Sulley
/
grimboite
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

update scaleway vs the world comparison

This commit is contained in:
Fred 2018-08-27 21:34:31 +02:00
parent 1dc9cf931d
commit 046b687017
9 changed files with 388 additions and 510 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

510
articles/dev/2015-12-29-scaleway.rst
View File

@ -1,510 +0,0 @@
# Comparaison de quelques VPS/Baremetal
Utilisation de `Sysbench <https://wiki.gentoo.org/wiki/Sysbench>`_ pour les tests.
Les serveurs comparés sont les suivants:
* Un VPS Cloud chez OVH: 4Go de Ram, single-core et SSD de 20Go.
* Un C1 chez Scaleway: 2Go de RAM, quad-core ARMv7 et SSD de 50Go.
* Un Kimsufi KS-1: Atom @ 1.86Ghz (2 coeurs, 4 threads), 2Go de RAM et 500 de disque.
* Un ARM64 2GB chez Scaleway
* Un VPS x86-64 2GB de RAM et SSD Nvme de chez Scaleway (les nouveaux modèles de 2018 :-) )
Au niveau des tests, j'ai comparé les performances des CPU en utilisant `sysbench`. J'ai ensuite démarré un test de lecture/écriture complètement biaisé *via* la commande `dd`, comme expliqué [ici](http://www.stevefortuna.com/check-disk-speed-quickly-and-easily-in-linux/). Cette étape permet juste de simuler une écriture sur le disque, dans des conditions idéales.
## En résumé
sysbench --test=cpu --cpu-max-prime=20000 run
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
dd if=/dev/zero of=test bs=1048576 count=2048
dd if=test of=/dev/null bs=1048576
Disk speed
----------
pour simuler une écriture sur disque. Chez OVH, le serveur tourne à **514Mo/s**, tandis qu'il plafonne à **115Mo/s** sur le C1.
Idem pour la vitesse de lecture: chez OVH, on atteint les **2.3Go/s**, tandis que le C1 se voit aux alentours de **130Mo/s**.
Sur le KS-1, on tombe sur les valeurs suivantes: **143Mo/s** en écriture, **145Mo/s** en lecture.
Processeur
----------
On commence par installer ``sysbench`` et on le lance sur les trois machines ``sysbench --test=cpu --cpu-max-prime=20000 run``:
VPS Cloud
.. code-block:: shell
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 29.8127s
total number of events: 10000
total time taken by event execution: 29.8107
per-request statistics:
min: 2.66ms
avg: 2.98ms
max: 6.78ms
approx. 95 percentile: 3.33ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.8107/0.00
Sur le C1
.. code-block:: shell
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 686.8824s
total number of events: 10000
total time taken by event execution: 686.8633
per-request statistics:
min: 68.62ms
avg: 68.69ms
max: 76.86ms
approx. 95 percentile: 68.73ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 686.8633/0.00
Sur le KS-1
.. code-block:: shell
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 209.6545s
total number of events: 10000
total time taken by event execution: 209.6493
per-request statistics:
min: 20.94ms
avg: 20.96ms
max: 34.01ms
approx. 95 percentile: 20.98ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 209.6493/0.00
4 Threads
---------
Et la même chose en spécifiant 4 threads (pour profiter des 4 coeurs du processeur ARM): ``sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run``.
Sur VPS:
.. code-block:: shell
Test execution summary:
total time: 27.9756s
total number of events: 10000
total time taken by event execution: 111.8676
per-request statistics:
min: 2.50ms
avg: 11.19ms
max: 33.53ms
approx. 95 percentile: 15.28ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 27.9669/0.01
Et sur le C1:
.. code-block:: shell
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 171.3275s
total number of events: 10000
total time taken by event execution: 685.2335
per-request statistics:
min: 68.46ms
avg: 68.52ms
max: 109.33ms
approx. 95 percentile: 68.54ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 171.3084/0.01
Sur le KS-1:
.. code-block:: shell
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 86.2656s
total number of events: 10000
total time taken by event execution: 345.0025
per-request statistics:
min: 23.36ms
avg: 34.50ms
max: 78.49ms
approx. 95 percentile: 34.55ms
Threads fairness:
events (avg/stddev): 2500.0000/1.87
execution time (avg/stddev): 86.2506/0.01
START-1S
root@scw-f0a063:~# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 20.8311s
total number of events: 10000
total time taken by event execution: 20.8279
per-request statistics:
min: 2.07ms
avg: 2.08ms
max: 3.39ms
approx. 95 percentile: 2.10ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 20.8279/0.00
