Linux Install Memo

サーバー管理者によるLinux関連ソフトのインストールメモ

Home » やっぱりDDR3は速いなぁ…STREAMでメモリベンチ

やっぱりDDR3は速いなぁ…STREAMでメモリベンチ

cd /usr/src/
mkdir stream
cd ./stream
wget http://www.cs.virginia.edu/stream/FTP/Code/stream.c
gcc -O stream.c -o stream
./stream

gcc -O2 stream.c -o stream
./stream

gcc -O3 stream.c -o stream
./stream

 Intel(R) Xeon(TM) CPU 2.80GHz、2GB(DDR2)の場合

————————————————————-
STREAM version $Revision: 5.9 $
————————————————————-
This system uses 8 bytes per DOUBLE PRECISION word.
————————————————————-
Array size = 2000000, Offset = 0
Total memory required = 45.8 MB.
Each test is run 10 times, but only
the *best* time for each is used.
————————————————————-
Printing one line per active thread….
————————————————————-
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 10828 microseconds.
   (= 10828 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
————————————————————-
WARNING — The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
————————————————————-
Function      Rate (MB/s)   Avg time     Min time     Max time
Copy:        2117.8006       0.0151       0.0151       0.0153
Scale:       2148.5149       0.0149       0.0149       0.0151
Add:         2566.0102       0.0188       0.0187       0.0190
Triad:       2566.5663       0.0187       0.0187       0.0188
————————————————————-
Solution Validates
————————————————————-

Intel(R) Core(TM) i3 CPU 550  @ 3.20GHz、4GB(DDR3)の場合

————————————————————-
STREAM version $Revision: 5.9 $
————————————————————-
This system uses 8 bytes per DOUBLE PRECISION word.
————————————————————-
Array size = 2000000, Offset = 0
Total memory required = 45.8 MB.
Each test is run 10 times, but only
the *best* time for each is used.
————————————————————-
Printing one line per active thread….
————————————————————-
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 2913 microseconds.
   (= 2913 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
————————————————————-
WARNING — The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
————————————————————-
Function      Rate (MB/s)   Avg time     Min time     Max time
Copy:        8513.1123       0.0040       0.0038       0.0041
Scale:       8510.9529       0.0041       0.0038       0.0043
Add:         9052.8617       0.0055       0.0053       0.0057
Triad:       9158.2856       0.0055       0.0052       0.0057
————————————————————-
Solution Validates
————————————————————-

ヽ(;´Д`)ノ

追伸:

先日買ったFB-DIMM(PC-5300F/DDR2)なマシンでも試してみました。

————————————————————-
STREAM version $Revision: 5.9 $
————————————————————-
This system uses 8 bytes per DOUBLE PRECISION word.
————————————————————-
Array size = 2000000, Offset = 0
Total memory required = 45.8 MB.
Each test is run 10 times, but only
the *best* time for each is used.
————————————————————-
Printing one line per active thread….
————————————————————-
Your clock granularity/precision appears to be 1 microseconds.
Each test below will take on the order of 7155 microseconds.
   (= 7155 clock ticks)
Increase the size of the arrays if this shows that
you are not getting at least 20 clock ticks per test.
————————————————————-
WARNING — The above is only a rough guideline.
For best results, please be sure you know the
precision of your system timer.
————————————————————-
Function      Rate (MB/s)   Avg time     Min time     Max time
Copy:        3028.5833       0.0106       0.0106       0.0109
Scale:       3052.5536       0.0105       0.0105       0.0106
Add:         3410.7543       0.0142       0.0141       0.0151
Triad:       3423.6305       0.0141       0.0140       0.0144
————————————————————-
Solution Validates
————————————————————-

やっぱりDDR3はすごいなー。

Name of author

Name: admin

コメントを残す

メールアドレスが公開されることはありません。 * が付いている欄は必須項目です