[SYMBOL]
[A]
[B]
[C]
[D]
[E]
[F]
[G]
[H]
[I]
[J]
[K]
[L]
[M]
[N]
[O]
[P]
[Q]
[R]
[S]
[T]
[U]
[V]
[W]
[X]
[Z]
machine code (native), compiling to
Macintosh, TCP/IP stacks
main( ) (ObjectCreationMonitoring)
Map interface
implementations of
optimizing queries
map( ) (FileChannel)
marshalling time, inferring for Web Services
Matcher class find( )
group( )
matches( ) (String)
MatchReader class (example)
Math class, replacing with StrictMath
maxMemory( ) (Runtime)
measurements analyzing statistics from
application response times and throughput
benchmarks
checklist for
communication layers
for distributed applications
of I/O operations
disk I/O
multithreading, timing
network speeds
profilers and
serialization and
skewing by rewriting previously written objects
timing exceptions
transmission statistics
what to measure
what, where, and how
memcntl( )
memory
[See also heap, tuning] cycling through
freeing with garbage collection
2nd
3rd
[See also garbage collection] in-memory database products
leaks in
caused by vector handles
physical, locking heap in
profiling usage by runtime system
RAM
reusing objects, balancing against free memory needs
runtime, reducing
session state information, holding in
shared
I/O performance and
size of
2nd
system
updates to
VM startup time and
memory mapping of files
memory-mapped filesystems
2nd
MemoryMonitor class (example)
MemorySampler class (example)
Mercury Interactive Corporation analysis of enterprise performance
message calls, reducing
2nd
CORBA example
performance checklist
proprietary communications layer
remote invocations of EJBs
RMI example
Message Façade pattern in EJB design
message-driven beans in EJB, pooling
MessageListener interface
messaging broadcast, point-to-point, and unicast
Java Messaging Service
[See JMS] multicast publish-and-subscribe
metadata
performance checklist
method calls eliminating for performance gains
inlining with compiler
in loop termination tests, avoiding
moving out of loops
to native code
profiling tools for
-Xhprof option
HotSpot and 1.3 -Xprof
Java 2 -Xaprof option
Java 2 cpu=samples output
JDK 1.1.x -prof and Java 2 cpu=old
reflection, avoiding use of
RMI
[See RMI] method parameters
methods access control for, altering to speed up calls
bean life cycle, monitoring for excessive use
entry points
granularity of
hierarchical invocation, overhead involved in
inlining by compiler
mapping requests to
remote method invocation
[See RMI] removing unused with compiler
renaming in compiler optimization
static
avoiding creation of objects
synchronized
tail-recursive
2nd
[See also recursion] Microsoft, reimplementation and optimization of Pet Store application in .NET
mmap( )
monitoring performance applications
2nd
performance checklist for
J2EE tools for
downloading from web site
features to look for
operating systems
monitors locking methods
locking/unlocking
synchronization and
mount command (Unix)
multicast publish-and-subscribe messaging
multiplexing
I/O channels with NIO
multiplication, replacing with bit-shifting
multithreading
[See also threads] for application responsiveness
performance checklist
timing
multiuser applications, tuning
|