root@scw-f0a063:~# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 3.65569 s, 587 MB/s
root@scw-f0a063:~# dd if=test of=/dev/null bs=1048576
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 1.98512 s, 1.1 GB/s
root@scw-f0a063:~# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 10.4649s
total number of events: 10000
total time taken by event execution: 41.8144
per-request statistics:
min: 2.07ms
avg: 4.18ms
max: 17.05ms
approx. 95 percentile: 10.09ms
Threads fairness:
events (avg/stddev): 2500.0000/4.12
execution time (avg/stddev): 10.4536/0.01
Et le C1,
root@dagger:~# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 443.7068s
total number of events: 10000
total time taken by event execution: 443.6852
per-request statistics:
min: 44.24ms
avg: 44.37ms
max: 68.35ms
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 443.7068s
total number of events: 10000
total time taken by event execution: 443.6852
per-request statistics:
min: 44.24ms
avg: 44.37ms
max: 68.35ms
approx. 95 percentile: 44.57ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 443.6852/0.00
root@dagger:~# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 22.3317 s, 96.2 MB/s
root@dagger:~# dd if=test of=/dev/null bs=1048576
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 22.0091 s, 97.6 MB/s
root@dagger:~# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 112.1245s
total number of events: 10000
total time taken by event execution: 448.3735
per-request statistics:
min: 44.13ms
avg: 44.84ms
max: 178.64ms
approx. 95 percentile: 44.42ms
Threads fairness:
events (avg/stddev): 2500.0000/5.15
execution time (avg/stddev): 112.0934/0.02
Et l'ARM64 de base:
root@scw-6a9bee:~# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 29.9709s
total number of events: 10000
total time taken by event execution: 29.9635
per-request statistics:
min: 2.94ms
avg: 3.00ms
max: 3.24ms
approx. 95 percentile: 3.04ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.9635/0.00
root@scw-6a9bee:~# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 10.2453 s, 210 MB/s
root@scw-6a9bee:~# dd if=test of=/dev/null bs=1048576
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.54965 s, 387 MB/s
root@scw-6a9bee:~# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 7.5875s
total number of events: 10000
total time taken by event execution: 30.3245
per-request statistics:
min: 2.94ms
avg: 3.03ms
max: 8.97ms
approx. 95 percentile: 3.06ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 7.5811/0.00
## Raspberry Pi 3
sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 370.1484s
total number of events: 10000
total time taken by event execution: 370.1422
per-request statistics:
min: 36.89ms
avg: 37.01ms
max: 74.89ms
approx. 95 percentile: 37.08ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 370.1422/0.00
dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 191,147 s, 11,2 MB/s
dd if=test of=/dev/null bs=1048576
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 120,917 s, 17,8 MB/s
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
µsysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 93.0836s
total number of events: 10000
total time taken by event execution: 372.2606
per-request statistics:
min: 36.88ms
avg: 37.23ms
max: 146.01ms
approx. 95 percentile: 37.18ms
Threads fairness:
events (avg/stddev): 2500.0000/3.74
execution time (avg/stddev): 93.0652/0.01
## Raspberry Pi
root@pimousse:/home/pi# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 874.7016s
total number of events: 10000
total time taken by event execution: 874.6414
per-request statistics:
min: 86.93ms
avg: 87.46ms
max: 197.29ms
approx. 95 percentile: 88.08ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 874.6414/0.00
root@pimousse:/home/pi# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 876.7125s
total number of events: 10000
total time taken by event execution: 3506.2652
per-request statistics:
min: 193.68ms
avg: 350.63ms
max: 682.34ms
approx. 95 percentile: 372.27ms
Threads fairness:
events (avg/stddev): 2500.0000/0.00
execution time (avg/stddev): 876.5663/0.04
root@pimousse:/home/pi# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 298,227 s, 7,2 MB/s
root@pimousse:/home/pi# dd if=test of=/dev/null bs
dd: opérande « bs » non reconnu
Saisissez « dd --help » pour plus d'informations.
root@pimousse:/home/pi# dd if=test of=/dev/null bs=1048576
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 99,6178 s, 21,6 MB/s

