Creates and populates the local db2 database

Assignment Help Database Management System

Reference no: EM131134033

IS 676: Information Integration

Assignment 3: Integration layer on top of two databases

DESCRIPTION

A major step in information integration is to provide access and retrieve information from different databases. You have already created an Inventory database in a previous homework. From now on we will use that database but we will refer to it in this assignment with a different name - let us call it local DB1 and it belongs to a fictitious company you own (MyComp).

MyComp has been doing really well for the past few years and it has merged with another company (BuddyComp) which already has an inventory database - let us call it local DB2. The local DB2 database is actually a set of tables and records that is provided to you as a script with SQL commands. You need to run this script in your Oracle account to create and populate the tables that comprise the local DB2 database. In reality you do not have two separate databases in Oracle; instead, you have several tables in your Oracle account:

Tables that make up your Inventory database (from HW2) which we now call local DB1
Tables that are generated by the script and make up local DB2

The purpose of this homework is to create an integration/metadata/ontology layer on top of these two inventory databases, local DB1 and local DB2. Think of the integration layer as the metadata layer which contains information describing the two databases. This homework is about creating this metadata layer.

Creation of a metadata layeri-

You need to create ametadata layer to be used for integration of information from the participating databases. This metadata layer is a table (or a set of couple of tables) which contain information about the local database schemas (local DB1 and local DB2, which both reside in your Oracle account. The information in the integration layer contains the following:

1. Canonical representation. This is the name you use to refer to an entity (table name, field name) at the metadata layer. For example, you may use the name"Client" in local DB1, and "Patron" in local DB2. However, these are "local names"representing customers at each local DB. At the metadata layer you may use the "Customer" name to identify either Clients or Patrons (this is just an example; there could be different field names in the actual databases). The canonical representation of an entity is the "global" name of that entity at the metadata layer. It can be used to describe the two corresponding names of the same entity, one in local DB1, and the other in local DB2. Customer (canonical name); Client (local DB1); Patron (local DB2)

2. Data Types and other semantic differences. This is a description of the local data types used for each column name in each DB. For example, varchar2(20), number, etc. You may also store function that converts data from the canonical representation to local databases.

3. Additional fields. If you want you may use additional information about the correspondences or translations between canonical to local.

You may use any structure of table(s) that you think it is appropriate to identify information that is stored as data in the two participating databases (local DB 1 and local DB 2). This is a very important task since it will provide the basis for the integration. I would suggest that for each concept (entity/field/etc.) that is present in each local db, you need to use three representations:

1. A canonical representation (global level) for the concept. This is a representation that you use to identify the concept globally (see first column in the example table below).

2. A local representation for local DB1: it is the name of the column in local DB1 that represents that same concept. Also the data type of that column in local DB1 (see columns 2 and 3 in example table below)

3. A local representation for local DB2: it is the name of the column in local DB2 that represents that same concept. Also the data type of that column in local DB1 (see columns 4 and 5 in example table below)

See an example table below for a very simple representation of two concepts: a customer (represented in both DBs) and a product ID(also in both DBs). Note that all data entered in that table are strings (varchar2):

Canonical Representation	Column name in local DB1	Data Type in local DB1	Column name in local DB2	Data Type in local DB2
'Customer'	'Client'	'Varchar2(20)'	'Patron'	'Char(50)'
'Product_ID'	'CD_ID'	'Varchar2(10)'	'CD'	'number'
...	...	...	...	...

You may need to add more fields and more tables to be able to capture semantic and syntactic differences that are present in the two local databases. In essence, you need to create another set of tables (one or more) that will act as the metadata/integration layer on top of DB1, and DB2. You don't create views on top of the existing tables of local DB1 and local DB2. You need to create actual tables with metadata about DB1 tables and fields, and DB2 tables and fields. All the above tables are in your own account (basically, they are just tables in your account).

How can one create a metadata/integration layer?

Look at the schema of local DB1.

Look at the schema of local DB2.

