topic title: Benchmark RAM?
3 posts
• Page 1 of 1
-
Posts: 65
- Joined: 14 Sep 2017
#1
My BIOS has a RAM setting which includes 'Turbo' and 'Ultra'. Ultra is not usable and gives long beeps during POST but I'd like to see what effect Turbo has, if any. What would be the easiest way to test this? The results for"benchmark" in SPM seem unrelated to this application. Thanks.
-
Posts: 1,445
- Joined: 09 Feb 2012
#2
debian package mbw
tiny 25kb utility +dependencies (mysql-client) ~= 3.4Mb
after installling, read `man mbw`
"MBW determines the copy memory bandwidth available to userspace programs.
Its simplistic approach models that of real applications.
It is not tuned to extremes and it is not aware of hardware architecture, just like your average software package.
.
.
debian package sysbench
tiny 150Kb utility ~~ after installing, refer to its manpage
tool for evaluating OS parameters that are important for a system running a database under intensive load.
The idea of this benchmark suite is to quickly get an impression about system performance
without setting up complex database benchmarks or even without installing a database at all.
Current features allow to test the following system parameters:
tiny 25kb utility +dependencies (mysql-client) ~= 3.4Mb
after installling, read `man mbw`
"MBW determines the copy memory bandwidth available to userspace programs.
Its simplistic approach models that of real applications.
It is not tuned to extremes and it is not aware of hardware architecture, just like your average software package.
.
.
debian package sysbench
tiny 150Kb utility ~~ after installing, refer to its manpage
tool for evaluating OS parameters that are important for a system running a database under intensive load.
The idea of this benchmark suite is to quickly get an impression about system performance
without setting up complex database benchmarks or even without installing a database at all.
Current features allow to test the following system parameters:
- file I/O performance
- scheduler performance
- memory allocation and transfer speed
- POSIX threads implementation performance
- database server performance (OLTP benchmark)
-
Posts: 65
- Joined: 14 Sep 2017
#3
Thanks. Sysbench didn't work and I was told to use a more up-to-date version than what Synaptic Package Manager gave me but mbw program worked:
This is the stock result:
This is the result with SDP set to Turbo and Command Rate changed from T2 to T1. It actually looks marginally worse for the MCBLOCK method but may something was going on in the background.
This is the stock result:
Code: Select all
sean@Teleprompter:~
$ mbw 512
Long uses 4 bytes. Allocating 2*134217728 elements = 1073741824 bytes of memory.
Using 262144 bytes as blocks for memcpy block copy test.
Getting down to business... Doing 10 runs per test.
0 Method: MEMCPY Elapsed: 173.65217 MiB: 512.00000 Copy: 2.948 MiB/s
1 Method: MEMCPY Elapsed: 104.83627 MiB: 512.00000 Copy: 4.884 MiB/s
2 Method: MEMCPY Elapsed: 201.39472 MiB: 512.00000 Copy: 2.542 MiB/s
3 Method: MEMCPY Elapsed: 328.69387 MiB: 512.00000 Copy: 1.558 MiB/s
4 Method: MEMCPY Elapsed: 331.46844 MiB: 512.00000 Copy: 1.545 MiB/s
5 Method: MEMCPY Elapsed: 308.48620 MiB: 512.00000 Copy: 1.660 MiB/s
6 Method: MEMCPY Elapsed: 328.49995 MiB: 512.00000 Copy: 1.559 MiB/s
7 Method: MEMCPY Elapsed: 340.92937 MiB: 512.00000 Copy: 1.502 MiB/s
8 Method: MEMCPY Elapsed: 407.30292 MiB: 512.00000 Copy: 1.257 MiB/s
9 Method: MEMCPY Elapsed: 369.14551 MiB: 512.00000 Copy: 1.387 MiB/s
AVG Method: MEMCPY Elapsed: 289.44094 MiB: 512.00000 Copy: 1.769 MiB/s
0 Method: DUMB Elapsed: 427.22051 MiB: 512.00000 Copy: 1.198 MiB/s
1 Method: DUMB Elapsed: 389.61728 MiB: 512.00000 Copy: 1.314 MiB/s
2 Method: DUMB Elapsed: 285.49182 MiB: 512.00000 Copy: 1.793 MiB/s
3 Method: DUMB Elapsed: 382.64656 MiB: 512.00000 Copy: 1.338 MiB/s
4 Method: DUMB Elapsed: 320.39583 MiB: 512.00000 Copy: 1.598 MiB/s
5 Method: DUMB Elapsed: 377.50293 MiB: 512.00000 Copy: 1.356 MiB/s
6 Method: DUMB Elapsed: 401.21181 MiB: 512.00000 Copy: 1.276 MiB/s
7 Method: DUMB Elapsed: 388.87290 MiB: 512.00000 Copy: 1.317 MiB/s
8 Method: DUMB Elapsed: 387.43153 MiB: 512.00000 Copy: 1.322 MiB/s
9 Method: DUMB Elapsed: 379.03232 MiB: 512.00000 Copy: 1.351 MiB/s
AVG Method: DUMB Elapsed: 373.94235 MiB: 512.00000 Copy: 1.369 MiB/s
0 Method: MCBLOCK Elapsed: 22.99905 MiB: 512.00000 Copy: 22.262 MiB/s
1 Method: MCBLOCK Elapsed: 7.79753 MiB: 512.00000 Copy: 65.662 MiB/s
2 Method: MCBLOCK Elapsed: 8.19003 MiB: 512.00000 Copy: 62.515 MiB/s
3 Method: MCBLOCK Elapsed: 4.19134 MiB: 512.00000 Copy: 122.157 MiB/s
4 Method: MCBLOCK Elapsed: 3.75465 MiB: 512.00000 Copy: 136.364 MiB/s
5 Method: MCBLOCK Elapsed: 3.74274 MiB: 512.00000 Copy: 136.798 MiB/s
6 Method: MCBLOCK Elapsed: 4.02775 MiB: 512.00000 Copy: 127.118 MiB/s
7 Method: MCBLOCK Elapsed: 3.81288 MiB: 512.00000 Copy: 134.282 MiB/s
8 Method: MCBLOCK Elapsed: 3.79867 MiB: 512.00000 Copy: 134.784 MiB/s
9 Method: MCBLOCK Elapsed: 3.80848 MiB: 512.00000 Copy: 134.437 MiB/s
AVG Method: MCBLOCK Elapsed: 6.61231 MiB: 512.00000 Copy: 77.431 MiB/s
sean@Teleprompter:~
$ `
This is the result with SDP set to Turbo and Command Rate changed from T2 to T1. It actually looks marginally worse for the MCBLOCK method but may something was going on in the background.
