Maximizing Performance and Scalability with IBM WebSphere

W

waterfall model, 28 “29

Web Archive file

Web module as WebSphere representation of, 63

Web container

as component of a J2EE environment, 59 “60

as component of a WebSphere queue, 279 “280

effect on performance, 8

vs. EJB container performance tuning, 374 “375

example implementation within WebSphere, 60

example of how it operates, 373

failover process, 330 “333

function of, 373 “374

key items that should be tuned , 386 “387

overview of, 311

queue modeling item to monitor, 284

steps for accessing for any application server, 384 “385

steps for accessing in WebSphere 5, 385 “386

tuning in WebSphere 5, 385 “394

tuning the number of threads processed by, 387 “389

in a WebSphere high availability environment, 304

WebSphere queue settings, 285

Web container failover, 329 “333

failover process, 330 “333

primary causes for failure in a typical environment, 329

Web container failover and high availability, 311 “333

Web container overview, 311

Web Container Service dialog box

in WebSphere 4, 385

Web container services

example of in WebSphere 5, 385 “386

Web module

function of, 63

Web module level

configuring session management in, 327

Web server

as component of a WebSphere queue, 279 “280

implementation, 308 “310

overview, 308

Web server failover and high availability

model for, 309

in WebSphere, 307 “310

Web servers (thin vs. thick models)

considerations for opting for one over the other, 166 “173

effect on performance, 8

Web servers

example of geographically distributed, 225

in high-level 2- tier WebSphere topology, 175

Web servers (thin vs. thick models)

in your WebSphere topological architecture, 165 “173

Web server tier/HTTP servers

in a WebSphere high availability environment, 303 “304

Web Service engine

function of in WebSphere 5, 74

Web Services

based on a HTTP Web Service stack, 465

diagram showing integration with WebSphere, 464

future of in legacy systems integration, 455

performance and scalability tips, 465

using standard SOAP data types, 465 “466

XML validation in, 467

Web Services Description Language (WSDL). See WSDL (Web Services Description Language)

Web Services integration

function of for distributed communications, 463 “468

Web site address

for downloading Jinsight, 516

for Quest Software JProbe profiling tool, 508

Web tier

example of abstraction layer for, 171

performance development considerations for, 397 “400

Web-based User Interface (UI)

