我是小小鸟
- 浏览: 239795 次
- 性别:
-
文章分类
最新评论
测试代码
使用serial + serial Old
full gc通过(System.gc())触发
年轻代垃垃圾收集器名字 DefNew
老年代Tenured 持久带 Perm
年轻代收集过程
从jstat看，只触发了YGC
老年代和持久带的收集过程
只触发了FGC，但是却没有YGC
如果开启Syste.gc()，也是一样，只有FGC，没有YGC
Parallel Scavenge +Parallel Old 组合显示的结果，full gc通过(System.gc())触发，
年轻代收集器名称 PSYoungGen
老年代ParOldGen， 持久带PSPermGen
年轻代的收集过程，只触发YGC
通过程序触发的full gc，此时YGC，FGC都会触发
Parallel GC的线程名
使用ParNew + ConcMarkSweep ，full gc通过(System.gc())触发，
ParNew GC的线程名
CMS线程名
年轻代收集过程，此时只会触发YGC
老年代和持久带收集过程，只会触发FGC
通过System.gc()触发，此时只有FGC，没有YGC
CMS中有，初始化，并发标记，重新标记，并发清除几个阶段
以下是初始化，并发标记
重新标记，并发清除
此外还有扫描中断：
处理失败的情况，此时使用serial old：
最后是G1，其线程名如下：
清扫过程：由于G1不能很好的和jstat配合工作，因此看不到gcutil情况
jdk1.7.0_04 打开G1，jstat，jmap等工具可以正常使用。
参考
Tenured 区并发垃圾回收器CMS介绍
http://blog.csdn.net/lxb_champagne/article/details/18357947
千丝万缕的FGC与Buffer Pool
http://fx114.net/qa-132-149018.aspx
public void test()throws Exception {
while(true) {
byte[] buf = new byte[3 * 1024 * 1024];
Thread.sleep(500);
//System.gc();
}
}
使用serial + serial Old
full gc通过(System.gc())触发
年轻代垃垃圾收集器名字 DefNew
老年代Tenured 持久带 Perm
年轻代收集过程
C:\>java -Xms20M -Xmx20M -Xmn10M -XX:+PrintGCDetails -XX:+ScavengeBeforeFullGC -
XX:+UseSerialGC Test2
[GC [DefNew: 6471K->146K(9216K), 0.0043318 secs] 6471K->146K(19456K), 0.0053627
secs] [Times: user=0.00 sys=0.02, real=0.01 secs]
[GC [DefNew: 6540K->146K(9216K), 0.0022464 secs] 6540K->146K(19456K), 0.0032348
secs] [Times: user=0.00 sys=0.00, real=0.00 secs]
Heap
def new generation total 9216K, used 3546K [0x315e0000, 0x31fe0000, 0x31fe000
0)
eden space 8192K, 41% used [0x315e0000, 0x31931f28, 0x31de0000)
from space 1024K, 14% used [0x31de0000, 0x31e04920, 0x31ee0000)
to space 1024K, 0% used [0x31ee0000, 0x31ee0000, 0x31fe0000)
tenured generation total 10240K, used 0K [0x31fe0000, 0x329e0000, 0x329e0000)
the space 10240K, 0% used [0x31fe0000, 0x31fe0000, 0x31fe0200, 0x329e0000)
compacting perm gen total 12288K, used 374K [0x329e0000, 0x335e0000, 0x369e000
0)
the space 12288K, 3% used [0x329e0000, 0x32a3dba0, 0x32a3dc00, 0x335e0000)
ro space 10240K, 54% used [0x369e0000, 0x36f5ee00, 0x36f5ee00, 0x373e0000)
rw space 12288K, 55% used [0x373e0000, 0x37a82800, 0x37a82800, 0x37fe0000)
从jstat看，只触发了YGC
C:\>jstat -gcutil 2528 1000 20 S0 S1 E O P YGC YGCT FGC FGCT GCT 14.29 0.00 37.50 0.00 3.04 12 0.029 0 0.000 0.029 0.00 14.29 37.50 0.00 3.04 13 0.032 0 0.000 0.032 14.29 0.00 37.50 0.00 3.04 14 0.034 0 0.000 0.034 0.00 14.29 37.50 0.00 3.04 15 0.036 0 0.000 0.036 0.01 0.00 37.50 1.43 3.04 16 0.038 0 0.000 0.038 0.00 0.00 37.50 1.43 3.04 17 0.040 0 0.000 0.040 0.00 0.00 37.50 1.43 3.04 18 0.041 0 0.000 0.041 0.00 0.00 37.50 1.43 3.04 19 0.042 0 0.000 0.042 0.00 0.00 37.50 1.43 3.04 20 0.043 0 0.000 0.043 0.00 0.00 37.50 1.43 3.04 21 0.044 0 0.000 0.044 0.00 0.00 37.50 1.43 3.04 22 0.044 0 0.000 0.044
老年代和持久带的收集过程
C:\>java -Xms20M -Xmx20M -Xmn10M -XX:+PrintGCDetails -XX:+ScavengeBeforeFullGC -
XX:PretenureSizeThreshold=2097152 -XX:+UseSerialGC Test2
[GC [Tenured: 9216K->146K(10240K), 0.0310260 secs] 9543K->146K(19456K), [Perm :
373K->373K(12288K)], 0.0322100 secs] [Times: user=0.03 sys=0.00, real=0.03 secs]
[GC [Tenured: 9362K->146K(10240K), 0.0238268 secs] 12762K->146K(19456K), [Perm :
373K->373K(12288K)], 0.0256443 secs] [Times: user=0.02 sys=0.00, real=0.01 secs
]
Heap
def new generation total 9216K, used 3563K [0x315e0000, 0x31fe0000, 0x31fe000