Code: Select all
sean@Teleprompter:~
$ mbw 512
Long uses 4 bytes. Allocating 2*134217728 elements = 1073741824 bytes of memory.
Using 262144 bytes as blocks for memcpy block copy test.
Getting down to business... Doing 10 runs per test.
0 Method: MEMCPY Elapsed: 101.34388 MiB: 512.00000 Copy: 5.052 MiB/s
1 Method: MEMCPY Elapsed: 100.03275 MiB: 512.00000 Copy: 5.118 MiB/s
2 Method: MEMCPY Elapsed: 105.45720 MiB: 512.00000 Copy: 4.855 MiB/s
3 Method: MEMCPY Elapsed: 117.65481 MiB: 512.00000 Copy: 4.352 MiB/s
4 Method: MEMCPY Elapsed: 116.01133 MiB: 512.00000 Copy: 4.413 MiB/s
5 Method: MEMCPY Elapsed: 73.83874 MiB: 512.00000 Copy: 6.934 MiB/s
6 Method: MEMCPY Elapsed: 72.36665 MiB: 512.00000 Copy: 7.075 MiB/s
7 Method: MEMCPY Elapsed: 162.53057 MiB: 512.00000 Copy: 3.150 MiB/s
8 Method: MEMCPY Elapsed: 192.22798 MiB: 512.00000 Copy: 2.664 MiB/s
9 Method: MEMCPY Elapsed: 55.38938 MiB: 512.00000 Copy: 9.244 MiB/s
AVG Method: MEMCPY Elapsed: 109.68533 MiB: 512.00000 Copy: 4.668 MiB/s
0 Method: DUMB Elapsed: 106.25649 MiB: 512.00000 Copy: 4.819 MiB/s
1 Method: DUMB Elapsed: 117.95109 MiB: 512.00000 Copy: 4.341 MiB/s
2 Method: DUMB Elapsed: 104.49727 MiB: 512.00000 Copy: 4.900 MiB/s
3 Method: DUMB Elapsed: 139.45409 MiB: 512.00000 Copy: 3.671 MiB/s
4 Method: DUMB Elapsed: 212.26644 MiB: 512.00000 Copy: 2.412 MiB/s
5 Method: DUMB Elapsed: 107.68112 MiB: 512.00000 Copy: 4.755 MiB/s
6 Method: DUMB Elapsed: 103.84393 MiB: 512.00000 Copy: 4.930 MiB/s
7 Method: DUMB Elapsed: 114.36213 MiB: 512.00000 Copy: 4.477 MiB/s
8 Method: DUMB Elapsed: 148.41306 MiB: 512.00000 Copy: 3.450 MiB/s
9 Method: DUMB Elapsed: 239.10295 MiB: 512.00000 Copy: 2.141 MiB/s
AVG Method: DUMB Elapsed: 139.38286 MiB: 512.00000 Copy: 3.673 MiB/s
0 Method: MCBLOCK Elapsed: 30.16894 MiB: 512.00000 Copy: 16.971 MiB/s
1 Method: MCBLOCK Elapsed: 6.20839 MiB: 512.00000 Copy: 82.469 MiB/s
2 Method: MCBLOCK Elapsed: 5.77876 MiB: 512.00000 Copy: 88.600 MiB/s
3 Method: MCBLOCK Elapsed: 4.41662 MiB: 512.00000 Copy: 115.926 MiB/s
4 Method: MCBLOCK Elapsed: 4.33213 MiB: 512.00000 Copy: 118.187 MiB/s
5 Method: MCBLOCK Elapsed: 4.26764 MiB: 512.00000 Copy: 119.973 MiB/s
6 Method: MCBLOCK Elapsed: 6.33055 MiB: 512.00000 Copy: 80.878 MiB/s
7 Method: MCBLOCK Elapsed: 3.86573 MiB: 512.00000 Copy: 132.446 MiB/s
8 Method: MCBLOCK Elapsed: 4.92002 MiB: 512.00000 Copy: 104.065 MiB/s
9 Method: MCBLOCK Elapsed: 5.02051 MiB: 512.00000 Copy: 101.982 MiB/s
AVG Method: MCBLOCK Elapsed: 7.53093 MiB: 512.00000 Copy: 67.986 MiB/s