b******y
发帖数: 1684 | |
A**o
发帖数: 1550 | 2 profiler?
【在 b******y 的大作中提到】
: 有什么好书么？
|
b******y
发帖数: 1684 | 3 nono...
it's about how to write performance-aware java
such as stringbuffer instead of string, factory instead of instances
【在 A**o 的大作中提到】
: profiler?
|
t*******e
发帖数: 684 | 4 performance bottleneck在数据库，Java programming把memory usage管好就行了。 |
m******t
发帖数: 2416 | 5
LOL, I think a good starting point to improve
your ejbs' performance is stop using them.
【在 b******y 的大作中提到】
: nono...
: it's about how to write performance-aware java
: such as stringbuffer instead of string, factory instead of instances
|
g*****g
发帖数: 34805 | 6 How is ejb3's performance, they look like spring bean and I expect
similar performance, shouldn't be bad right?
【在 m******t 的大作中提到】
:
: LOL, I think a good starting point to improve
: your ejbs' performance is stop using them.
|
m******t
发帖数: 2416 | 7
I don't have first-hand experience with ejb3
so I can't really comment on it.
Although I think that the performance problem
of classic ejbs mostly comes from not the
wiring, but the invocation boundary - all
the marshalling/going over the wire/unmarshalling,
which probably hardly changed in ejb3.
It's not really that bad for stateless session
beans though, their API being very coarse
grained.
【在 g*****g 的大作中提到】
: How is ejb3's performance, they look like spring bean and I expect
: similar performance, shouldn't be bad right?
|
b******y
发帖数: 9224 | 8 Right, EJB3 is a waste, IMHO |
t*******e
发帖数: 684 | 9 EJB3还是可以的,从Spring那儿学了不少好的东西。EJB3.1 light weight EJB会更有点
意思. |
b******y
发帖数: 9224 | 10 偶是被Sun的玩意儿吓怕了。每次Sun推出的东东，总是感觉笨重，而且不合developer
的要求，呵呵 |
t*******e
发帖数: 684 | 11 以前是这样的，现在SUN改变了不少, 新的JSR规范都不再是由一两个公司提出了，整个
Java Community都可以参与. Java.net和GlassFish都搞的很不错 |
b******y
发帖数: 9224 | |
