Hallo,
wärt Ihr so lieb mal aktuelle Benchmarks der von Euch gebuchten VPS und RS Produkte zu posten? Bitte mit Angabe der Uhrzeit! Danke!
Hallo,
wärt Ihr so lieb mal aktuelle Benchmarks der von Euch gebuchten VPS und RS Produkte zu posten? Bitte mit Angabe der Uhrzeit! Danke!
Irgendwelche Wünsche/Skripts? Ich persönlich bin Fan von yabs.sh bzw wget.racing/nench.sh
Hi!
Hier ein paar Benchmarks unter Ubuntu 18.04 mit sysbench 1.0.11:
VPS 200 G8 (etwa 9:00 mit Steal von ~21%)
CPU 1 Thread
root@vps200:~# sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 214.12
General statistics:
total time: 46.6994s
total number of events: 10000
Latency (ms):
min: 2.62
avg: 4.67
max: 42.65
95th percentile: 7.98
sum: 46653.33
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 46.6533/0.00
Alles anzeigen
VPS 200 G8 (etwa 9:00 mit Steal von ~21%)
RAM 1 Thread
root@vps200:~# sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 ( 7357.65 per second)
102400.00 MiB transferred (7357.65 MiB/sec)
General statistics:
total time: 13.9148s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.14
max: 32.03
95th percentile: 0.16
sum: 13830.10
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 13.8301/0.00
Alles anzeigen
VPS 200 G8 (etwa 9:00 mit Steal von ~21%)
DD Write
VPS 200 G8 (etwa 9:00 mit Steal von ~21%)
DD Read
RS 2000 SSD G8 (etwa 9:00, kein Steal)
CPU 1 Thread
root@rs2000:~$ sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 343.93
General statistics:
total time: 29.0734s
total number of events: 10000
Latency (ms):
min: 2.59
avg: 2.90
max: 4.06
95th percentile: 3.55
sum: 29045.45
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.0454/0.00
Alles anzeigen
RS 2000 SSD G8 (etwa 9:00, kein Steal)
CPU 4 Thread
root@rs2000:~$ sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 1351.42
General statistics:
total time: 7.3971s
total number of events: 10000
Latency (ms):
min: 2.59
avg: 2.95
max: 15.15
95th percentile: 3.55
sum: 29527.00
Threads fairness:
events (avg/stddev): 2500.0000/31.92
execution time (avg/stddev): 7.3817/0.01
Alles anzeigen
RS 2000 SSD G8 (etwa 9:00, kein Steal)
RAM 1 Thread
root@rs2000:~$ sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (11865.60 per second)
102400.00 MiB transferred (11865.60 MiB/sec)
General statistics:
total time: 8.6276s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.08
max: 7.53
95th percentile: 0.12
sum: 8547.96
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 8.5480/0.00
Alles anzeigen
RS 2000 SSD G8 (etwa 9:00, kein Steal)
RAM 4 Thread
root@rs2000:~$ sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.11 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (26344.86 per second)
102400.00 MiB transferred (26344.86 MiB/sec)
General statistics:
total time: 3.8847s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.15
max: 8.34
95th percentile: 0.18
sum: 15162.87
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 3.7907/0.08
Alles anzeigen
RS 2000 SSD G8 (etwa 9:00, kein Steal)
DD Write
RS 2000 SSD G8 (etwa 9:00, kein Steal)
DD Read
Sind halt künstliche Benchmarks, aber man sieht schön den Unterschied zwischen RS und VPS. Bei den DD Befehlen spielt wahrscheinlich ein Cache mit rein, aber ich habe alle Befehle extra mehrfach ausgeführt. Nur beim RS waren die DD Werte ziemlich inkonsistent (vielleicht zu kleine Werte?). Anfangs Schreiben so 600-800MB/s, dann konstant die 1,1GB/s. Beim lesen anfangs nur 700MB/s, dann 2,4GB/s und dann konstant über 4GB/s (nach ein paar Minuten wieder 2,6GB/s). Für zufälliges I/O Gedöns bin ich jetzt zu faul
Hier Benchmarks vom VPS Ostern M, jedoch mit ZFS :
CPU 1 Thread
sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 421.76
General statistics:
total time: 23.7091s
total number of events: 10000
Latency (ms):
min: 1.58
avg: 2.37
max: 52.38
95th percentile: 1.76
sum: 23700.84
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 23.7008/0.00
CPU 4 Thread
sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 1021.79
General statistics:
total time: 9.7849s
total number of events: 10000
Latency (ms):
min: 1.58
avg: 3.91
max: 250.68
95th percentile: 6.21
sum: 39123.53
Threads fairness:
events (avg/stddev): 2500.0000/133.85
execution time (avg/stddev): 9.7809/0.00
RAM 1 Thread
sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (18746.14 per second)
102400.00 MiB transferred (18746.14 MiB/sec)
General statistics:
total time: 5.4615s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.05
max: 13.43
95th percentile: 0.05
sum: 5445.53
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 5.4455/0.00
RAM 4 Thread
sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (15724.09 per second)
102400.00 MiB transferred (15724.09 MiB/sec)
General statistics:
total time: 6.5114s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.25
max: 57.40
