内存测试工具-Lmbench

简介：

lmbench是个用于评价系统综合性能的多平台开源benchmark，能够测试包括文档读写、内存操作、进程创建销毁开销、网络等性能，测试方法简单。

Lmbench是个多平台软件，因此能够对同级别的系统进行比较测试，反映不同系统的优劣势，通过选择不同的库函数我们就能够比较库函数的性能；更为重要的是，作为一个开源软件，lmbench提供一个测试框架，假如测试者对测试项目有更高的测试需要，能够通过少量的修改源代码达到目的（比如现在只能评测进程创建、终止的性能和进程转换的开销，通过修改部分代码即可实现线程级别的性能测试）。

下载：

www.bitmover.com/lmbench，最新版本3.0-a9

主要功能：

*带宽测评工具

—读取缓存文件

—拷贝内存

—读内存

—写内存

—管道

—TCP

* 反应时间测评工具

—上下文切换

—网络： 连接的建立，管道，TCP，UDP和RPC hot potato

—文件系统的建立和删除

—进程创建

—信号处理

—上层的系统调用

—内存读入反应时间

* 其他

—处理器时钟比率计算

测试过程：

 

1. To run the benchmark, you should be able to say:

cd src
# make results实际调用了两个脚本，scripts/lmbench 和 scripts/config-run
make results

 

 

1. If you want to see how you did compared to the other system resultsincluded here, say

# 实际调用命令为cd results && make summary percent 2>/dev/null | more
make see

 

其他：

Be warned that many of these benchmarks are sensitive to other things being run on the system, mainly from CPU cache and CPU cycle effects. So make sure your screen saver is not running, etc. It's a good idea to do several runs and compare the output like so

make results
make rerun
make rerun
make rerun
cd Results && make LIST=/*

 

报错及解决办法  

• make报错：

make[2]: *** 没有规则可以创建“bk.ver”需要的目标“../SCCS/s.ChangeSet”。 停止。
  make[2]:正在离开目录 `/home/hero/lmbench3/src'
  make[1]: *** [lmbench] 错误 2
  make[1]:正在离开目录 `/home/hero/lmbench3/src'
  make: *** [build] 错误 2

 

• 解决办法：

lmbench3目录下
  #mkdir SCCS
  #touch  ./SCCS/s.ChangeSet
  #make

 

测试结果：

 

cd results && make summary percent 2>/dev/null | more
make[1]: Entering directory `/root/test/lmbench3/results'

                 L M B E N C H  3 . 0   S U M M A R Y
                 ------------------------------------
     (Alpha software, do not distribute)


Processor, Processes - times in microseconds - smaller is better
------------------------------------------------------------------------------
Host                 OS  Mhz null null      open slct sig  sig  fork exec sh
                             call  I/O stat clos TCP  inst hndl proc proc proc
--------- ------------- ---- ---- ---- ---- ---- ---- ---- ---- ---- ---- ----
localhost Linux 3.10.0- 2304 0.37 0.48 1.49 2.79 8.57 0.56 1.68 155. 612. 2090

Basic integer operations - times in nanoseconds - smaller is better
-------------------------------------------------------------------
Host                 OS  intgr intgr  intgr  intgr  intgr
                          bit   add    mul    div    mod
--------- ------------- ------ ------ ------ ------ ------
localhost Linux 3.10.0- 0.4500 0.0500 1.4200   11.0   12.2

Basic float operations - times in nanoseconds - smaller is better
-----------------------------------------------------------------
Host                 OS  float  float  float  float
                         add    mul    div    bogo
--------- ------------- ------ ------ ------ ------
localhost Linux 3.10.0- 1.3500 2.2600 6.0600 3.2900

Basic double operations - times in nanoseconds - smaller is better
------------------------------------------------------------------
Host                 OS  double double double double
                         add    mul    div    bogo
--------- ------------- ------  ------ ------ ------
localhost Linux 3.10.0- 1.3400 2.2100 9.0700 6.3900

Context switching - times in microseconds - smaller is better
-------------------------------------------------------------------------
Host                 OS  2p/0K 2p/16K 2p/64K 8p/16K 8p/64K 16p/16K 16p/64K
                         ctxsw  ctxsw  ctxsw ctxsw  ctxsw   ctxsw   ctxsw
--------- ------------- ------ ------ ------ ------ ------ ------- -------
localhost Linux 3.10.0- 2.1300 2.3500 2.4200 2.8300 3.8200 3.16000 4.21000

*Local* Communication latencies in microseconds - smaller is better
---------------------------------------------------------------------
Host                 OS 2p/0K  Pipe AF     UDP  RPC/   TCP  RPC/ TCP
                        ctxsw       UNIX         UDP         TCP conn
--------- ------------- ----- ----- ---- ----- ----- ----- ----- ----
localhost Linux 3.10.0- 2.130 7.756 8.96  12.7        14.7        38.

File & VM system latencies in microseconds - smaller is better
-------------------------------------------------------------------------------
Host                 OS   0K File      10K File     Mmap    Prot   Page   100fd
                        Create Delete Create Delete Latency Fault  Fault  selct
--------- ------------- ------ ------ ------ ------ ------- ----- ------- -----
localhost Linux 3.10.0-   17.2   31.5  248.9   79.8   13.2K 0.754 1.95200 1.743

*Local* Communication bandwidths in MB/s - bigger is better
-----------------------------------------------------------------------------
Host                OS  Pipe AF    TCP  File   Mmap  Bcopy  Bcopy  Mem   Mem
                             UNIX      reread reread (libc) (hand) read write
--------- ------------- ---- ---- ---- ------ ------ ------ ------ ---- -----
localhost Linux 3.10.0- 2369 1745 2586 4274.1 7622.8 4708.3 3248.1 7833 5057.

Memory latencies in nanoseconds - smaller is better
    (WARNING - may not be correct, check graphs)
------------------------------------------------------------------------------
Host                 OS   Mhz   L1 $   L2 $    Main mem    Rand mem    Guesses
--------- -------------   ---   ----   ----    --------    --------    -------
localhost Linux 3.10.0-  2304 1.7310 8.6750   20.5       159.7
make[1]: Leaving directory `/root/test/lmbench3/results'

 

结果解析：

关键参数：

1. Memory latencies（内存操作延时）

L1：缓存1

L2：缓存2

Main Mem：连续内存

Rand Mem：内存随机访问延时

Guesses

假如L1和L2近似，会显示“No L1 cache?”

假如L2和Main Mem近似，会显示“No L2 cache?”

1. Local Communication bandwidths（本地通信带宽）

Pipe：管道操作

AF UNIX

Unix协议

TCP

TCP通信

File reread：文档重复读

MMap reread：内存映射重复读

Bcopy(libc)：内存拷贝

Bcopy(hand)：内存拷贝

Mem read：内存读

Mem write：内存写