52
articles/sys/2018-08-27-scaleway.md Normal file
View File

@ -0,0 +1,52 @@
# Comparaison de quelques VPS/Baremetal
Utilisation de `Sysbench <https://wiki.gentoo.org/wiki/Sysbench>`_ pour les tests.
Les serveurs comparés sont les suivants:
* Un VPS Cloud chez OVH: 4Go de Ram, single-core et SSD de 20Go.
* Un C1 chez Scaleway: 2Go de RAM, quad-core ARMv7 et SSD de 50Go.
* Un Kimsufi KS-1: Atom @ 1.86Ghz (2 coeurs, 4 threads), 2Go de RAM et 500 de disque.
* Un ARM64 2GB chez Scaleway
* Un VPS x86-64 2GB de RAM et SSD Nvme de chez Scaleway (les nouveaux modèles de 2018 :-) )
Au niveau des tests, j'ai comparé les performances des CPU en utilisant `sysbench`. J'ai ensuite démarré un test de lecture/écriture complètement biaisé *via* la commande `dd`, comme expliqué [ici](http://www.stevefortuna.com/check-disk-speed-quickly-and-easily-in-linux/). Cette étape permet juste de simuler une écriture sur le disque, dans des conditions idéales.
## En résumé
```shell
sysbench --test=cpu --cpu-max-prime=20000 run
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
dd if=/dev/zero of=test bs=1048576 count=2048
dd if=test of=/dev/null bs=1048576
```
| | VPS Cloud | C1 | KS-1 | Start-1S | ARM64-2G |
|-------------------|-------------|-----------|-----------|--------------|--------------|
| CPU | 29.8127s | 686.8824s | 209.6545s | **20.8311s** | 29.9709s |
| CPU multithreaded | 27.9756s | 171.3275s | 86.2656s | 10.4649s | **7.5875s** |
| Read | **2.3Go/s** | 97.6 MB/s | 145Mo/s | 1.1GB/s | 387 MB/s |
| Write | 514Mo/s | 96.2 MB/s | 143Mo/s | **587 MB/s** | 210 MB/s |
A titre de comparaison, un Rapsberry Pi 3 obtient respectivement les valeurs suivantes:
* CPU: 370.1484s
* CPU MT: 93.0836s
* Disk read: 17,8 MB/s
* Disk write: 11,2 MB/s
Et pour un Raspberry Pi B+ dans les mêmes conditions, on obtient ceci:
* CPU: 874.7016s
* CPU MT: 876.7125s
* Disk read: 21,6 MB/s
* Disk write: 7,2 MB/s
En conclusion, et en se basant sur ces résultats (théoriques), l'ARM64 semble être une option valable pour un petit serveur hébergeant quelques applications.
Surtout en prenant le prix en considération; à 3.67€ TVAC le petit processeur, 2Go de RAM et 50Go de SSD, c'est une offre qui semble plutôt sympa.
Si les besoins en espace de stockage sont grands, il vaudra peut-être mieux s'orienter vers un Kimsufi (pour peu qu'ils soient disponibles...).
Reste à vérifier quelques points:
* La disponibilité des noyaux pour ces archis un peu hétéroclites - le C1 tourne toujours avec un noyau en 4.3.5.
* Le support IPv6 qu'on dit bancal ou partiel (pas vérifié).

33
results/vps/ovh-ks1.bench Normal file
View File

@ -0,0 +1,33 @@
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 209.6545s
total number of events: 10000
total time taken by event execution: 209.6493
per-request statistics:
min: 20.94ms
avg: 20.96ms
max: 34.01ms
approx. 95 percentile: 20.98ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 209.6493/0.00
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 86.2656s
total number of events: 10000
total time taken by event execution: 345.0025
per-request statistics:
min: 23.36ms
avg: 34.50ms
max: 78.49ms
approx. 95 percentile: 34.55ms
Threads fairness:
events (avg/stddev): 2500.0000/1.87
execution time (avg/stddev): 86.2506/0.01

67
results/vps/rpi.bench Normal file
View File

@ -0,0 +1,67 @@
root@pimousse:/home/pi# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 874.7016s
total number of events: 10000
total time taken by event execution: 874.6414
per-request statistics:
min: 86.93ms
avg: 87.46ms
max: 197.29ms
approx. 95 percentile: 88.08ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 874.6414/0.00
root@pimousse:/home/pi# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 876.7125s
total number of events: 10000
total time taken by event execution: 3506.2652
per-request statistics:
min: 193.68ms
avg: 350.63ms
max: 682.34ms
approx. 95 percentile: 372.27ms
Threads fairness:
events (avg/stddev): 2500.0000/0.00
execution time (avg/stddev): 876.5663/0.04
root@pimousse:/home/pi# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 298,227 s, 7,2 MB/s
root@pimousse:/home/pi# dd if=test of=/dev/null bs
dd: opérande « bs » non reconnu
Saisissez « dd --help » pour plus d'informations.
root@pimousse:/home/pi# dd if=test of=/dev/null bs=1048576
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 99,6178 s, 21,6 MB/s

