ClientServer and Middleware

1
Client/Server and Middleware

• Chapter 8
• MIS 3413 Advanced Data Base Concepts
• Dr. Segall
• Spring 2002

2
Client/Server Environments

• Client/server environments use a local area
  network (LAN) to support a network of personal
  computers, each with its own storage, that are
  also able to share common devices (such as hard
  disks or printers) and software (such as a DBMS)
  attached to the LAN. At least one PC is
  designated as a file server, where the shared
  database is stored.

3
Client/Server Architectures

• The several client/server architectures that
  have evolved can be distinguished by the
  distribution of application logic components
  across clients and servers.
• There are 3 components of application logic
• 1. Presentation Logic
• 2. Processing Logic
• 3. Storage Logic

See Figure 8.1
4
Application Logic Components

• Presentation Logic
• Concerned with input and output (I/O), this
  component is responsible for formatting and
  presenting data to the user (e.g., screen or
  printer) and managing input from the user (e.g.,
  keyboard or mouse)

5
Application Logic Components

• Processing Logic
• Data processing logic
• data validation
• identification of processing errors
• Business rules logic
• those rules not coded at the DBMS level
• Data management logic
• identification of data necessary for processing
  of transactions or queries

6
Application Logic Components

• Storage Logic
• Responsible for data storage and retrieval from
  the physical storage devices associated with the
  application being used. (Activities of a DBMS
  occur in this component.)

7
Client/Server Architectures

• File Server Architecture
• Database Server Architecture
• Three-Tier or N-Tier Architecture

See next slides ?
8
File Server Architecture

• The first of the client/server architectures that
  were developed
• The client (a PC, sometimes called a fat client)
  handles the presentation logic, processing logic,
  and much of the storage logic (that part
  associated with a DBMS)
• One or more file servers (devices that manage
  file operations and are shared by each of the
  client PCs attached to the LAN) act as additional
  hard disks

9
File Servers (contd)

• In a file server environment, each client PC is
  authorized to use the DBMS when a database
  application program runs on that PC. Thus, there
  is one database but many concurrently running
  copies of the DBMS, one on each of the active
  PCs. The primary characteristic of file server
  architecture is that all data manipulation is
  performed at the client PCs, not at the file
  server. The file server acts simply as a shared
  storage device.

10
Limitations of File Servers

• Considerable data movement is generated across
  the network. (E.g., when a table is queried, the
  entire table is transferred from the file server
  to the client PC.)
• Each client workstation must perform all
  processing and devote memory to a full version of
  the DBMS (and thus must be relatively powerful).
• The DBMS copy in each PC must handle shared
  database integrity and security.

11
Database Server Architecture

• An improvement on file server architecture
• Sometimes called two-tier architecture
• The client is responsible for managing
  presentation logic, data processing logic, and
  business rules logic
• The database server is responsible for database
  storage, access, and processing
• See next slide for schematic

12
Database Server Architecture
13
Database Server Terminology

• Back-end Functions - functions performed by the
  central DBMS on the database server
• Front-End Programs - the application programs
  running on the client PCs
• Stored Procedures - modules of code (e.g.,
  proprietary SQL dialects) that imple-ment
  application logic or a business rule that are
  stored on the database server

14
Database Server Advantages

• Only the database server requires processing
  power adequate to handle the database (and it can
  be tuned to optimize database-specific
  processing)
• Only selected data (e.g., the answers to queries)
  are sent across the network, greatly reducing the
  telecommunications traffic
• User authorization and other security issues, and
  data integrity checking and sharing is handled at
  one location, the database server

15
Limitations of Database Servers

• The writing of stored procedures to implement
  application logic, because of the proprietary
  nature of them, may reduce flexibility in
  changing DBMSs.
• Each client must still be loaded with all
  application programs that will be used at that
  location.
• Upgrades to an application will require that each
  client be upgraded separately.

16
Three-Tier Architecture

• Includes another server layer in addition to the
  client and database server layers already
  described
• The additional server in a three-tier
  architec-ture may serve different purposes, but a
  common use is as an application server
• This application server will serve as a central
  location for storage of application programs
• See next slide for schematic

17
Three-Tier Architecture
18
Three-Tier Architecture

• Most application code is stored on the
  application server rather than on client PCs (and
  thus upgrades to applications need only be loaded
  on the server, not each PC)
• Resulting workstations are referred to as thin
  clients PCs configured only for handling user
  interface (e.g., Windows) and minimal application
  processing (and often with limited or no local
  data storage)

19
N-Tier Advantages

• Scalability - greater ability to upgrade a system
  without having to redesign it (e.g., can add
  additional application servers as demands on
  system grows)
• Technological Flexibility - with most application
  logic handled by the application server, the DBMS
  can be changed more easily, it is easier to
  interface with the Web