Identify semantically similar fields and for each field create a column in the metadata table:

Column 1: Provide a name (canonical representation) of that field.
Column 2: Identify the corresponding name (local DB1 name) of the same field in DB1
Column 3: Identify the data type of that field in DB1
Column 4: Identify a function to map the canonical name to the DB2 name (if applicable)
Column 5: Identify the corresponding name (local DB2 name) of the same field in DB2
Column 6: Identify the data type of that field in DB2
Column 7: Identify a function to map the canonical name to the DB2 name (if applicable)

The results from bullet list above are going to be inserted as a single record in the table which represents the metadata layer. If you need additional fields, (e.g., how to convert from local to canonical - such as CONCAT (f1, f2), back and forth) feel free to add additional column(s) in your integration table and populate them accordingly.

Fields that are not present in both databases have NULL values in some of the columns in the metadata table.
For aggregate fields type the function that puts them together (e.g. CONCAT, or SUM, etc.)
Add table names also in the metadatatable (just as you have added fields). The data type for tables should be the string 'TABLE'

Let me reiterate that this assignment is about metadata. The final output of this assignment is to make sure that you have enough information in your metadata (integration) layer.

DELIVERABLES

Create a file in MS Word or pdf and upload it on Blackboard. This file should contain the following:

1. Populating Local DB2. Run the script that creates and populates the local DB2 database. It has been posted on Bb, under Homework Assignments. Download it, study the relational schema and run it on your Oracle account.

Preservation of local schemas. You are NOT allowed to modify the schema of the two existing databases (local DB1 and local DB2). However, you may insert more data records in any existing table.

a. Insert records in local DB1: 2 CDs of your choice and 2 books of your choice.

b. Insert the same records in the schema of local DB2. Submit the INSERT commands that you have used for a. and b., and also show a screen shot with the new records.

2. Creation of the integration (metadata) layer

c. E-R diagram of your integration layer (this is a very very simple one)

d. Screen-shots with queries showing "select * from ..." for each table you created in the metadata (integration) layer (not the local DB2).

Detailed Question: Please use homework 2 for assignment 3. all queries are to be done in SQL and provide screenshots in Oracle.

Homework 2 - https://www.dropbox.com/s/pewqshz7oylv26w/Assignment.rar?dl=0.

Reference no: EM131134033

Questions Cloud

In what way did japans relationship with larger world change : In what ways did Japan's relationship with the larger world change during its modernization process? China and the Ottoman Empire fell prey to European imperialism in the 19th century. How so?

What are differences between an inner join and an outer join : What are the differences between an inner join and an outer join? Describe in your own words. When would you use one type of join versus the other? Provide at least two examples of each type of join.

Ethics and professional conduct in business : The Southwest Division of Texcaliber Inc. uses absorption costing for profit reporting. The general manager of the Southwest Division is concerned about meeting the income objectives of the division. At the beginning of the reporting period, the d..

Stimulates demand for a product category : Advertising that stimulates demand for a product category by informing potential customers about the product’s features, uses, and benefits is called ________________.

Creates and populates the local db2 database : Populating Local DB2. Run the script that creates and populates the local DB2 database. It has been posted on Bb, under Homework Assignments. Download it, study the relational schema and run it on your Oracle account

Experienced a hazardous materials spill at one of its plants : Several months ago, Ayers Industries Inc. experienced a hazardous materials spill at one of its plants. As a result, the Environmental Protection Agency (EPA) fined the company

Describe the seven step sdlc : The paper will be five pages: Describe the 7 Step SDLC and Describe a different SDLC Model (4 step or 12 step).

What were permanent long-term us objectives in latin america : Discuss what were the permanent long-term U.S. Objectives in Latin America, and how successive U.S. governments sought to achieve them before 1898; between 1898-1933; after WWII; during the Cold War; and since 1981?

Products equals the amount spent producing the products : The money a company brings in from selling products equals the amount spent producing the products. the total fixed costs are exactly equal to the total variable costs. profits are exactly equal to the difference between revenue and total variable co..

User Account

All Pages