0)
eden space 8192K, 43% used [0x315e0000, 0x3195ae88, 0x31de0000)
from space 1024K, 0% used [0x31de0000, 0x31de0000, 0x31ee0000)
to space 1024K, 0% used [0x31ee0000, 0x31ee0000, 0x31fe0000)
tenured generation total 10240K, used 3218K [0x31fe0000, 0x329e0000, 0x329e00
00)
the space 10240K, 31% used [0x31fe0000, 0x32304a00, 0x32304a00, 0x329e0000)
compacting perm gen total 12288K, used 374K [0x329e0000, 0x335e0000, 0x369e000
0)
the space 12288K, 3% used [0x329e0000, 0x32a3dba0, 0x32a3dc00, 0x335e0000)
ro space 10240K, 54% used [0x369e0000, 0x36f5ee00, 0x36f5ee00, 0x373e0000)
rw space 12288K, 55% used [0x373e0000, 0x37a82800, 0x37a82800, 0x37fe0000)
只触发了FGC，但是却没有YGC
C:\>jstat -gcutil 2572 1000 20 S0 S1 E O P YGC YGCT FGC FGCT GCT 0.00 0.00 39.50 31.43 3.04 0 0.000 23 0.621 0.621 0.00 0.00 39.50 31.43 3.04 0 0.000 25 0.675 0.675 0.00 0.00 39.50 31.43 3.04 0 0.000 27 0.728 0.728 0.00 0.00 39.50 31.43 3.04 0 0.000 29 0.782 0.782 0.00 0.00 39.50 31.43 3.04 0 0.000 31 0.837 0.837 0.00 0.00 39.50 31.30 3.00 0 0.000 33 0.891 0.891
如果开启Syste.gc()，也是一样，只有FGC，没有YGC
C:\>java -Xms20M -Xmx20M -Xmn10M -XX:+PrintGCDetails -XX:+ScavengeBeforeFullGC -
XX:PretenureSizeThreshold=2097152 -XX:+UseSerialGC Test2
[Full GC (System) [Tenured: 3072K->3218K(10240K), 0.0281815 secs] 3399K->3218K(1
9456K), [Perm : 373K->373K(12288K)], 0.0297979 secs] [Times: user=0.03 sys=0.00,
real=0.03 secs]
[Full GC (System) [Tenured: 3218K->3218K(10240K), 0.0269135 secs] 6618K->3218K(1
9456K), [Perm : 373K->373K(12288K)], 0.0280661 secs] [Times: user=0.02 sys=0.00,
real=0.03 secs]
Heap
def new generation total 9216K, used 3563K [0x315e0000, 0x31fe0000, 0x31fe000
0)
eden space 8192K, 43% used [0x315e0000, 0x3195ae88, 0x31de0000)
from space 1024K, 0% used [0x31de0000, 0x31de0000, 0x31ee0000)
to space 1024K, 0% used [0x31ee0000, 0x31ee0000, 0x31fe0000)
tenured generation total 10240K, used 3218K [0x31fe0000, 0x329e0000, 0x329e00
00)
the space 10240K, 31% used [0x31fe0000, 0x32304930, 0x32304a00, 0x329e0000)
compacting perm gen total 12288K, used 374K [0x329e0000, 0x335e0000, 0x369e000
0)
the space 12288K, 3% used [0x329e0000, 0x32a3dbd8, 0x32a3dc00, 0x335e0000)
ro space 10240K, 54% used [0x369e0000, 0x36f5ee00, 0x36f5ee00, 0x373e0000)
rw space 12288K, 55% used [0x373e0000, 0x37a82800, 0x37a82800, 0x37fe0000)
C:\>jstat -gcutil 3724 1000 20 S0 S1 E O P YGC YGCT FGC FGCT GCT 0.00 0.00 0.00 31.43 3.04 0 0.000 20 0.535 0.535 0.00 0.00 39.50 31.43 3.04 0 0.000 21 0.551 0.551 0.00 0.00 39.50 31.43 3.04 0 0.000 23 0.604 0.604 0.00 0.00 39.50 31.43 3.04 0 0.000 25 0.659 0.659 0.00 0.00 39.50 31.43 3.04 0 0.000 27 0.715 0.715 0.00 0.00 39.50 31.43 3.04 0 0.000 29 0.768 0.768 0.00 0.00 39.50 31.43 3.04 0 0.000 31 0.823 0.823
Parallel Scavenge +Parallel Old 组合显示的结果，full gc通过(System.gc())触发，
年轻代收集器名称 PSYoungGen
老年代ParOldGen， 持久带PSPermGen
年轻代的收集过程，只触发YGC
C:\>java -Xmx15M -Xms15M -Xmn10M -XX:+PrintGCDetails -XX:+ScavengeBeforeFullGC - XX:PretenureSizeThreshold=2097152 -XX:+UseParallelOldGC -XX:ParallelGCThreads=10 Test2 [GC [PSYoungGen: 4419K->176K(8960K)] 8515K->4272K(15104K), 0.0034451 secs] [Time s: user=0.03 sys=0.00, real=0.02 secs] [GC [PSYoungGen: 4425K->168K(8960K)] 8521K->4264K(15104K), 0.0032476 secs] [Time s: user=0.00 sys=0.00, real=0.00 secs] Heap PSYoungGen total 8960K, used 4571K [0x06ff0000, 0x079f0000, 0x079f0000) eden space 7680K, 57% used [0x06ff0000,0x0743cd38,0x07770000) from space 1280K, 13% used [0x078b0000,0x078da030,0x079f0000) to space 1280K, 0% used [0x07770000,0x07770000,0x078b0000) ParOldGen total 6144K, used 4096K [0x069f0000, 0x06ff0000, 0x06ff0000) object space 6144K, 66% used [0x069f0000,0x06df0010,0x06ff0000) PSPermGen total 12288K, used 2104K [0x029f0000, 0x035f0000, 0x069f0000) object space 12288K, 17% used [0x029f0000,0x02bfe3c8,0x035f0000)
