FINALTERM EXAMINATION
Spring 2010
CS614- Data Warehousing (Session - 3)
Time: 90 min
M a r k s: 60
Question No: 1 ( M a r k s: 1 ) http://vuzs.net
A data warehouse may include
► Legacy systems
► Only internal data sources
► Privacy restrictions
► Small data mart
Question No: 2 ( M a r k s: 1 ) http://vuzs.net
De-Normalization normally speeds up
► Data Retrieval
► Data Modification
► Development Cycle
► Data Replication
Question No: 3 ( M a r k s: 1 ) http://vuzs.net
In horizontal splitting, we split a relation into multiple tables on the basis of
► Common Column Values
► Common Row Values
► Different Index Values
► Value resulted by ad-hoc query
Question No: 4 ( M a r k s: 1 ) http://vuzs.net
Multidimensional databases typically use proprietary __________ format to store pre-summarized cube structures.
► File
► Application
► Aggregate
► Database
Question No: 5 ( M a r k s: 1 ) http://vuzs.net
A dense index, if fits into memory, costs only ______ disk I/O access to locate a record by given key.
► One
► Two
► lg (n)
► n
Question No: 6 ( M a r k s: 1 ) http://vuzs.net
All data is ______________ of something real.
IAn Abstraction
IIA Representation
Which of the following option is true?
► I Only
► II Only
► Both I & II (P# 181)
► None of I & II
Question No: 7 ( M a r k s: 1 ) http://vuzs.net
The
key idea behind ___________ is to take a big task and break it into
subtasks that can be processed concurrently on a stream of data inputs
in multiple, overlapping stages of execution.
► Pipeline Parallelism
► Overlapped Parallelism
► Massive Parallelism
► Distributed Parallelism
Question No: 8 ( M a r k s: 1 ) http://vuzs.net
Non uniform distribution, when the data is distributed across the processors, is called ______.
► Skew in Partition (P # 218)
► Pipeline Distribution
► Distributed Distribution
► Uncontrolled Distribution
Question No: 9 ( M a r k s: 1 ) http://vuzs.net
The
goal of ideal parallel execution is to completely parallelize those
parts of a computation that are not constrained by data dependencies.
The smaller the portion of the program that must be executed __________,
the greater the scalability of the computation.
► None of these
► Sequentially
► In Parallel
► Distributed
Question No: 10 ( M a r k s: 1 ) http://vuzs.net
If
‘M’ rows from table-A match the conditions in the query then table-B is
accessed ‘M’ times. Suppose table-B has an index on the join column. If
‘a’ I/Os are required to read the data block for each scan and ‘b’ I/Os
for each data block then the total cost of accessing table-B is
_____________ logical I/Os approximately.
► (a + b)M
► (a - b)M
► (a + b + M)
► (a * b * M)
Question No: 11 ( M a r k s: 1 ) http://vuzs.net
Data
mining is a/an __________ approach, where browsing through data using
data mining techniques may reveal something that might be of interest to
the user as information that was unknown previously.
► Exploratory
► Non-Exploratory
► Computer Science
Question No: 12 ( M a r k s: 1 ) http://vuzs.net
Data
mining evolve as a mechanism to cater the limitations of ________
systems to deal massive data sets with high dimensionality, new data
types, multiple heterogeneous data resources etc.
► OLTP
► OLAP
► DSS
► DWH
Question No: 13 ( M a r k s: 1 ) http://vuzs.net
________ is the technique in which existing heterogeneous segments are reshuffled, relocated into homogeneous segments.
► Clustering
► Aggregation
► Segmentation
► Partitioning
Question No: 14 ( M a r k s: 1 ) http://vuzs.net
To
measure or quantify the similarity or dissimilarity, different
techniques are available. Which of the following option represent the
name of available techniques?
► Pearson correlation is the only technique
► Euclidean distance is the only technique
► Both Pearson correlation and Euclidean distance
► None of these
Question No: 15 ( M a r k s: 1 ) http://vuzs.net
For a given data set, to get a global view in un-supervised learning we use
► One-way Clustering (P# 271)
► Bi-clustering
► Pearson correlation
► Euclidean distance
Question No: 16 ( M a r k s: 1 ) http://vuzs.net
In DWH project, it is assured that ___________ environment is similar to the production environment
► Designing
► Development
► Analysis
► Implementation
Question No: 17 ( M a r k s: 1 ) http://vuzs.net
For a DWH project, the key requirement are ________ and product experience.