effect of images and components on response times, 238 “239

WebSphere

administrative consoles, 53

availability model, 35 “36

base configurations provided by for testing and example purposes, 52

basic J2EE application operating under, 198 “199

building block rules, 155 “157

the business bottom line, 13 “14

causes of downtime, 41 “42

clusterable EJBs available in versions 4 and 5, 341 “342

combined horizontal and vertical scaling with, 153 “154

common components between versions 4 and 5, 51 “64

database performance and optimization, 411 “451

databases supported by, 412 “414

deployment and network architecture, 149 “229

dual Web, single application server, dual JVM topology, 200 “202

EJB and Web container performance, 371 “394

example of dual application server configuration, 313

example of horizontal scaling with, 151

example of session management in a failover scenario, 326

example of session management with, 325

example Web container implementation within, 60

external system availability, 343 “370

failover and high availability considerations, 291 “342

failover techniques with, 301 “307

function of HTTP server with, 53 “54

generic database performance architecture with, 428 “431

high availability and failover fundamentals, 292 “293

horizontal scaling with, 150 “152

infrastructure design, 81 “147

JDBC connection pool tuning, 445 “447

key components that affect performance, 8 “9

MQ Series communications with, 462

performance of Java 2 Enterprise Edition (J2EE) applications, 1 “5

profiling and benchmarking, 503 “517

scalability and availability, 33 “50

session identifiers supported in versions 4 and 5, 326 “329

split Web and EJB container JVM configuration, 199 “200

SQL implementation methods used within, 414 “427

SSL ID-based tracking supported by, 329

types of database usage you may require with, 431

vertical scaling with, 152 “153

virtual host capability of, 52 “53

Web server failover and high availability in, 307 “310

Web Service communications with, 464

WebSphere 4

administration console, 378

administrative repository, 68 “69

administrative repository considerations, 356 “358

cloning and workload management flowchart, 68

cloning in, 66 “68

code example of key CloneID changes, 316

configuring the pool manager connection in, 447

EJB Container Service dialog box in, 379

EJB container tuning in, 378 “380

function of administration server in, 69

function of EJB container failover in, 336 “337

IBM-certified relational databases in, 412

improved concept of server groups in, 65 “66

JDBC connection pool tuning, 447 “448

managed by a Java/Swing-based Graphical User Interface (GUI), 53

parameter to help improve ORB performance, 389

repository database failover, 356 “358

session persistence options, 324

steps for making changes to keep- alive I/O time-out values, 393

steps for making changes to keep-alive requests , 391

steps for setting the thread size value in, 388

Web container performance tuning, 384 “385

Web module introduced in, 63

XMLConfig tool in, 65

WebSphere 4 and 5

component architectures, 51 “79

three forms of session persistence standard with, 363

WebSphere 4-specific component architecture, 64 “69

WebSphere 5

code example of key CloneID changes, 316 “317

ConnectTime setting, 332 “333

ConnectTimeout setting in, 322 “323

contents of node in, 72 “73

EJB container services in, 381

EJB container tuning in, 380 “381

external load balancers in, 320 “322

function of administration service in, 76

function of administrative services in, 76

function of deployment manager in, 73

function of EJB container failover in, 337 “338

function of JCA container in, 75

function of JMS server in, 75 “76

function of names services server in, 75

function of security services server in, 75

function of UDDI Registry in, 74

IBM-certified relational databases in, 413

JDBC connection pool tuning, 449 “451

key components of, 69 “79

managed by a Web-based console, 53

network deployment component architecture, 71

new configuration settings available for, 449 “451

new session management scope supported in, 327

node agent in, 73

RetryInterval setting in, 331 “332

session persistence options, 324 “325

standard deployment component architecture, 70

steps for accessing EJB container settings in, 380 “381

steps for accessing the Web container settings, 385 “386

steps for making changes to keep-alive I/O time-out values, 393 “394

steps for making changes to keep-alive requests, 391 “392

steps for setting the thread size value in, 388

tying ConnectTime setting with other setting in, 332 “333

Web container tuning, 385 “394

weighted load distribution in, 318 “320

WebSphere 5.0.1

changes to weighted load distribution in, 320

WebSphere 5-specific component architecture vs. WebSphere 4-specific component architecture, 69 “79

WebSphere application environments

batch components and external applications within, 469 “472

categories of services provided by, 411 “412

effect of large database workload on your network, 433 “434

example of correctly configured funnel or carrot -shaped model, 29 “30

ROI from reducing processing over-head, 23 “24

running batch scripts and external applications within, 469 “472

using iostat command to pick a hot disk in, 487 “488

WebSphere application performance

high CPU utilization of JVM as cause of poor, 498

monitoring, 497 “500

WebSphere applications

available stress and volume testing tools for, 246

availability as key performance indicator, 36

avoiding use of large session objects in, 397 “398

business tier performance in, 400 “406

database server failover in, 356 “370

defining the methods of monitoring and testing, 240 “243

developing high performance, 395 “410

importance of not using single-threaded models with servlets in, 400

importance of using <jsp: usebean ()> tag carefully in, 400

importance of using constants in, 405

importance of using WebSphere data sources in, 407

importance of writing local method calls, 405 “406

insulating from failover delays, 361 “362

invalidating HTTP sessions when they aren't required, 398 “399

monitors and probes for testing, 244 “245

a practical approach to profiling, 505 “507

purpose of, 34

questioning use of doGet() and doPost() methods in, 399

stress and volume testing, 245 “246

tools for monitoring performance of, 473 “500

top 20 performance development considerations, 396 “397

using container-managed persistence only for high-read database access, 408

using Data Access Objects for general database access in, 409

using init() servlet method in, 399

WebSphere application server level

configuring session management in, 327

WebSphere application servers

common symptoms for poor performance of, 497 “500

developing high-performance applications for, 395 “410

having all CORBA components with a shared, 459 “460

how Web Services may be implemented within, 464 “465

importance of not using all of your memory for scripts, 470

WebSphere application server topologies

database server tier as important component of, 305 “306

WebSphere availability

understanding, 40 “43

WebSphere availability index

measuring for components within your environment, 44 “45

WebSphere bin directory

WebSphere Control Program in, 65

WebSphere business tier. See business tier

WebSphere cluster environment, 78 “79

WebSphere clustering

typical configuration and function of, 298 “299

WebSphere cluster networking

using for a multinode or multiapplication server environment, 141 “142

WebSphere Command Program (WSCP)

listing the running state of each application with, 480

WebSphere connection pool manager

function of, 442 “445

how a pooled connection infrastructure operates, 443

for WebSphere-based applications, 442 “451

WebSphere Control Program (WSCP)