20
N-Tier Advantages (contd)

• Lower Long-Term Costs - changes to the
  applications in the middle tier can be made
  faster and less expensively. Use of off the
  shelf products is easier, and small chunks of
  code can be written to address needs
• Improved User Satisfaction - multiple interfaces
  on different clients can access the same business
  process

21
N-Tier Challenges

• High Short-Term Costs - implementation costs
  associated with programming necessary to convert
  to this architecture
• Tools and Training - the newness of this
  tech-nology means that tools are not yet fully
  developed and training programs are not yet
  widely available
• Experienced Personnel - few trained people
• Incompatible Standards - standards for some
  components of this technology have not yet
  evolved
• Lack of End-User Tools that work with Middle-Tier
  Services - problems with generic applications
  such as spreadsheet programs (see middleware)

22
Other Client/Server IssuesApplication
Partitioning

• Defined as The process of assigning portions of
  application code to client or server partitions
  after it is written, in order to achieve better
  performance and interoperability (ability of a
  component to function on different platforms).
• Basically refers to the ability to optimize a
  particular installations performance by placing
  code where it will run the fastest (client or
  server)

23
Other Client/Server IssuesParallel Architectures

• Useful in environments characterized by high
  transaction volumes, complex queries, and new
  data types (e.g., video)
• Supported by SQL since individual queries can be
  split into parts which can then be run
  simultaneously on multiple processors
• Also, multiple queries can be run simul-taneously
  on parallel processors

24
Multiprocessor Hardware Architectures

• Symmetric Multiprocessing (SMP)A design
  characterized by all processors sharing a common
  memory and single copy of the operating system.
• All processors share one copy of OS
• Fewer bottlenecks due to multiprocessing
• Improved performance because data stays in memory
  during processing

25
Multiprocessor Hardware Architectures

• Massively Parallel Processors (MPP)A design
  characterized by all processors having their own
  dedicated memory and copy of the operating
  system.
• Eliminates of memory contention
• Easy scalability (small increments possible)

26
Things to Consider when using SMP or MPP

• Data Analysis Requirements
• Large Databases
• Complex Queries
• Large of Users
• Cost Justification
• Investment not cost savings (often complements
  and does not replace)
• Traditional Technologies
• Less Risky

27
Middleware Terminology

• Middleware software which allows an application
  to interoperate with other, otherwise
  incompatible, software without requiring the user
  to understand and code the low-level operations
  necessary to achieve interoperability.
• Synchronous (real-time)
• Asynchronous

28
More Middleware Terminology

• Application Program Interface (API) sets of
  routines that an application program uses to
  direct the performance of procedures by the
  computers operating system
• Open Database Connectivity (ODBC) similar
  definition to API, but refers to Windows-based
  client/server applications used for accessing
  relational database data

29
More Middleware Terminology

• Common Object Request Broker Architecture
  (CORBA) a specification for a standard,
  universal, object-oriented middleware
• Distributed Component Object Model (DCOM) a
  competitor to CORBA developed and supported by
  Microsoft

30
Client/Server Security Issues

• Because networks are susceptible to breaches of
  security (e.g., through electronic eavesdropping
  and unauthorized connections), establishing
  database security is more difficult in a
  distributed environment than it is in a
  centralized environ-ment. Security measures to
  consider include
• System-Level Security - user names passwords
  required to access a multiuser client/server
  system
• Database-Level Security - user names passwords
  required to access a particular database (2nd
  layer)
• Secure Client/Server Communication - use of
  encryp-tion to transform readable data into
  ciphertext

31
Issues to Address Before Implementing
Client/Server

• Perform Accurate Business Problem Analysis(dont
  force a problem to fit a particular technology)
• Perform Detailed Architecture Analysis(make sure
  existing components can work together)
• Avoid Tool-Driven Architectures(again, define
  problem first, then choose tools to use)
• Consider Scalability Needs in Choosing
  Architecture(e.g., two-tier may be easier to
  implement for modest needs)
• Determine Proper Placement of Services(seek best
  balance of fat/thin servers and fat/thin clients)
• Analyze Throughput of Networks(make sure that
  existing networks can handle volume)
• Plan for Hidden Costs(allow generous development
  times anticipate training costs)

32
Benefits of Moving to a Client/Server Architecture

• System functionality can usually be rolled out in
  increments - speeds delivery
• GUI interfaces on most systems are familiar and
  welcome to end-users
• Flexibility and scalability allows for pursuit of
  business process re-engineering (BPR)
• With proper design, can reduce network traffic
  and improve response times
• Facilitates Intranet/Internet applications

33
Client/Server and Middleware

• Chapter 8
• THE END!!!