► Tools
► Industry (P# 320)
► Software
► None of these
Question No: 18 ( M a r k s: 1 ) http://vuzs.net
Pipeline parallelism focuses on increasing throughput of task execution, NOT on __________ sub-task execution time.
► Increasing
► Decreasing (P# 215)
► Maintaining
► None of these
Question No: 19 ( M a r k s: 1 ) http://vuzs.net
Many data warehouse project teams waste enormous amounts of time searching in vain for a ___________________.
► Silver Bullet
► Golden Bullet
► Suitable Hardware
► Compatible Product
Question No: 20 ( M a r k s: 1 ) http://vuzs.net
Focusing on data warehouse delivery only often end up _________.
► Rebuilding
► Success
► Good Stable Product
► None of these
Question No: 21 ( M a r k s: 1 ) http://vuzs.net
Pakistan is one of the five major ________ countries in the world.
► Cotton-growing
► Rice-growing
► Weapon Producing
Question No: 22 ( M a r k s: 1 ) http://vuzs.net
_____________
is a process which involves gathering of information about column
through execution of certain queries with intention to identify
erroneous records.
► Data profiling (P# 439)
► Data Anomaly Detection
► Record Duplicate Detection
► None of these
Question No: 23 ( M a r k s: 1 ) http://vuzs.net
Relational
databases allow you to navigate the data in ____________ that is
appropriate using the primary, foreign key structure within the data
model.
► Only One Direction
► Any Direction
► Two Direction
► None of these
Question No: 24 ( M a r k s: 1 ) http://vuzs.net
DSS queries do not involve a primary key
► True
► False
Question No: 25 ( M a r k s: 1 ) http://vuzs.net
__________________
contributes to an under-utilization of valuable and expensive
historical data, and inevitably results in a limited capability to
provide decision support and analysis.
► The lack of data integration and standardization (P# 330)
► Missing Data
► Data Stored in Heterogeneous Sources
Question No: 26 ( M a r k s: 1 ) http://vuzs.net
DTS allows us to connect through any data source or destination that is supported by ____________
► OLE DB
► OLAP
► OLTP
► Data Warehouse
Question No: 27 ( M a r k s: 1 ) http://vuzs.net
Data
Transformation Services (DTS) provide a set of _____ that lets you
extract, transform, and consolidate data from disparate sources into
single or multiple destinations supported by DTS connectivity.
► Tools
► Documentations
► Guidelines
Question No: 28 ( M a r k s: 1 ) http://vuzs.net
Execution
can be completed successfully or it may be stopped due to some error.
In case of successful completion of execution all the transactions will
be ___________
► Committed to the database
► Rolled back
Question No: 29 ( M a r k s: 1 ) http://vuzs.net
If
some error occurs, execution will be terminated abnormally and all
transactions will be rolled back. In this case when we will access the
database we will find it in the state that was before the ____________.
► Execution of package
► Creation of package
► Connection of package
Question No: 30 ( M a r k s: 1 ) http://vuzs.net
To judge effectiveness we perform data profiling twice.
► One before Extraction and the other after Extraction
► One before Transformation and the other after Transformation
► One before Loading and the other after Loading
Question No: 31 ( M a r k s: 2 )
What are the two extremes for technical architecture design? Which one is better?
Theoretically there can be two extremes i.e. free space and free performance. If storage is
not an issue, then just pre-compute every cube at every unique combination of dimensions
at every level as it does not cost anything. This will result in maximum query
performance. But in reality, this implies huge cost in disk space and the time for
constructing the pre-aggregates. In the other case where performance is free i.e. infinitely
fast machines and infinite number of them, then there is not need to build any summaries.
Meaning zero cube space and zero pre-calculations, and in reality this would result in
minimum performance boost, in the presence of infinite performance.
Question No: 32 ( M a r k s: 2 )
What is value validation process?
Value validation is the process of ensuring that each value that is sent to the data
warehouse is accurate.
Question No: 33 ( M a r k s: 2 )
What is the difference between training data and test data?
Question No: 34 ( M a r k s: 2 )
Do
you think it will create the problem of non-standardized attributes, if
one source uses 0/1 and second source uses 1/0 to store male/female
attribute respectively? Give a reason to support your answer.
Question No: 35 ( M a r k s: 3 )
Why
building a data warehouse is a challenging activity? What are the three
broad categories of data warehouse development methods?