通过程序触发的full gc，此时YGC，FGC都会触发
C:\>java -Xms20M -Xmx20M -Xmn10M -XX:+PrintGCDetails -XX:-ScavengeBeforeFullGC -XX:PretenureSizeThreshold=2097152 -XX:+UseParallelOldGC Test2 [Full GC (System) [PSYoungGen: 3395K->0K(8960K)] [ParOldGen: 0K->3218K(10240K)] 3395K->3218K(19200K) [PSPermGen: 2091K->2091K(12288K)], 0.0317060 secs] [Times: user=0.05 sys=0.00, real=0.03 secs] [Full GC (System) [PSYoungGen: 3379K->0K(8960K)] [ParOldGen: 3218K->3218K(10240K )] 6597K->3218K(19200K) [PSPermGen: 2097K->2097K(12288K)], 0.0238544 secs] [Time s: user=0.01 sys=0.00, real=0.02 secs] [Full GC (System) [PSYoungGen: 3225K->0K(8960K)] [ParOldGen: 3218K->3218K(10240K )] 6443K->3218K(19200K) [PSPermGen: 2097K->2097K(12288K)], 0.0254747 secs] [Time s: user=0.02 sys=0.00, real=0.02 secs] Heap PSYoungGen total 8960K, used 3532K [0x073f0000, 0x07df0000, 0x07df0000) eden space 7680K, 46% used [0x073f0000,0x077633a0,0x07b70000) from space 1280K, 0% used [0x07cb0000,0x07cb0000,0x07df0000) to space 1280K, 0% used [0x07b70000,0x07b70000,0x07cb0000) ParOldGen total 10240K, used 3218K [0x069f0000, 0x073f0000, 0x073f0000) object space 10240K, 31% used [0x069f0000,0x06d14928,0x073f0000) PSPermGen total 12288K, used 2104K [0x029f0000, 0x035f0000, 0x069f0000) object space 12288K, 17% used [0x029f0000,0x02bfe108,0x035f0000)
Parallel GC的线程名
"GC task thread#0 (ParallelGC)" prio=6 tid=0x0031e000 nid=0x988 runnable
使用ParNew + ConcMarkSweep ，full gc通过(System.gc())触发，
ParNew GC的线程名
"Gang worker#0 (Parallel GC Threads)" prio=10 tid=0x0031bc00 nid=0x604 runnable
CMS线程名
"Concurrent Mark-Sweep GC Thread" prio=10 tid=0x0836e400 nid=0xe18 runnable "Gang worker#0 (Parallel CMS Threads)" prio=10 tid=0x082cc000 nid=0xd18 runnable
年轻代收集过程，此时只会触发YGC
C:\>java -Xmx20M -Xms20M -Xmn10M -XX:+PrintGCDetails -XX:+UseConcMarkSweepGC -X X:ParallelGCThreads=10 Test2 [GC [ParNew: 4423K->168K(9216K), 0.0055490 secs] 4423K->168K(19456K), 0.0062150 secs] [Times: user=0.00 sys=0.00, real=0.00 secs] [GC [ParNew: 4428K->176K(9216K), 0.0032652 secs] 4428K->176K(19456K), 0.0044014 secs] [Times: user=0.03 sys=0.00, real=0.02 secs] Heap par new generation total 9216K, used 4600K [0x02e10000, 0x03810000, 0x0381000 0) eden space 8192K, 54% used [0x02e10000, 0x03261f28, 0x03610000) from space 1024K, 17% used [0x03610000, 0x0363c3c0, 0x03710000) to space 1024K, 0% used [0x03710000, 0x03710000, 0x03810000) concurrent mark-sweep generation total 10240K, used 0K [0x03810000, 0x04210000, 0x04210000) concurrent-mark-sweep perm gen total 12288K, used 2104K [0x04210000, 0x04e10000 , 0x08210000)
老年代和持久带收集过程，只会触发FGC