95th percentile: 0.24
sum: 25466.11
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 6.3665/0.09
DD Write
dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 10.5927 s, 990 MB/s
DD Read
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 1.9778 s, 5.3 GB/s
Geekbench 5
Hier mal die Ergebnisse von meine VPS 2000 G9 (21:25 auf einem Samstag)
CPU 1 Thread
sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 567.58
General statistics:
total time: 17.6177s
total number of events: 10000
Latency (ms):
min: 1.58
avg: 1.76
max: 15.47
95th percentile: 1.61
sum: 17614.54
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 17.6145/0.00
CPU 4 Thread
sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 1344.13
General statistics:
total time: 7.4389s
total number of events: 10000
Latency (ms):
min: 1.58
avg: 2.97
max: 54.13
95th percentile: 1.64
sum: 29747.38
Threads fairness:
events (avg/stddev): 2500.0000/45.28
execution time (avg/stddev): 7.4368/0.00
RAM 1 Thread
sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (18761.19 per second)
102400.00 MiB transferred (18761.19 MiB/sec)
General statistics:
total time: 5.4572s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.05
max: 13.93
95th percentile: 0.05
sum: 5446.30
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 5.4463/0.00
RAM 4 Thread
sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (17334.97 per second)
102400.00 MiB transferred (17334.97 MiB/sec)
General statistics:
total time: 5.9063s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.23
max: 52.73
95th percentile: 0.21
sum: 23340.28
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 5.8351/0.04
DD Write
dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9,8 GiB) copied, 10,1049 s, 1,0 GB/s
DD Read
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9,8 GiB) copied, 13,5206 s, 776 MB/s
Messung2
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9,8 GiB) copied, 1,19978 s, 8,7 GB/s
root@v2202005120832117197:/srv/bots# dd if=test of=/dev/null bs=1M count=10000
Messung3
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9,8 GiB) copied, 0,924867 s, 11,3 GB/s
root@v2202005120832117197:/srv/bots# dd if=test of=/dev/null bs=1M count=10000
Meint Ihr auf einem RS (z.B. RS 2000 G8) hätte ich bessere CPU Werte? Kann man auch mit besserer Single Core Leistung rechnen wenn man einen VPS mit mehr Codes (z.B. VPS 4000 G9) wählt?
Wo kann man sich eigentlich den Steal anzeigen lassen?
Irgendwelche Wünsche/Skripts? Ich persönlich bin Fan von yabs.sh bzw wget.racing/nench.sh
Sysbench würde mir erstmal reichen
Die auch noch der Speedtest vom VPS 2000 G9 (Samstag, 22 Uhr)
Retrieving speedtest.net configuration...
Testing from netcup GmbH (xxx.xxx.xxx.xxx)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Thueringer Netkom GmbH (Weimar) [132.12 km]: 10.774 ms
Testing download speed................................................................................
Download: 800.10 Mbit/s
Testing upload speed......................................................................................................
Upload: 631.96 Mbit/s
Hier ein RS 2000 Plus SAS G8:
sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 334.52
General statistics:
total time: 29.8914s
total number of events: 10000
Latency (ms):
min: 2.51
avg: 2.99
max: 19.38
95th percentile: 3.62
sum: 29854.98
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.8550/0.00
sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 890.88
General statistics:
total time: 11.2229s
total number of events: 10000
Latency (ms):
min: 2.52
avg: 4.48
max: 23.91
95th percentile: 14.73
sum: 44834.89
Threads fairness:
events (avg/stddev): 2500.0000/82.23
execution time (avg/stddev): 11.2087/0.01
sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (10015.52 per second)
102400.00 MiB transferred (10015.52 MiB/sec)
General statistics:
total time: 10.2219s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.10
max: 16.31
95th percentile: 0.14
sum: 10129.98
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 10.1300/0.00
sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (16077.58 per second)
102400.00 MiB transferred (16077.58 MiB/sec)
General statistics:
total time: 6.3675s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.24
max: 27.73
95th percentile: 0.29
sum: 24313.92
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 6.0785/0.25
dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 23.3543 s, 449 MB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 87.7457 s, 120 MB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 3.2306 s, 3.2 GB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 2.61867 s, 4.0 GB/s
(Während des ersten DD-Read lief wahrscheinlich gerade rsnapshot)
Und Speedtest:
speedtest-cli
Retrieving speedtest.net configuration...