function of in WebSphere 4, 64 “65

WebSphere database. See also database

performance and optimization, 411 “451

tuning and optimization, 427 “442

WebSphere database architecture

database considerations, 209

WebSphere database tuning

for two most common database products, 427 “442

WebSphere EJB container

function of within the application server, 61 “62

WebSphere environment

options for using NFS in, 348

WebSphere environments

ability to scale down infrastructure with an optimized, 13

building block rules for, 154 “164

considerations for scalability and performance, 157 “164

data tier performance in, 406 “409

disk storage recommendations for larger implementations , 133 “134

example of end-to-end response time measurement, 235

HTTP plug-in as key component of, 54 “58

importance of knowing what you are changing, 232 “233

indicative operational costs, initial outlay, 12 “13

key areas susceptible to failures, 40

main categories of networks within, 162 “164

the methodology of optimization in, 232

the need for speed in, 231 “233

network requirements for, 141 “145

networking as a key component in, 140 “145

optimization and tuning checklist for one running under Linux, 255 “260

primary causes for Web container failover, 329 “330

primary reasons for using multiple JVMs in, 183

recommended testing applications for, 243 “244

selecting the right CPU for yours, 116 “118

summarizing storage technologies in, 130 “140

tools for testing, 243 “246

tuning the JVM, 287 “288

using Microsoft Excel for analyzing tuning data, 247

WebSphere high availability environment

EJB container in, 304 “305

example of a complex, 302

levels of, 302 “303

Web container in, 304

WebSphere HTTP plug-in

contents of configuration file, 54 “56

example of topology, 55

five main sections of configuration file, 56 “58

function of, 54 “58

WebSphere implementations

cost comparisons for, 21 “23

cost items to be included in, 20 “21

failover and high availability considerations, 291 “342

key indicators of performance improvements in, 7 “8

overview of key components of, 149 “150

WebSphere Java process

steps for drilling down and viewing, 483 “484

WebSphere JDBC connection pool tuning

primary settings common to WebSphere 4 and 5, 445

WebSphere physical implementation

single Web server, single application server topology, 178 “183

WebSphere platform

comparing versions in, 79

example CORBA implementation within, 457

exploring advanced features, 145 “147

improving performance for greater user capacity, 18 “19

methods for achieving high availability, 293 “307

performance testing, 233 “251

performance, tuning, and optimization, 231 “289

testing response times for an application, 234 “236

topological architecture, 154 “164

WebSphere queue

EJB container queue settings, 286

JDBC connection pool manager queue settings, 286 “287

Web container queue settings, 285

WebSphere queues

Carrot mode showing correct queuing configuration, 280

determining sizes for, 282 “283

essential components for, 279 “281

an example queuing model, 283

HTTP Web server queue settings, 285

as part of performance and capacity management, 279 “287

queue configuration baselining, 283 “284

queue modeling items to monitor, 284

reasons for using them, 281

updating and configuring, 284

WebSphere repository database

impact of on a WebSphere environment, 431

WebSphere Resource Analyzer tool

capturing performance management testing results with, 26 “27

using to monitor and test queues, 283 “284

WebSphere topological architecture

blueprints, 165 “229

and example of a building block for, 156 “157

example of a high-level, single-channel, 2-tier topology, 175

example of a high-level, single-channel topology, 174

major network types within, 163

WebSphere-based applications

factors to consider when implementing a database for, 429 “431

IBM DB2 parameters, 440 “441

optimization and tuning checklists for, 251 “287

products that provide solid, robust NFS solutions for, 349

using baselining and profiling for performance optimization, 242

WebSphere-based environments

factors to consider when implementing a database for, 429 “431

profiling and benchmarking tools for, 507 “517

WebSphere-based J2EE applications

interfacing with Java Message Service (JMS), 455 “456

the need for performance in, 1 “5

WebSphere/J2EE application

example showing integration with a legacy application, 454

WebSphere-to-customer (or Web server) networking

recommendations for, 144 “145

WebSphere-to-database networking

basic recommendations for, 142 “144

weighted load distribution

changes to in WebSphere 5.0.1, 320

in WebSphere 5, 318 “320

weighted round- robin routing algorithm

used within WebSphere 5's EJB WLM routing, 335

Windows 2000 Active Directory (with LDAP extensions)

supported by WebSphere, 344

Windows NT/2000/XP

CPU and system load performance monitoring for, 481

Windows NT/2000/XP/2003

performance tuning checklist for, 252 “255

real and virtual memory usage, 484 “485

steps for modeling the usage of real and virtual memory, 484 “485

Windows Performance tool

example showing network utilization output, 496

measuring CPU and system load with, 481

monitoring network utilization with, 495 “497

for monitoring Oracle and DB2 for performance issues, 442

selecting the monitor agents in, 482 “483

steps for setting up active monitoring on your CPU, 482 “483

Windows server environment

disk storage technologies, 488 “490

Windows server operating system

performance tuning checklist for, 253 “255

Windows-based file systems

system and hardware tuning and optimization for, 275

WinRunner (Mercury Interactive)

profiling and load testing tool, 243

WLM availability

different levels of for configuring EJB containers, 305

workload management. See load balancing

and cloning flowchart in WebSphere 4, 68

of EJB container in WebSphere, 334 “336

and EJB containers, 340 “342

requirements for user sessions to participate in, 329

Workload Management (WLM) pool, 67

workload summary matrix

for generic WebSphere database sizing, 434 “435

WSAD profiler/Jinsight

features of, 517

function of, 516 “517

WSAD Profile view, 516

wsadmin tool

listing the running state of each application with, 480

WSCP. See WebSphere Control Program (WSCP)

WSDL (Web Services Description Language)

caching of to improve performance, 467 “468