C:\>java -Xmx20M -Xms20M -Xmn10M -XX:+PrintGCDetails -XX:+UseConcMarkSweepGC -X X:ParallelGCThreads=10 -XX:PretenureSizeThreshold=2097152 Test2 [GC [CMS: 8192K->146K(10240K), 0.0272448 secs] 8519K->146K(19456K), [CMS Perm : 2091K->2091K(12288K)], 0.0294118 secs] [Times: user=0.01 sys=0.02, real=0.03 sec s] [GC [CMS: 8338K->146K(10240K), 0.0229166 secs] 12762K->146K(19456K), [CMS Perm : 2097K->2097K(12288K)], 0.0248618 secs] [Times: user=0.01 sys=0.00, real=0.01 se cs] [GC [1 CMS-initial-mark: 8338K(10240K)] 12519K(19456K), 0.0010387 secs] [Times: user=0.00 sys=0.00, real=0.00 secs] [CMS-concurrent-mark: 0.009/0.009 secs] [Times: user=0.03 sys=0.00, real=0.02 se cs] [CMS-concurrent-preclean: 0.000/0.000 secs] [Times: user=0.00 sys=0.00, real=0.0 0 secs] [GC [CMS[CMS-concurrent-abortable-preclean: 0.000/0.035 secs] [Times: user=0.00 sys=0.00, real=0.03 secs] (concurrent mode failure): 8338K->146K(10240K), 0.0236156 secs] 12519K->146K(19 456K), [CMS Perm : 2097K->2097K(12288K)], 0.0257522 secs] [Times: user=0.02 sys= 0.00, real=0.03 secs] [GC [CMS: 8338K->146K(10240K), 0.0213714 secs] 12491K->146K(19456K), [CMS Perm : 2097K->2097K(12288K)], 0.0231105 secs] [Times: user=0.03 sys=0.00, real=0.03 se cs] [GC [CMS: 8338K->146K(10240K), 0.0224847 secs] 12472K->146K(19456K), [CMS Perm : 2097K->2097K(12288K)], 0.0243117 secs] [Times: user=0.03 sys=0.00, real=0.03 se cs] [GC [CMS: 8338K->146K(10240K), 0.0219254 secs] 12459K->146K(19456K), [CMS Perm : 2097K->2097K(12288K)], 0.0242763 secs] [Times: user=0.02 sys=0.01, real=0.03 se cs] Heap par new generation total 9216K, used 4440K [0x02e10000, 0x03810000, 0x0381000 0) eden space 8192K, 54% used [0x02e10000, 0x03266128, 0x03610000) from space 1024K, 0% used [0x03610000, 0x03610000, 0x03710000) to space 1024K, 0% used [0x03710000, 0x03710000, 0x03810000) concurrent mark-sweep generation total 10240K, used 4242K [0x03810000, 0x042100 00, 0x04210000) concurrent-mark-sweep perm gen total 12288K, used 2104K [0x04210000, 0x04e10000 , 0x08210000)
通过System.gc()触发，此时只有FGC，没有YGC
C:\>java -Xmx20M -Xms20M -Xmn10M -XX:+PrintGCDetails -XX:+UseConcMarkSweepGC -X X:ParallelGCThreads=10 -XX:PretenureSizeThreshold=2097152 -XX:+ScavengeBeforeFul lGC Test2 [Full GC (System) [CMS: 4096K->4242K(10240K), 0.0274647 secs] 4423K->4242K(19456 K), [CMS Perm : 2091K->2091K(12288K)], 0.0289805 secs] [Times: user=0.03 sys=0.0 0, real=0.03 secs] [Full GC (System) [CMS: 4242K->4242K(10240K), 0.0256547 secs] 8666K->4242K(19456 K), [CMS Perm : 2097K->2097K(12288K)], 0.0265903 secs] [Times: user=0.02 sys=0.0 0, real=0.02 secs] [Full GC (System) [CMS: 4242K->4242K(10240K), 0.0258008 secs] 8423K->4242K(19456 K), [CMS Perm : 2097K->2097K(12288K)], 0.0271459 secs] [Times: user=0.02 sys=0.0 0, real=0.01 secs] Heap par new generation total 9216K, used 4480K [0x02e10000, 0x03810000, 0x0381000 0) eden space 8192K, 54% used [0x02e10000, 0x032700b0, 0x03610000) from space 1024K, 0% used [0x03610000, 0x03610000, 0x03710000) to space 1024K, 0% used [0x03710000, 0x03710000, 0x03810000) concurrent mark-sweep generation total 10240K, used 4242K [0x03810000, 0x042100 00, 0x04210000) concurrent-mark-sweep perm gen total 12288K, used 2104K [0x04210000, 0x04e10000 , 0x08210000)
CMS中有，初始化，并发标记，重新标记，并发清除几个阶段
以下是初始化，并发标记
C:\>java -Xmx15M -Xms15M -Xmn10M -XX:+PrintGCDetails -XX:+UseConcMarkSweepGC -X X:ParallelGCThreads=10 -XX:PretenureSizeThreshold=2097152 -XX:+ScavengeBeforeFul lGC Test2
[GC [1 CMS-initial-mark: 4242K(6144K)] 4242K(15360K), 0.0007012 secs] [Times: us er=0.00 sys=0.00, real=0.00 secs] [CMS-concurrent-mark: 0.009/0.009 secs] [Times: user=0.02 sys=0.01, real=0.02 se cs]
重新标记，并发清除
