Merge pull request #363 from apache/add_max_open_file

Add max open file

README.md

@@ -81,6 +81,8 @@ Then the binary version (including both server and client) can be found at **dis
 
 ### Configurations
 
+Set the max open file num as 65535 to avoid "too many open files" problem.
+
 configuration files are under "conf" folder
 
   * environment config module (`iotdb-env.bat`, `iotdb-env.sh`), 

docs/Documentation-CHN/QuickStart.md

@@ -61,6 +61,8 @@ IoTDB支持多种安装途径。用户可以使用三种方式对IoTDB进行安
 
 ### 启动IoTDB
 
+设置最大文件打开数为65535。
+
 用户可以使用$IOTDB_HOME/bin文件夹下的start-server脚本启动IoTDB。
 
 Linux系统与MacOS系统启动命令如下：

docs/Documentation/QuickStart.md

@@ -70,6 +70,8 @@ Then the binary version (including both server and client) can be found at **dis
 
 ### Configurations
 
+Set the max open file num as 65535 to avoid "too many open files" problem.
+
 configuration files are under "conf" folder
 
   * environment config module (`iotdb-env.bat`, `iotdb-env.sh`), 

server/src/assembly/resources/conf/iotdb-engine.properties

@@ -129,12 +129,6 @@ timestamp_precision=ms
 # The parameter form is a:b:c, where a, b and c are integers. for example: 1:1:1 , 6:3:1
 write_read_free_memory_proportion=6:3:1
 
-# The maximum concurrent thread number for merging
-# Increase this value, it will increase IO and CPU consumption
-# Decrease this value, when there is much unsequence data, it will increase disk usage, which will reduce read speed
-# When the value<=0 or > CPU core number, use the CPU core number.
-merge_concurrent_threads=0
-
 # The amount of data read each time in batch (the number of data strips, that is, the number of different timestamps.)
 fetch_size=10000