Testing from netcup GmbH (193.30.121.79)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Thueringer Netkom GmbH (Weimar) [132.12 km]: 11.306 ms
Testing download speed................................................................................
Download: 787.64 Mbit/s
Testing upload speed......................................................................................................
Upload: 4.16 Mbit/s
Der Upload ist offensichtlich falsch.
Uhrzeit kurz nach 22 Uhr (Samstag)
Danke für die Übersicht. Hast Du gerade gut was laufen auf der Kiste oder hatte die zu dem Zeitpunkt 0 Last? Bin überrascht, dass der RS mit gleichen Kernen und RAM deutlich weniger CPU Power bei Single Core und Multi-Core hat als das gleiche VPS.
Danke für die Übersicht. Hast Du gerade gut was laufen auf der Kiste oder hatte die zu dem Zeitpunkt 0 Last? Bin überrascht, dass der RS mit gleichen Kernen und RAM deutlich weniger CPU Power bei Single Core und Multi-Core hat als das gleiche VPS.
Apache mit ca 40 vHosts, Redis, MariaDB, drei Minecraft-Server, TS3, und ein paar Docker-Container. Die Load war bei ca. 2,5 (4 Kerne).
Außerdem ist der RS ja noch die 8te Generation.
Apache mit ca 40 vHosts, Redis, MariaDB, drei Minecraft-Server, TS3, und ein paar Docker-Container. Die Load war bei ca. 2,5 (4 Kerne).
Außerdem ist der RS ja noch die 8te Generation.
Also wenn man die Benchmarks Deines vollgepackten RS mit meinem VPS (nackt) vergleicht, könnte ich wohl beim RS mehr CPU Performance rausholen?
könnte ich wohl beim RS mehr CPU Performance rausholen
Genau darum geht es bei der RS Reihe. Um dedizierte Performance. Klar, es ist immernoch ein shared System, aber dir wird mehr Leistung zugesichert.
Du hast stärkere Hardware bei der RS Reihe und gleichzeitig weniger "Nachbarn".
Alles anzeigenHier ein RS 2000 Plus SAS G8:
sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 334.52
General statistics:
total time: 29.8914s
total number of events: 10000
Latency (ms):
min: 2.51
avg: 2.99
max: 19.38
95th percentile: 3.62
sum: 29854.98
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 29.8550/0.00
sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 890.88
General statistics:
total time: 11.2229s
total number of events: 10000
Latency (ms):
min: 2.52
avg: 4.48
max: 23.91
95th percentile: 14.73
sum: 44834.89
Threads fairness:
events (avg/stddev): 2500.0000/82.23
execution time (avg/stddev): 11.2087/0.01
sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (10015.52 per second)
102400.00 MiB transferred (10015.52 MiB/sec)
General statistics:
total time: 10.2219s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.10
max: 16.31
95th percentile: 0.14
sum: 10129.98
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 10.1300/0.00
sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (16077.58 per second)
102400.00 MiB transferred (16077.58 MiB/sec)
General statistics:
total time: 6.3675s
total number of events: 102400
Latency (ms):
min: 0.07
avg: 0.24
max: 27.73
95th percentile: 0.29
sum: 24313.92
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 6.0785/0.25
dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 23.3543 s, 449 MB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 87.7457 s, 120 MB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 3.2306 s, 3.2 GB/s
dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 2.61867 s, 4.0 GB/s
(Während des ersten DD-Read lief wahrscheinlich gerade rsnapshot)
Auch wenn es so früh wahrscheinlich nicht besonders aussagekräftig ist, hier mal als Vergleich die Werte eines RS2000 G9:
root@TEST:/home/jeff# sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 620.18
General statistics:
total time: 16.1234s
total number of events: 10000
Latency (ms):
min: 1.58
avg: 1.61
max: 2.32
95th percentile: 1.70
sum: 16118.06
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 16.1181/0.00
root@TEST:/home/jeff# sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 2462.40
General statistics:
total time: 4.0601s
total number of events: 10000
Latency (ms):
min: 1.57
avg: 1.62
max: 2.27
95th percentile: 1.73
sum: 16224.79
Threads fairness:
events (avg/stddev): 2500.0000/32.33
execution time (avg/stddev): 4.0562/0.00
root@TEST:/home/jeff# sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (20480.53 per second)
102400.00 MiB transferred (20480.53 MiB/sec)
General statistics:
total time: 4.9988s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.05
max: 0.18
95th percentile: 0.06
sum: 4984.37
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 4.9844/0.00
root@TEST:/home/jeff# sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (33147.15 per second)
102400.00 MiB transferred (33147.15 MiB/sec)
General statistics:
total time: 3.0883s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.12
max: 4.38
95th percentile: 0.19
sum: 12162.77
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 3.0407/0.03
root@TEST:/home/jeff# dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 6.69543 s, 1.6 GB/s
root@TEST:/home/jeff# dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 7.11509 s, 1.5 GB/s
root@TEST:/home/jeff# dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 1.14207 s, 9.2 GB/s
root@TEST:/home/jeff# dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 0.905321 s, 11.6 GB/s
root@TEST:/home/jeff# speedtest-cli
Retrieving speedtest.net configuration...