[GC[YG occupancy: 4106 K (9216 K)][Rescan (parallel) , 0.0013488 secs][weak refs processing, 0.0000218 secs] [1 CMS-remark: 4242K(6144K)] 8349K(15360K), 0.00288 98 secs] [Times: user=0.00 sys=0.00, real=0.00 secs] [CMS-concurrent-sweep: 0.000/0.000 secs] [Times: user=0.01 sys=0.00, real=0.01 s ecs] [CMS-concurrent-reset: 0.001/0.001 secs] [Times: user=0.00 sys=0.00, real=0.00 s ecs]
此外还有扫描中断：
[Full GC (System) [CMS[CMS-concurrent-abortable-preclean: 0.001/0.499 secs] [Tim es: user=0.00 sys=0.00, real=0.50 secs] (concurrent mode interrupted): 4242K->4242K(6144K), 0.0266688 secs] 8363K->4242 K(15360K), [CMS Perm : 2097K->2097K(12288K)], 0.0283159 secs] [Times: user=0.03 sys=0.00, real=0.03 secs]
处理失败的情况，此时使用serial old：
(concurrent mode failure): 8338K->146K(10240K), 0.0260159 secs] 12519K->146K(19 456K), [CMS Perm : 2097K->2097K(12288K)], 0.0277560 secs] [Times: user=0.02 sys= 0.01, real=0.03 secs]
最后是G1，其线程名如下：
"Gang worker#0 (Parallel GC Threads)" prio=10 tid=0x08a4f800 nid=0x108d runnable "G1 Main Concurrent Mark GC Thread" prio=10 tid=0x08a6a400 nid=0x1090 runnable "Gang worker#0 (G1 Parallel Marking Threads)" prio=10 tid=0x08a7bc00 nid=0x1091 runnable "G1 Concurrent Refinement Thread#0" prio=10 tid=0x08a62800 nid=0x108f runnable "G1 Concurrent Refinement Thread#1" prio=10 tid=0x08a61000 nid=0x108e runnable
清扫过程：由于G1不能很好的和jstat配合工作，因此看不到gcutil情况
[GC pause (young), 0.00875300 secs]
[Parallel Time: 8.6 ms]
[GC Worker Start Time (ms): 19253.4]
[Update RS (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Processed Buffers : 0
Sum: 0, Avg: 0, Min: 0, Max: 0]
[Ext Root Scanning (ms): 7.5
Avg: 7.5, Min: 7.5, Max: 7.5]
[Mark Stack Scanning (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Scan RS (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Object Copy (ms): 0.1
Avg: 0.1, Min: 0.1, Max: 0.1]
[Termination (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Termination Attempts : 1
Sum: 1, Avg: 1, Min: 1, Max: 1]
[GC Worker End Time (ms): 19261.2]
[Other: 0.9 ms]
[Clear CT: 0.0 ms]
[Other: 0.1 ms]
[Choose CSet: 0.0 ms]
[ 12M->12M(20M)]
[Times: user=0.00 sys=0.00, real=0.01 secs]
[Full GC 12M->114K(20M), 0.0280980 secs]
[Times: user=0.02 sys=0.00, real=0.02 secs]
[GC pause (young), 0.00837900 secs]
[Parallel Time: 8.1 ms]
[GC Worker Start Time (ms): 21412.9]
[Update RS (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Processed Buffers : 0
Sum: 0, Avg: 0, Min: 0, Max: 0]
[Ext Root Scanning (ms): 6.8
Avg: 6.8, Min: 6.8, Max: 6.8]
[Mark Stack Scanning (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Scan RS (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Object Copy (ms): 0.1
Avg: 0.1, Min: 0.1, Max: 0.1]
[Termination (ms): 0.0
Avg: 0.0, Min: 0.0, Max: 0.0]
[Termination Attempts : 1
Sum: 1, Avg: 1, Min: 1, Max: 1]
[GC Worker End Time (ms): 21419.9]
[Other: 1.1 ms]
[Clear CT: 0.1 ms]
[Other: 0.1 ms]
[Choose CSet: 0.0 ms]
[ 12M->12M(20M)]
[Times: user=0.00 sys=0.00, real=0.01 secs]
[Full GC 12M->101K(20M), 0.0368190 secs]
[Times: user=0.04 sys=0.00, real=0.02 secs]
Heap
garbage-first heap total 20480K, used 8293K [0xaff00000, 0xb1300000, 0xb1300000)
region size 1024K, 1 young (1024K), 0 survivors (0K)
compacting perm gen total 12288K, used 1724K [0xb1300000, 0xb1f00000, 0xb5300000)
the space 12288K, 14% used [0xb1300000, 0xb14af300, 0xb14af400, 0xb1f00000)
No shared spaces configured.
