Maximizing Performance and Scalability with IBM WebSphere
S
sales and order management application
in a single-channel 3- tier topology, 178 “180
SAN (Storage Area Network) technology. See Storage Area Network (SAN) technology
S-ATA (Serial ATA)
function of, 128 “129
vs. Parallel-ATA (P-ATA), 128
scalability
as key measure of high-performance computing, 33
understanding your needs for, 47 “48
what it is, 34 “35
scalability and availability
a primer, 34 “43
ten rules to live by, 49
scalability and performance
considerations for scalability and performance, 157 “164
script factory
considerations for implementing, 470
SCSI disk technologies
common implementations , 121
SCSI-2 disk technology
function of, 121 “122
SecondaryServer directive
using to help support failover configurations, 317
Secure Socket Layer (SSL). See SSL (Secure Socket Layer)
security services server
function of in WebSphere 5, 75
Select type
in Oracle Transparent Application Failover, 368
sendspace attribute
setting for TCP send and receive buffer, 270
Serial ATA (S-ATA). See S-ATA (Serial ATA)
server affinity, 184
using in split-brain WebSphere configuration, 207 “208
server cluster directive
code example of, 314
server group directive
code example of, 314
server groups
vs. application servers, 65
association with WebSphere 4 components , 66
vs. clones , 66 “67
improvement of concept of in WebSphere 4, 65 “66
server sizes and configurations
deciding on an operational preference for, 172
ServerGroup directives
in an HTTP plug-in configuration file, 58
service availability vs. operational availability, 292 “293
Service Level Agreement (SLA), 15
costs associated with for the pilot application, 16 “17
session database
function of, 63
as key part in WebSphere availability strategy, 362 “365
use of for session persistence, 51
session EJBs vs. entity EJBs, 376
session failover and persistence, 323 “326
performance and operational considerations when implementing, 324
session identifier types
supported in WebSphere 4 and 5, 327
session management
configuration levels in WebSphere 5, 327
with cookies, 327
example of with WebSphere, 325
example with WebSphere in a failover scenario, 326
with SSL ID session linkage, 328 “329
with URL rewriting, 328
session management scope
new supported in WebSphere 5, 327
Session Manager
configuring to support a session identifier type, 327
session persistence
option differences between WebSphere 4 and 5, 363 “364
three forms of standard with WebSphere 4 and 5, 363
use of session database for, 51
session state
considerations for multiple JVM environments, 185
Session type
in Oracle Transparent Application Failover, 368
shared memory synchronization
role of Solaris process semaphores in, 265 “266
shared_pool_size parameter
for Oracle system configuration and tuning, 438
Simple Object Access Protocol (SOAP)
legacy systems integration via XML, 455
standard data types for use in SOAP envelopes, 466
Single Instruction/Multiple Data (SIMD) instructions
added to the PowerPC 970 processor, 115
single-channel 3-tier topology
advantages of over the 2-tier model, 181
contexts in, 179
database transaction load considerations, 180 “183
disadvantages of over the 2-tier model, 182
example of high level, 178
final considerations for using, 189 “190
final recommendations for using, 183
how it works, 178 “183
single-channel topologies
advantages of, 176 “177
combined application server and database server in, 175 “176
components of, 174 “178
disadvantages of, 177
vs. dual or multichannel, 174
example of a high level, 174
example of a high-level 2-tier, 175
example of high-level 3-tier, 178
final considerations for using, 177 “178
multiapplication JVM, 183 “190
in WebSphere, 173 “204
single-threaded model
importance of not using with servlets, 400
SLA. See Service Level Agreement (SLA)
SOAP. See Simple Object Access Protocol (SOAP)
socket buffer memory allocation
increasing, 257 “258
soft costs
associated with TCO, 21
software clustering (proprietary clustering)
function of, 293 “294
software-based errors
in Windows and Unix networks, 493
Solaris. See Sun Solaris
Solaris process semaphores. See process semaphores
Solaris swap space command, 480
sort_area_size parameter
for Oracle system configuration and tuning, 438
SortHeap parameter
setting in IBM DB2, 441
SPARC Compliance Definition
creation of, 95
SPARC RISC platform (Sun Microsystems)
examining, 94 “102
overview, 95
platform architecture, 95 “102
SPARC v9 processors