65
results/vps/rpi3.bench Normal file
View File

@ -0,0 +1,65 @@
sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 370.1484s
total number of events: 10000
total time taken by event execution: 370.1422
per-request statistics:
min: 36.89ms
avg: 37.01ms
max: 74.89ms
approx. 95 percentile: 37.08ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 370.1422/0.00
dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 191,147 s, 11,2 MB/s
dd if=test of=/dev/null bs=1048576
2048+0 enregistrements lus
2048+0 enregistrements écrits
2147483648 bytes (2,1 GB, 2,0 GiB) copied, 120,917 s, 17,8 MB/s
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
µsysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 93.0836s
total number of events: 10000
total time taken by event execution: 372.2606
per-request statistics:
min: 36.88ms
avg: 37.23ms
max: 146.01ms
approx. 95 percentile: 37.18ms
Threads fairness:
events (avg/stddev): 2500.0000/3.74
execution time (avg/stddev): 93.0652/0.01

63
results/vps/scaleway-arm64.bench Normal file
View File

@ -0,0 +1,63 @@
root@scw-6a9bee:~# sysbench --test=cpu --cpu-max-prime=20000 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 1
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 29.9709s
total number of events: 10000
total time taken by event execution: 29.9635
per-request statistics:
min: 2.94ms
avg: 3.00ms
max: 3.24ms
approx. 95 percentile: 3.04ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.9635/0.00
root@scw-6a9bee:~# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 10.2453 s, 210 MB/s
root@scw-6a9bee:~# dd if=test of=/dev/null bs=1048576
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 5.54965 s, 387 MB/s
root@scw-6a9bee:~# sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
sysbench 0.4.12: multi-threaded system evaluation benchmark
Running the test with following options:
Number of threads: 4
Doing CPU performance benchmark
Threads started!
Done.
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 7.5875s
total number of events: 10000
total time taken by event execution: 30.3245
per-request statistics:
min: 2.94ms
avg: 3.03ms
max: 8.97ms
approx. 95 percentile: 3.06ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 7.5811/0.00

43
results/vps/scaleway-c1.bench Normal file
View File

@ -0,0 +1,43 @@
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 686.8824s
total number of events: 10000
total time taken by event execution: 686.8633
per-request statistics:
min: 68.62ms
avg: 68.69ms
max: 76.86ms
approx. 95 percentile: 68.73ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 686.8633/0.00
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 171.3275s
total number of events: 10000
total time taken by event execution: 685.2335
per-request statistics:
min: 68.46ms
avg: 68.52ms
max: 109.33ms
approx. 95 percentile: 68.54ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 171.3084/0.01
root@dagger:~# dd if=/dev/zero of=test bs=1048576 count=2048
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 22.3317 s, 96.2 MB/s
root@dagger:~# dd if=test of=/dev/null bs=1048576
2048+0 records in
2048+0 records out
2147483648 bytes (2.1 GB, 2.0 GiB) copied, 22.0091 s, 97.6 MB/s

32
results/vps/scaleway-start-1-s.bench Normal file
View File

@ -0,0 +1,32 @@
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 20.8311s
total number of events: 10000
total time taken by event execution: 20.8279
per-request statistics:
min: 2.07ms
avg: 2.08ms
max: 3.39ms
approx. 95 percentile: 2.10ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 20.8279/0.00
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
Test execution summary:
total time: 10.4649s
total number of events: 10000
total time taken by event execution: 41.8144
per-request statistics:
min: 2.07ms
avg: 4.18ms
max: 17.05ms
approx. 95 percentile: 10.09ms
Threads fairness:
events (avg/stddev): 2500.0000/4.12
execution time (avg/stddev): 10.4536/0.01

33
results/vps/vps-cloud.bench Normal file
View File

@ -0,0 +1,33 @@
sysbench --test=cpu --cpu-max-prime=20000 run
Maximum prime number checked in CPU test: 20000
Test execution summary:
total time: 29.8127s
total number of events: 10000
total time taken by event execution: 29.8107
per-request statistics:
min: 2.66ms
avg: 2.98ms
max: 6.78ms
approx. 95 percentile: 3.33ms
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.8107/0.00
sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run
Test execution summary:
total time: 27.9756s
total number of events: 10000
total time taken by event execution: 111.8676
per-request statistics:
min: 2.50ms
avg: 11.19ms
max: 33.53ms
approx. 95 percentile: 15.28ms
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 27.9669/0.01