jdk1.7.0_04 打开G1，jstat，jmap等工具可以正常使用。
C:\jdk1.7.0_04\bin>java -Xmx20M -Xms20M -Xmn10M -XX:+UseG1GC -XX:+PrintGCDet
ails Test2
[GC pause (young) (initial-mark), 0.00661676 secs]
[Parallel Time: 5.7 ms]
[GC Worker Start (ms): 747.9 747.9
Avg: 747.9, Min: 747.9, Max: 747.9, Diff: 0.0]
[Ext Root Scanning (ms): 2.1 1.9
Avg: 2.0, Min: 1.9, Max: 2.1, Diff: 0.2]
[Update RS (ms): 0.0 0.1
Avg: 0.1, Min: 0.0, Max: 0.1, Diff: 0.1]
[Processed Buffers : 0 5
Sum: 5, Avg: 2, Min: 0, Max: 5, Diff: 5]
[Scan RS (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Object Copy (ms): 3.3 3.3
Avg: 3.3, Min: 3.3, Max: 3.3, Diff: 0.1]
[Termination (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Termination Attempts : 1 1
Sum: 2, Avg: 1, Min: 1, Max: 1, Diff: 0]
[GC Worker End (ms): 753.4 753.4
Avg: 753.4, Min: 753.4, Max: 753.4, Diff: 0.0]
[GC Worker (ms): 5.5 5.5
Avg: 5.5, Min: 5.5, Max: 5.5, Diff: 0.0]
[GC Worker Other (ms): 0.3 0.3
Avg: 0.3, Min: 0.3, Max: 0.3, Diff: 0.0]
[Clear CT: 0.0 ms]
[Other: 0.9 ms]
[Choose CSet: 0.0 ms]
[Ref Proc: 0.1 ms]
[Ref Enq: 0.0 ms]
[Free CSet: 0.0 ms]
[Eden: 1024K(10M)->0B(9216K) Survivors: 0B->1024K Heap: 4663K(20M)->4560K(20M
)]
[Times: user=0.03 sys=0.00, real=0.02 secs]
[GC concurrent-root-region-scan-start]
[GC concurrent-root-region-scan-end, 0.0019206]
[GC concurrent-mark-start]
[GC concurrent-mark-end, 0.0012521 sec]
[GC remark [GC ref-proc, 0.0000235 secs], 0.0031993 secs]
[Times: user=0.00 sys=0.00, real=0.00 secs]
[GC cleanup 8697K->4601K(20M), 0.0008915 secs]
[Times: user=0.00 sys=0.02, real=0.02 secs]
[GC concurrent-cleanup-start]
[GC concurrent-cleanup-end, 0.0004232]
[GC pause (young) (initial-mark), 0.00516323 secs]
[Parallel Time: 4.3 ms]
[GC Worker Start (ms): 1281.0 1281.0
Avg: 1281.0, Min: 1281.0, Max: 1281.0, Diff: 0.0]
[Ext Root Scanning (ms): 1.5 1.4
Avg: 1.4, Min: 1.4, Max: 1.5, Diff: 0.1]
[Update RS (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Processed Buffers : 0 2
Sum: 2, Avg: 1, Min: 0, Max: 2, Diff: 2]
[Scan RS (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Object Copy (ms): 2.7 2.7
Avg: 2.7, Min: 2.7, Max: 2.7, Diff: 0.0]
[Termination (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Termination Attempts : 1 1
Sum: 2, Avg: 1, Min: 1, Max: 1, Diff: 0]
[GC Worker End (ms): 1285.2 1285.2
Avg: 1285.2, Min: 1285.2, Max: 1285.2, Diff: 0.0]
[GC Worker (ms): 4.2 4.2
Avg: 4.2, Min: 4.2, Max: 4.2, Diff: 0.0]
[GC Worker Other (ms): 0.1 0.1
Avg: 0.1, Min: 0.1, Max: 0.1, Diff: 0.0]
[Clear CT: 0.0 ms]
[Other: 0.9 ms]
[Choose CSet: 0.0 ms]
[Ref Proc: 0.1 ms]
[Ref Enq: 0.0 ms]
[Free CSet: 0.0 ms]
[Eden: 1024K(9216K)->0B(9216K) Survivors: 1024K->1024K Heap: 4601K(20M)->4544
K(20M)]
[Times: user=0.03 sys=0.00, real=0.03 secs]
[GC concurrent-root-region-scan-start]
[GC concurrent-root-region-scan-end, 0.0016821]
[GC concurrent-mark-start]
[GC concurrent-mark-end, 0.0005808 sec]
[GC remark [GC ref-proc, 0.0000212 secs], 0.0029291 secs]
[Times: user=0.00 sys=0.00, real=0.00 secs]
[GC cleanup 8640K->4544K(20M), 0.0008895 secs]
