|
- cat /proc/user_beancounters | grep 'shmpage\|failcnt'
- uid resource held maxheld barrier limit failcnt
- shmpages 524288 524288 524288 524288 10
复制代码 # dd if=/dev/zero of=./memtest bs=1M count=100
100+0 records in
100+0 records out
104857600 bytes (105 MB) copied, 0.0692311 s, 1.5 GB/s
写100M,没事,速度挺快的
failcnt也没变化
- free
- total used free shared buffers cached
- Mem: 2621440 183940 2437500 0 0 34112
- -/+ buffers/cache: 149828 2471612
- Swap: 0 0 0
复制代码 写600M
# dd if=/dev/zero of=./memtest bs=1M count=600
600+0 records in
600+0 records out
629145600 bytes (629 MB) copied, 3.63775 s, 173 MB/s
明显优势, 就是往硬盘里写了
- # cat /proc/user_beancounters | grep 'shmpage\|failcnt'
- uid resource held maxheld barrier limit failcnt
- shmpages 524288 524288 524288 524288 12
复制代码 failcnt也增加了
2.5 G内存,我实际用了 100M不到
- # free
- total used free shared buffers cached
- Mem: 2621440 71368 2550072 0 0 33444
- -/+ buffers/cache: 37924 2583516
- Swap: 0 0 0
复制代码 太黑心了
|
|