Rainbow: Bridging XML and Relational Databases Design, Implementation, and Evaluation

1
Rainbow Bridging XML and Relational Databases
Design, Implementation, and Evaluation

• MQP Advisor
• Prof. Elke A. Rundensteiner
• Sponsor
• Verizon Laboratories Incorporated

MQP Project Members Tien Vu, Mirek Cymer, John
Lee
2
HTML vs. XML

• Microsoft, IBM, Informix, Oracle, Sun, ...

3
XML Data Management by RDBMS

• Advantages
• Efficient query and analysis tools.
• Matured database tools available.
• Easy integration with existing business
  databases.
• Issues
• Map between XML and Relational Model.
• Update Propagation.
• Query Translation and Optimization.

4
Motivation for Mapping
car
ltEMPTYgt
Alternate Mapping
car
make
make
model
year
Ford
model
Ford
Mustang
2001
Mustang
year
2001

• Query Performance vary with respect to how data
  is mapped.
• Flexible mapping fixed translation and
  restructure

5
Rainbow Architecture
Legend
XMLQuery
XML
User
RDBMS
XML Query Engine
XML Data
Subsystem
Restructuring Subsystem
DTDM Manager
XML Manager
DTD
XML
6
Goals of our MPQ

• What
• Implement and evaluate restructuring subsystems
  within the large-scale Rainbow system.
• How
• Learn about the database technologies and web
  tools.
• Translate research ideas to software system
  design.
• Practice software engineering techniques
• UML, engineer and reuse code.
• Design an experimental test plan and test bed.
• Conduct performance study and analysis.

7
Restructuring Subsystem
XMLQuery
XML
User
Legend
XML Query Engine
XML Model
Mapping
Query Storage
Subsystem
Restructuring
Restructure Operator Library
Restructurer
Relational Model
DTDM Manager
XML Manager
Internal Process
DTD
XML
8
Restructuring Operators

• 11 Restructuring Operators
• Rename Item/Attribute
• Switch Nesting
• Pushup/Pushdown Attribute
• Pushup/Pushdown Nesting
• Split/Merge Nesting
• Reference/Dereference

9
Mapping Sequence of Restructure Operators
invoice
summary
ltemptygt
account_num
bill_period
account_num
bill_period
value
value

• Mapping is modeled as a sequence of reversable
  restructuring operators, Operator Name
  Parameters.
• For Example

pushUpAttribute(account_number, value,
invoice, account_number) pushUpAttribute(bil
l_period, value, invoice, bill_peroid) ren
ameItem(invoice, summary)
10
SQLs for Push-Up Attributes
A
A
Push-up
B
a
B
b

• CREATE VIEW new.A (ltall-columnsgt, a) AS
• SELECT A.ltall_columnsgt, B.b
• FROM old.A, old.B
• WHERE B.pid A.iid
• CREATE VIEW new.B (ltall-columns-but-bgt) AS
• SELECT B.ltall-columns-but-bgt
• FROM old.B

11
Example SQLs

• Inline make.value into car as Attribute make.
• Mapping
• pushUpAttribute(account_number, value,
  invoice, account_number)
• SQL statements
• CREATE VIEW new.invoice (iid, pid,
  account_number) AS
• SELECT invoice.iid, invoice.pid,
  account_number.value
• FROM old.invoice, old.account_number
• WHERE account_number.pid invoice.iid
• CREATE VIEW new.account_number (iid, pid) AS
• SELECT account_number.iid, account_number.pid
• FROM old.account_number

12
Rainbow Implementation

• Development Tools
• Java Visual Café2, Javadocs, JAVA2
• Oracle 8i, XML 4J, JDBC1.2, SQL Queries
• Code Facts
• 44 total system classes
• 17 classes of Rainbow
• 27 classes reused
• ? lines of system code
• ? lines of Rainbow code
• ? lines of code reused

13
Screen Shot
14
Screen Shot
15
Rainbow Test Experimental Evaluation

• Experimental Setup
• Oracle 8i
• Windows NT
• Data
• Created a DTD
• Randomly generated XML
• Hand translated queries
• Factors
• Type of query
• Number of operations

16
Query Performance Evaluation
17
Rainbow Conclusions

• Technical accomplishments
• Functional prototype system
• Feasibility of Rainbow concepts
• Automated test bed designed
• Performance evaluations show that
• (Ideal) Moving up data on the embedded-relational-
  level yields better query performance for Join
  queries.
• Knowledge gained
• OO, Java, JDBC, SQL, RDBMS, XML, DTD
• Teamwork S/W Engineering Software Reuse
• Logistics of setting up an experiment
• Future work
• Experiment test plans and test beds to realize
  the full potential of the restructuring component.

18
Rainbow XML and Relational Database Design,
Implementation, and Evaluation

• Project Members
• Tien Vu, Mirek Cymer, John Lee
• Advisor
• Elke A. Rundensteiner
• Ph. D Student
• Xin Zhang
• Sponsor By
• Verizon Laboratories Incorporated
• Visit Rainbow at http//davis.wpi.edu/dsrg/TJM/

19
Recycled!!!
20
XML The Future of the Web

• Benefits
• Efficient query and analysis tools.
• Matured Data Warehousing support.
• Easy Integration with existing business database.
• Applications
• E-commerce
• Web-based industries

• ltinvoicegt
• ltaccount_numbergt555 777-3158 573 234
  lt/account_numbergt
• ltbill_periodgtJun 9 - Jul 8, 2000lt/bill_periodgt
• ltcarriergtSprintlt/carriergt
• ltitemized_call no1 dateJUN 10
  number_called973 555-8888 time1017pm
  rateNIGHT min1 amount0.05 /gt
• ltitemized_call no2 dateJUN 13
  number_called973 650-2222 time1019pm
  rateNIGHT min1 amount0.05 /gt
• ltitemized_call no3 dateJUN 15
  number_called206 365-9999 time1025pm
  rateNIGHT min3 amount0.15 /gt
• lttotalgt0.25lt/totalgt
• lt/invoicegt

21
XML and Relational Database

• Problem
• Many Application usually change its data very
  frequently.
• e.g., flight reservation, online billing,
  inventory.
• Current Solution
• Reloading the complete XML document when changed
  which is very expensive.
• Rainbow Solution
• Incrementally propagate XML Document Updates to
  Stored XML Data.
• Goal XML Repository Implemented using RDBMS
• Approach Flexible Mapping
• Features
• DTD Metadata Management in RDB
• Automatic Schema Creation
• Incremental Update Propagation
• XML Query Optimization

22
Rainbow Analysis
23
Rainbow Analysis Cont..
24
HTML vs. XML

• HTML
• lth1gtCarlt/h1gt
• lth2gtMakelt/h2gt
• ltpgtFord Mustang
• lth2gtSeatslt/h2gt
• ltpgt5
• lth2gtTop Speedlt/h2gt
• ltpgt70 m.p.h

• XML
• lth1gtCarlt/h1gt
• ltmakegtFord Mustanglt/makegt
• ltseatsgt5ltseatsgt
• ltspeed unitsmphgt70lt/speedgt