[bigdata] flume file channel CPU消耗比 memory channel高的原因
2015-12-25 17:44
323 查看
https://www.quora.com/Why-does-flume-take-more-resource-CPU-when-file-channel-is-used-compared-to-when-memory-channel-is-used
In case of File channel , the CPU would be used for the following
Serializing/Deserializing Events from/to file channel . In memory channel , this is plainly stored in RAM , so no serialization is required.
A small cpu overhead per disk write in determining the disk location where it needs to write. Typically this will be backed by swap file . So you will not see this per event. But will see CPU spikes when moving from physical to swap.
So this increase is not entirely unexpected. However , you should look closer if this increase in CPU is high enough to slow down the system.
In case of File channel , the CPU would be used for the following
Serializing/Deserializing Events from/to file channel . In memory channel , this is plainly stored in RAM , so no serialization is required.
A small cpu overhead per disk write in determining the disk location where it needs to write. Typically this will be backed by swap file . So you will not see this per event. But will see CPU spikes when moving from physical to swap.
So this increase is not entirely unexpected. However , you should look closer if this increase in CPU is high enough to slow down the system.
相关文章推荐
- 数据结构课程设计--“模拟图书管理系统的开发”
- sharesdk for android集成调试的几个问题
- Mount usb device
- 欢迎使用CSDN-markdown编辑器
- Log4j2使用总结
- Hdu 2045之解题报告
- redis memcache
- PHP - 魔术方法
- pngCanvas 是一个使用纯Python代码的生成png图像的工具
- hdu acm 3836 Equivalent Sets
- 熵,相对熵和互信息
- bcm4330蓝牙分析总结 .
- Spring MVC 系统异常处理方式
- Linux进程实践(2) --僵尸进程与文件共享
- leetcode笔记:First Bad Version
- OC中Foundation框架的基本对象之数字对象
- java hashCode
- My to do 12.25
- [转]Android之Context和Activity互相转换
- 面向对象的六大原则