[Times: user=0.00 sys=0.00, real=0.02 secs]
[GC concurrent-cleanup-start]
[GC concurrent-cleanup-end, 0.0005897]
[GC pause (young) (initial-mark), 0.00520737 secs]
[Parallel Time: 4.3 ms]
[GC Worker Start (ms): 1812.2 1812.2
Avg: 1812.2, Min: 1812.2, Max: 1812.2, Diff: 0.0]
[Ext Root Scanning (ms): 1.5 1.3
Avg: 1.4, Min: 1.3, Max: 1.5, Diff: 0.2]
[Update RS (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Processed Buffers : 0 0
Sum: 0, Avg: 0, Min: 0, Max: 0, Diff: 0]
[Scan RS (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Object Copy (ms): 2.6 2.8
Avg: 2.7, Min: 2.6, Max: 2.8, Diff: 0.2]
[Termination (ms): 0.0 0.0
Avg: 0.0, Min: 0.0, Max: 0.0, Diff: 0.0]
[Termination Attempts : 1 1
Sum: 2, Avg: 1, Min: 1, Max: 1, Diff: 0]
[GC Worker End (ms): 1816.4 1816.4
Avg: 1816.4, Min: 1816.4, Max: 1816.4, Diff: 0.0]
[GC Worker (ms): 4.2 4.2
Avg: 4.2, Min: 4.2, Max: 4.2, Diff: 0.0]
[GC Worker Other (ms): 0.1 0.1
Avg: 0.1, Min: 0.1, Max: 0.1, Diff: 0.0]
[Clear CT: 0.0 ms]
[Other: 0.9 ms]
[Choose CSet: 0.0 ms]
[Ref Proc: 0.1 ms]
[Ref Enq: 0.0 ms]
[Free CSet: 0.0 ms]
[Eden: 0B(9216K)->0B(9216K) Survivors: 1024K->1024K Heap: 4544K(20M)->4544K(2
0M)]
[Times: user=0.03 sys=0.00, real=0.03 secs]
[GC concurrent-root-region-scan-start]
[GC concurrent-root-region-scan-end, 0.0014873]
[GC concurrent-mark-start]
[GC concurrent-mark-end, 0.0006668 sec]
[GC remark [GC ref-proc, 0.0000263 secs], 0.0028615 secs]
[Times: user=0.00 sys=0.00, real=0.00 secs]
[GC cleanup 8640K->4544K(20M), 0.0007965 secs]
[Times: user=0.00 sys=0.00, real=0.00 secs]
[GC concurrent-cleanup-start]
[GC concurrent-cleanup-end, 0.0006168]
Heap
garbage-first heap total 20480K, used 4544K [0x02d00000, 0x04100000, 0x041000
00)
region size 1024K, 2 young (2048K), 1 survivors (1024K)
compacting perm gen total 12288K, used 1572K [0x04100000, 0x04d00000, 0x081000
00)
the space 12288K, 12% used [0x04100000, 0x042893d0, 0x04289400, 0x04d00000)
No shared spaces configured.
参考
Tenured 区并发垃圾回收器CMS介绍
http://blog.csdn.net/lxb_champagne/article/details/18357947
千丝万缕的FGC与Buffer Pool
http://fx114.net/qa-132-149018.aspx
分享到：
发表评论
-
JVM字节码指令
2014-06-26 13:49 435实例分析Java Class的文件结构 http ... -
JVM字节码
2014-06-26 13:49 450实例分析Java Class的文件结构 h ... -
JVM参数调优
2014-06-23 20:00 486java -XX:+PrintFlags ... -
JVM自带的工具
2014-06-23 16:57 395appletviewer idlj ja ... -
JVM监控工具
2014-06-23 12:54 424JPS(虚拟机进程状态工具) 参数 含义 ... -
GC调优
2014-05-22 20:43 19-XX:+CMSIncrementalMode启用垃圾回收增 ... -
虚拟机相关
2012-12-21 19:03 658.Net 垃圾回收和大对象处理 http://blog.job ... -
Java字节码格式
2012-10-24 17:06 781实例分析Java Class的文件结构 http://cool ... -
jvm bug
2012-09-10 18:20 771jvm进程明明存在，但是执行jps，jstat却找不到 js ... -
JVM，内存相关
2012-04-04 17:20 934从 Java 代码到 Java 堆 http://www.ib ... -
调优实践
2012-03-29 20:43 69关于CMS垃圾回收 http://www.iteye.com/ ... -
JVM工具
2012-02-23 19:24 213JVM监控工具介绍jstack, jconsole, jinf ... -
JVM文章汇总
2012-02-03 17:46 881JVM线程知多少 http://www.goldendoc.o ... -
JPDA文章汇总
2012-02-02 16:15 1313深入 Java 调试体系: 第 1 部分，JPDA 体系概览 ... -
GC调优文章
2012-02-01 17:14 763官方： http://www.oracle.com/techn ...
相关推荐
性能工程师指南：玩转OpenJDK HotSpot垃圾收集器.ppt性能工程师指南：玩转OpenJDK HotSpot垃圾收集器.ppt性能工程师指南：玩转OpenJDK HotSpot垃圾收集器.ppt性能工程师指南：玩转OpenJDK HotSpot垃圾收集器.ppt性能...
性能工程师指南：玩转OpenJDK HotSpot垃圾收集器（83页）.pdf
Performance Engineering Performance Requirements & Test Plan & Development Performance Analysis
JVM，JVM内存结构、HotSpot 虚拟机对象探秘、垃圾收集策略与算法、HotSpot 垃圾收集器、内存分配与回收策略、JVM 性能调优、类文件结构。。
gcviz 是垃圾收集器可视化工具/框架。gcviz 是帮助生成 gc.log 可视化的程序集合。HotSpot，一个 Java 可视化机制，用以下标记配置： -verbose:gc -verbose:sizes -Xloggc:/apps/tomcat/logs/gc.log -XX: ...
Cache、Perf Data、Crash 分析方法、转储分析方法、 垃圾收集器的设计演进、CMS 和 G1 收集器、栈、JVM 对硬件寄存器的利用、栈顶缓存技术、解释器、字节 码表、转发表、Stubs、Code Cache、Code 生成器、JIT 编译器...
包括OpenJDK与HotSpot项目、编译和调试HotSpot的方法、HotSpot内核结构、Launcher、OOP-Klass对象表示系统、链接、运行时数据区、方法区、常量池和常量池Cache、Perf Data、Crash分析方法、转储分析方法、垃圾收集器...
本教程介绍了垃圾收集如何与Hotspot JVM一起使用的基础知识。了解垃圾收集器的工作原理后，学习如何使用Visual VM监视垃圾收集过程。最后，了解Java SE 7 Hotspot JVM中可用的垃圾收集器。
HotSpot正是目前世界上java虚拟机的最好的实现。 HotSpot的基础代码是许多人辛勤劳动的结晶，这个过程迄今已持续了超过10年的时间（当然...当然还有大约3个垃圾收集器以及一系列用来实现同步的高性能运行库文件等等。
随着垃圾收集器的发展，每一代 GC 与其上一代相比，都带来了巨大的进步和改善。parallel GC 与 serial GC 相比，它让垃圾收集器以多线程的方式工作，充分利用了多核计算机的计算能力。CMS(“Concurrent Mark-...
比如：字节码的编译原理、字节码的内部组成结构、通过源码的方式剖析HotSpot VM 的启动过程和初始化过程、Java 虚拟机的运行时内存、垃圾收集算法、垃圾收集器（重点讲解了Serial 收集器、ParNew 收集器、Parallel ...
Jvm中有多种垃圾收集器，如串行（Serial）垃圾收集器、并行（Parallel）垃圾收集器、并发（Concurrent）垃圾收集器等。其中，串行垃圾收集器是Jvm默认的垃圾收集器，它使用单线程来执行垃圾回收。并行垃圾收集器使用...
本书以极其精练的语句诠释了HotSpot VM 的方方面面，比如：字节码的编译原理、字节码的内部组成结构、通过源码的方式剖析HotSpot VM 的启动过程和初始化过程、Java 虚拟机的运行时内存、垃圾收集算法、垃圾收集器...
本书以极其精练的语句诠释了HotSpot VM 的方方面面，比如：字节码的编译原理、字节码的内部组成结构、通过源码的方式剖析HotSpot VM 的启动过程和初始化过程、Java 虚拟机的运行时内存、垃圾收集算法、垃圾收集器...
本书以极其精练的语句诠释了 HotSpot VM的方方面面，比如：字节码的编译原理、字节码的内部组成结构、通过源码的方式剖析 HotSpot VM 的启动过程和初始化过程、Java 虚拟机的运行时内存、垃圾收集算法、垃圾收集器...
一个干净的垃圾收集器接口，用来改善垃圾收集器源代码之间的隔离效果，这样可以为HotSpot 虚拟机中的内部垃圾收集代码提供更好的模块化功能，也可以更容易向 HotSpot 添加新的垃圾收集器。 并行、完整的 G1 垃圾...
一个干净的垃圾收集器接口，用来改善垃圾收集器源代码之间的隔离效果，这样可以为HotSpot 虚拟机中的内部垃圾收集代码提供更好的模块化功能，也可以更容易向 HotSpot 添加新的垃圾收集器。 并行、完整的 G1 垃圾...
一个干净的垃圾收集器接口，用来改善垃圾收集器源代码之间的隔离效果，这样可以为HotSpot 虚拟机中的内部垃圾收集代码提供更好的模块化功能，也可以更容易向 HotSpot 添加新的垃圾收集器。 并行、完整的 G1 垃圾...
GC 日志用于记录垃圾收集器的执行情况，可以用于调试和优化垃圾收集器。 内存分配 Java 中的内存分配可以分为以下几种： * 栈上分配（Stack Allocation）：在栈上分配内存。 * 堆上分配（Heap Allocation）：在堆...