-
Waterfall model
-
RAD model
-
Spiral Model
Question No: 36 ( M a r k s: 3 )
What are three fundamental reasons for warehousing Web data?
1. Web data is unstructured and dynamic, Keyword search is insufficient.
2. Web log contain wealth of information as it is a key touch point.
3. Shift from distribution platform to a general communication platform.
Question No: 37 ( M a r k s: 3 )
What types of operations are provided by MS DTS?
-
Providing connectivity to different databases
-
Building query graphically
-
Extraction data from disparate databases
-
Transforming data
-
Copying database objects
-
Providing support of different scripting languages (by default VB-script and Java –
Question No: 38 ( M a r k s: 3 )
What problems may be faced during Change Data Capture (CDC) while reading a log/journal tape?
Problems with reading a log/journal tape are many:
-
Contains lot of extraneous data
-
Format is often arcane
-
Often contains addresses instead of data values and keys
-
Sequencing of data in the log tape often has deep and complex
-
implications
-
Log tape varies widely from one DBMS to another.
Question No: 39 ( M a r k s: 5 )
What are seven steps for extracting data using the SQL server DTS wizard?
SQL Server Data Transformation Services (DTS) is a set of graphical
tools
and programmable objects that allow you extract, transform, and
consolidate data from disparate sources into single or multiple
destinations. SQL Server Enterprise .Manager provides an easy access to
the tools of DTS.
Question No: 40 ( M a r k s: 5 )
Explain Analytic Applications Development Phase of Analytic Applications Track of Kimball’s Model?
Ans:
The DWH development lifecycle (Kimball’s Approach)
has three parallel tracks emanating from requirements definition.
These are
-
technology track,
-
data track and
-
Analytic applications track.
Analytic Applications Track:
Analytic applications also serve to encapsulate the analytic expertise of
the organization, providing a jump-start for the less analytically inclined.
It consists of two phases.
-
Analytic applications specification
-
Analytic applications development
Analytic applications specification:
The main features of Analytic applications specification are:
-
Starter set of 10-15 applications.
-
Prioritize and narrow to critical capabilities.
-
Single template use to get 15 applications.
-
Set standards: Menu, O/P, look feel.
-
From standard: Template, layout, I/P variables, calculations.
-
Common understanding between business & IT users.
Following
the business requirements definition, we need to review the findings
and collected sample reports to identify a starter set of approximately
10 to 15 analytic applications. We want to narrow our initial focus to
the most critical capabilities so that we can manage expectations and
ensure on-time delivery. Business community input will be critical to
this prioritization process. While 15 applications may not sound like
much,
Before designing the initial applications, it's important to establish standards for the applications, such as
-
common pull-down menus and
-
Consistent output look and feel.
Using the standards, we specify each application
-
template,
-
capturing sufficient Information about the layout,
-
input variables,
-
calculations, and
-
breaks
so that both the application developer and business representatives share a common understanding.
During
the application specification activity, we also must give
consideration to the organization of the applications. We need to
identify structured navigational paths to access the applications,
reflecting the way users think about their business. Leveraging the Web
and customizable information portals are the dominant strategies for
disseminating application access.
Analytic applications development:
The main features of Analytic applications development consisits of:
-
Standards: naming, coding, libraries etc.
-
Coding begins AFTER DB design complete, data access tools installed,
subset of historical data loaded.
-
Tools: Product specific high performance tricks, invest in tool-specific
education.
-
Benefits: Quality problems will be found with tool usage => staging.
-
Actual performance and time gauged.
When we do work into the development phase for the analytic applications, we again need to focus on standards. Standards for
-
naming conventions,
-
calculations,
-
libraries, and
-
coding
should be established to minimize future rework. The application development
activity
can begin once the database design is complete, the data access tools
and metadata are installed, and a subset of historical data has been
loaded. The application template specifications should be revisited to
account for the inevitable changes to the data model since the
specifications were completed.
We should take approperiate-specific education or supplemental resources
for the development team.
While
the applications are being developed, several ancillary benefits
result. Application developers, should have a robust data access tool,
quickly will find needling problems in the data haystack despite the
quality assurance performed by the staging application. we need to allow
time in the schedule to
address any flaws identified by the analytic applications.
After
realistically test query response times developers now reviewing
performance-tuning strategies. The application development
quality-assurance activities cannot be completed until the data is
stabilized. We need to make sure that there is adequate time in the
schedule beyond the final data staging cutoff to allow for an orderly
wrap-up of the application development tasks. 
