A
  • Semantic synchronization (incoming replica transaction update must pass acceptance test)
17
Q

Which are the 4 alternatives for replication activity?

A
  • continuous
  • interval-based
  • triggered by DB-events
  • one-time snapshot
18
Q

Which are the two types of queues in TP-Monitors?

A
  • Short-term queue for load control

- Persistent queues

19
Q

Advantages of stratified transactions?

A
  • less resource contention
  • higher throughput
  • if all transactions in a stratum execute on the same node, there is no network traffic for 2PC
20
Q

What to do if client is non-transactional?

A
  • queue operation guarantees message is stored before returning to client
  • provide ID of last message of which a request was submitted
  • allow client to explicitly acknowledge receiving a reply
21
Q

Which are the main characteristics of MQS?

A
  • asynchronous transaction processing
  • persistent message queues
  • loosely coupled (client not blocked, server works when wants, components don´t need to be active)
22
Q

What can affect the sending order of messages in MQS?

A
  • message priority

- transaction order of messages

23
Q

What can affect the receiving order of messages?

A
  • selectors (on header and property fields only)
24
Q

What is the difference between transactional sessions and non-transactional sessions?

A
  • Transactional: messages are automatically acked at TA commit. Queues are recovered automatically at rollback.
  • Non-transactional: needs ack:
    - lazy acks
    - auto-ack (after successful receive)
    - client ack (explicitly)
25
Q

Which are the main characteristics of Message Brokers?

A
  • focus on application integration
  • message transformation
  • message routing and transport
  • message repository
  • message warehouse
26
Q

Which are the three topologies of Message Brokers?

A
  • Hub-and-spoke
  • Multi-hub
  • Federation
27
Q

Which are the two roles of DBMS in a messaging world?

A
  • persistence manager

- provide advanced DBMS capabilities (buffer, index, row-level locking, logging services)

28
Q

Which are the three integration challenges?

A
  • Autonomy (design, interface, access, judicial)
  • Distribution (physical or logical)
  • Heterogeneity
29
Q

Architecture components of a Federated DB.

A
  • local schema (local schema design)
  • component schema (data model transformation)
  • export schema (defining schema sub-set)
  • federated schema (schema integration, mapping)
  • external schema (tailored to app / user group)
30
Q

Which are the 5 steps in information integration?

A
  • Analysis (determined requirements of integrated schema)
  • Discovery (find/identify relevant data sources)
  • Planning (resolve heterogeneity)
  • Deployment (set integration plan in runtime environment)
  • Runtime
31
Q

Which are the 3 information integration approaches?

A
  • Bottom-up (analysis = discovery)
  • Top-down (analysis before discovery)
  • Hybrid (analysis and discovery intertwined)

Enterprise Information Systems (2 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions