Parallel Database Systems Engineering

Date: Sunday, February 26, 1995

Speaker : Dr. Kam-Fai Wong

Department of Systems Engineering

Chinese University

Shatin, N.Y.

Hong Kong.

Email - kfwong@se.cuhk.hk

Tel: +852 6098332

Fax: +852 6035505

Abstract:

=========

Today, very large databases may easily involve over tera-

bytes of data. This trend shows no sign of diminishing.

Albeit the advancements in processor technology, handling

such large volume of information is becoming increasingly

difficult for conventional database management systems which

run on sequential computers. To overcome this predicament,

a number of research projects are investigating the use of

parallel computers. The inherent parallelisms behind its

data model (e.g. relational) render database suitable for

parallel implementation. In this tutorial, the concept of

parallel database systems (PDS) which is based on the

extended dataflow computation model will be presented. In

addition, few engineering issues regarding to the implemen-

tation of the model will be reviewed.

Target Audience:

================

* Database developers who are interested in parallel

implementations

* Parallel software developers who are planning to

develop a database system

* First year Postgraduate students in database or paral-

lel computing

- 2 -

Course Structure (provisional):

===============================

Section I: Introduction

I.1 Why Parallel Database Systems (PDS)?

I.2 Overview of the existing parallel machine architectures

suitable for PDS implementation

Section II: PDS Computation Model

Aim(s): introduce the extended dataflow paradigm which is

the computation model of PDS and identify the parallelism

therein.

II.1 Extended Dataflow Paradigm

II.2 Forms of Parallelisms

Section III: Engineering Model

Aim(s): Look at various implementation issues in extended

dataflow graphs.

III.1 Data Placement

III.2 Control Mechanism in execution of an Extended Dataflow

Graph (EDG)

III.3 Self Scheduling in EDG Execution

III.4 Localisation of EDG Operations

III.5 Forms of Pipelines between EDG Operators

III.6 Queuing and De-Queuing

III.7 How to Schedule Work

Section IV: System Architecture

Aim(s): introduce a classical PDS system architecture and

review different implementation techniques

IV.1 A PDS Environment

IV.2 Route for Query Compilation

IV.3 Dynamic versus Static Program Loading and Execution

IV.4 Parallel Query Optimization

- 3 -

Section V: Transaction Model

Aim(s): Because of parallel execution, PDS must maintain the

ACID properties. What are these properties and how to

implement them?

V.1 Definition of the Transaction Model

V.2 How to achieve Atomicity?

V.3 How to achieve Consistency?

V.2 How to achieve Isolation?

V.2 How to achieve Durability?

Section VI: Existing PDS Systems

Aim(s): If time allows, some prominent PDS will be reviewed.

This will include: EDS (European Declarative System), Gamma,

Bubba and XRPS.

Biography

=========

Kam-Fai Wong obtained his PhD from the University of Edin-

burgh, Scotland, in 1987, in the area of computer architec-

tures. After his PhD, he has performed research in Heriot-

Watt University (Edinburgh, Scotland), UniSys (Livingston,

Scotland) and ECRC (Munich, Germany). At present he is a

Project Coordinator at the Chinese University of Hong Kong,

in charge of the IPOC (Intelligent Processing Of Chinese)

project. His research interests are parallel database and

information systems. He has published over 20 technical

papers in these areas in various international journals,

conferences and books.

During his 7 years postdoctoral research period, he has

given many seminars. In 1993/94, he is one of the ACM lec-

turers worldwide. He is a member of IEEE-CS, ACM and

IEE(UK) and have served as the AI/DB track chair in 1994 ACM

Symposium on Applied Computing, the Asian Coordinator of the

1994 Parallel and Distributed Information Systems and PC

members of TOOLS94, GWIC94, PARLE94, VLDB94, SPDP94, ICDCS95

and DASFAA95.