comparison chart, 102 “104
SPARC V9 RISC architecture
UltraSPARC II and III processors based on, 96
SPARC-based server platform
selecting for your needs, 104
split-brain WebSphere configuration
failure issues with, 207 “208
in multi-Web server, multiapplication server topologies, 207 “208
non-Web-based application considerations, 213 “214
spreadsheet applications
plotting and analyzing tuning data with, 247
SQL Query Analyzer
capturing performance management testing results with, 26 “27
SQL statements
management of, 419 “421
using prepared vs. embedded SQL commands, 419
using prepared where possible in data tier, 409
SQL transactions
determining type you're looking to support, 430
sqlplus tool
use of in Oracle, 438
SSL (Secure Socket Layer)
considerations in system and hardware tuning, 277 “279
increasing the number of requests per connection, 277 “279
SSL ciphers
availability of, 279
SSL crypto interface adapters
for increasing SSL-based performance, 278
SSL ID session linkage
session management with, 328 “329
stand-alone database instances
database availability option, 305
stand-alone databases
function of, 219
stateful EJBs
importance of avoiding where possible, 404
sticky routing
using in split-brain WebSphere configuration, 207 “208
stmt.close() command
closing the statement object with, 421
Storage Area Network (SAN) technology
function of, 124 “125
overview of high-level implementation, 124
storage technologies. See disk storage technologies
stress and volume testing`
importance of, 15
of your WebSphere applications, 245 “246
Stress and Volume Testing (SVT) environment
testing your applications in, 504 “505
stress testing vs. volume testing, 245 “246
striping disks
calculating performance of, 136 “137
Structured Query Language (SQL) statement controls
effect on performance, 8
stub, 59
Sun HotSpot JVM
managing the minimum and maximum heap sizes for, 288 “289
tuning and performance options, 288 “289
Sun Microsystems
SPARC platform architectures, 95 “102
SPARC platform overview, 95
Sun Microsystems SunCluster
clustering solution, 294
as solution for applications using NFS, 349
Sun ONE Directory Server
supported by WebSphere, 344
Sun server models
CPU and server comparison matrix, 104
Sun Solaris
command for getting network interface statistics in, 491
networking tuning and optimization, 260 “266
setting TCP keepalive settings, 261 “262
Sun Solaris operating system
as next best for your WebSphere implementation, 108 “109
Sun Solaris WebSphere-based applications
changing the TCP keepalive settings in, 261 “262
enabling or disabling the TCP time stamp, 264
FIN_WAIT_2 problems with, 263
maximum TCP buffer size , 264
process semaphores, 265 “266
setting size of TCP connection hash table in, 265
setting the TCP TIME_WAIT interval for, 262
Sun SPARC-based server platform
selecting for your needs, 104
summary, 108 “109
Sun SPARC platform
feature of, 145 “147
grid computing in, 146 “147
hot-swappable components, 146
Sun SPARC v9 processors
comparison chart, 102 “104
Sun UltraSPARC II
architecture and function of, 97 “98
best-use model for, 102
Sun UltraSPARC IIe
architecture and function of, 99
best-use model for, 103
Sun UltraSPARC IIi
architecture and function of, 98 “99
Sun UltraSPARC III
architecture and function of, 99 “101
Sun UltraSPARC IIi
best-use model for, 102 “103
Sun UltraSPARC III
best-use model for, 103
key components of, 100
vs. UltraSPARC IIi, 99
Sun UltraSPARC IIIi
architecture and function of, 102
best-use model for, 103 “104
Sun-based server models
identifying, 96
SunFire Enterprise class servers
hot-swappable and hot-pluggable components in, 146
from Sun Microsystems, 95
SunFire range of servers
choosing for medium production environments, 106 “107
hot-swappable and/or redundant parts in, 106 “107
recommendation for large production environments, 108
swap space
in Unix, 480 “481
SX processors, 83
sync() method call
manually persisting session data with, 364 “365
system and hardware tuning and optimization
disk layout approaches, 272 “275
network configuration, 276 “277
for WebSphere-based platforms, 271 “277
for Windows-based file systems, 275
system configuration
effect of workload on, 433
System Interface Unit (SIU)
in UltraSPARC III processor, 100
system performance
impact of buggy software or unstable hardware on, 239 “240
system redundancy
understanding, 37
system utilization
monitoring, 474 “485
system.out.println()
using a multithreaded logger instead of, 402