Deck 12: Web Technology and DBMSs, Semistructured Data and XML and Data Warehousing Concepts

Question 1

The World Wide Web is a distributed information system based on hypertext. Discuss how the two-tier client-server architecture may not be entirely suitable for this environment and propose an alternative architecture.

Answer 1

Discussion of traditional Two-Tier Client-Server Architecture versus Three-Tier Architecture - see Section 3.1.

Question 2

Discuss what you consider to be the four most important advantages and the four most important disadvantages of the World Wide Web as a distributed information system. Justify your selection in each case.

Answer 2

See Section 29.2.7, but looking for justification for advantages and disadvantages selected, not just regurgitation.

Question 3

Despite the excitement surrounding XML, it is important to note that most operational business data, even for new Web-based applications, continues to be stored in relational DBMSs. This is unlikely to change in the foreseeable future because of their reliability, scalability, tools, and performance. Consequently, if XML is to fulfil its potential, some mechanism is required to publish relational data in the form of XML documents. The SQL:2003 standard has defined extensions to SQL to enable the publication of XML, commonly referred to as SQL/XML. Discuss in detail these extensions.

Answer 3

SQL/XML contains:
• a new native XML data type, XML, which allows XML documents to be treated as relational values in columns of tables, attributes in user-defined types, variables, and parameters to functions;
• a set of operators for the type:
▪ XMLELEMENT, to generate an XML value with a single element as a child of its root item. The element can have zero or more attributes specified using an XMLATTRIBUTES subclause.
▪ XMLFOREST, to generate an XML value with a list of elements as children of a root item.
▪ XMLCONCAT, to concatenate a list of XML values.
▪ XMLPARSE, to perform a non-validating parse of a character string to produce an XML value.
▪ XMLROOT, to create an XML value by modifying the properties of the root item of another XML value.
▪ XMLCOMMENT, to generate an XML comment.
▪ XMLPI, to generate an XML processing instruction.
▪ XMLSERIALIZE, to generate a character or binary string from an XML value;
▪ XMLAGG, an aggregate function, to generate a forest of elements from a collection of elements.
• an implicit set of mappings from relational data to XML. The mapping may take as its source an individual table, all the tables in a particular schema, or all the tables in a given catalog. The standard does not specify a syntax for the mapping; instead it is provided for use by applications and as a reference for other standards. The mapping produces two XML documents: one that contains the mapped table data and the other that contains an XML Schema describing the first document.

Question 4

Provide XQuery expressions for the following queries based on the sample XML file (books.xml) below.

(a) List the title of the first book.
(b) List the titles of all the books along with a count of the number of books.
(c) List the title and price of each book published in the year 2003.
(d) List the title and numbers of authors of each book.
(e) List the titles of books whose price is less than $60.
(f) List the titles of books in alphabetical order.
(g) List the authors, sorted in reverse order of surname, then first name.
(h) List the authors as a single string, sorted in reverse order of surname, then first name.
(I) List the titles of books that have at least one author called Thomas Connolly.
(j) List the titles of books that have every author called Thomas Connolly.
(k) List books by author.
(l) Test whether the most expensive book is also the book with the largest number of authors/editors.
(m) List the books where Begg is an author but is not listed as the first author.
(n) List the titles of books that are more expensive than the average book price.

Question 5

State the underlying type of the following expressions:
(a) "hello"
(b) 36
(c) 36.00
(d) 3.6e1
(e) 36 + 69
(f) 36.00 + 69.00
(g) 3.6e1 + 6.9e1
(h) 36 + 69.00
(i) 3.6e1 + 69
(j) 36.00 + 6.9e1
(k) "abc" + "def"
(l) "36" + 69.00
(m) "abc" < "def"
(n) 1 < 2
(o) 1 < 2.0
(p) "abc" < 2

Question 6

State the underlying type of the following expressions:
(a) let $a := 1 + 2 return $a + $a
(b) let $a := 1e0 + 1.0 return $a + $a
(c) let $a as xs:integer := 1 + 2 return $a + $a
(d) let $a as xs:decimal := 1 + 2 return $a + $a
(e) let $a as xs:double := 1 + 2 return $a + $a
(f) let $a := xs:double(1 + 2) return $a + $a
(g) let $a as xs:double := xs:double (1 + 2) return $a + $a
(h) let $a as xs:decimal := 1, $b as xs:integer := $a + $a return $b + $b
(i) if ($a < $b) then 3 else 4 (assume $a and $b are xs:integer)
(j) if ($a < $b) then "abc" else 4.00 (assume $a and $b are xs:integer)
(k) if ($a < $b) then 3.6e1 else 4 (assume $a and $b are xs:integer)
(l) (if ($a < $b) then 3.6e1 else 4) + 0.5 (assume $a and $b are xs:integer)

Question 7

'Data warehouse and the underlying support for informational processing will emerge as a growing trend in the 1990s. With the advent of the data warehouse, some basic ideas about data management will change' (Inmon, W.H., 1993). Briefly discuss why the emergence of the data warehouse phenomenon is causing such interest in the business world.

Question 8

'A data warehouse is a subject-oriented, integrated, time-variant and non-volatile collection of data in support of management's decision-making process' (W.H. Inmon, 1993). Describe the major characteristics of the data held in a data warehouse.

Question 9

Discuss the relationship between online transaction processing (OLTP) and data warehousing and identify the major differences between these systems.

Question 10

Describe the architecture and major components of a data warehouse.

Question 11

Discuss the current issues associated with the development and management of a data warehouse.

Question 12

Discuss the major problems associated with the design, development, and management of a data warehouse.

Question 13

Describe how data marts differ from a data warehouse and identify the major issues associated with the development and management of data marts.

Question 14

Explain why businesses have shown a growing interest in data warehousing in recent years.

Question 15

Discuss the reasons why an organisation may have one or more Online Transaction Processing (OLTP) system but only a single data warehouse.

Question 16

"A subject-oriented, integrated, time-variant, and non-volatile collection of data in support of management's decision-making process." (Inmon, 1993). Discuss what this statement is saying about the data in a data warehouse and contrast the purpose of such systems with OLTP systems.

Question 17

Explain why businesses have shown a growing interest in technologies that support their decision makers.

Question 18

The main characteristics for describing Online Transaction Processing (OLTP) systems and data warehousing systems are listed below.
- Main purpose
- Data age
- Data latency
- Data granularity
- Data processing
- Reporting
- Users