Testing from netcup GmbH (45.83.105.8)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Keyweb AG (Erfurt) [112.91 km]: 4.099 ms
Testing download speed................................................................................
Download: 903.78 Mbit/s
Testing upload speed......................................................................................................
Upload: 4.18 Mbit/s
RS8000 G9 :
root@deadpool:/home/caspar# sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 625.49
General statistics:
total time: 15.9865s
total number of events: 10000
Latency (ms):
min: 1.57
avg: 1.60
max: 5.68
95th percentile: 1.61
sum: 15980.45
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 15.9804/0.00
root@deadpool:/home/caspar# sysbench cpu --threads=4 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 2484.99
General statistics:
total time: 4.0232s
total number of events: 10000
Latency (ms):
min: 1.57
avg: 1.61
max: 5.93
95th percentile: 1.64
sum: 16080.83
Threads fairness:
events (avg/stddev): 2500.0000/6.28
execution time (avg/stddev): 4.0202/0.00
root@deadpool:/home/caspar# sysbench memory --threads=1 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (20425.14 per second)
102400.00 MiB transferred (20425.14 MiB/sec)
General statistics:
total time: 5.0125s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.05
max: 0.30
95th percentile: 0.06
sum: 4997.03
Threads fairness:
events (avg/stddev): 102400.0000/0.00
execution time (avg/stddev): 4.9970/0.00
root@deadpool:/home/caspar# sysbench memory --threads=4 --memory-block-size=1M --memory-total-size=100G --time=0 --events=10000 run
sysbench 1.0.20 (using bundled LuaJIT 2.1.0-beta2)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Running memory speed test with the following options:
block size: 1024KiB
total size: 102400MiB
operation: write
scope: global
Initializing worker threads...
Threads started!
Total operations: 102400 (26093.49 per second)
102400.00 MiB transferred (26093.49 MiB/sec)
General statistics:
total time: 3.9234s
total number of events: 102400
Latency (ms):
min: 0.05
avg: 0.15
max: 8.33
95th percentile: 0.29
sum: 15273.48
Threads fairness:
events (avg/stddev): 25600.0000/0.00
execution time (avg/stddev): 3.8184/0.11
root@deadpool:/home/caspar# dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 11.5052 s, 911 MB/s
root@deadpool:/home/caspar# dd if=/dev/zero of=test bs=1M count=10000 oflag=direct
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 11.5052 s, 911 MB/s
root@deadpool:/home/caspar# #
root@deadpool:/home/caspar#
root@deadpool:/home/caspar#
root@deadpool:/home/caspar#
root@deadpool:/home/caspar#
root@deadpool:/home/caspar#
root@deadpool:/home/caspar# dd if=test of=/dev/null bs=1M count=10000
10000+0 records in
10000+0 records out
10485760000 bytes (10 GB, 9.8 GiB) copied, 15.2275 s, 689 MB/s
root@deadpool:/home/caspar# speedtest-cli
Retrieving speedtest.net configuration...
Testing from netcup GmbH (45.83.104.221)...
Retrieving speedtest.net server list...
Selecting best server based on ping...
Hosted by Keyweb AG (Erfurt) [112.91 km]: 4.737 ms
Testing download speed................................................................................
Download: 1773.97 Mbit/s
Testing upload speed......................................................................................................
Upload: 4.17 Mbit/s
sysbench cpu --threads=1 --cpu-max-prime=20000 --time=0 --events=10000 run
sysbench 1.0.18 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 355.89
General statistics:
total time: 28.0963s
total number of events: 10000
Latency (ms):
min: 2.61
avg: 2.81
max: 14.27
95th percentile: 3.55
sum: 28082.79
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 28.0828/0.00
Kann mir bitte jemand erklären, warum die G9 CPUs (AMD EPYC 7702) bei den Benchmarks viel besser abschneiden als meine G8 CPU (Intel Xeon Gold 6230)? Laut https://www.cpubenchmark.net/c…Xeon-Gold-6230/3719vs3468 sind die Single Thread Ratings:
AMD EPYC 7702: 2067
Intel Xeon Gold 6230: 2362
Demnach wäre ein umstieg von G8 auf G9 ein Downgrade (bzgl. CPU Leistung). Die Benchmarks hingegen sagen etwas anderes. Warum ist das so?
Benchmarks sind nur untereinander vergleichbar wenn man genau das gleiche Programm nimmt.
Und dann messen diese nur die Geschwindigkeit von genau sich selbst
Das Problem ist: Der eine Prozessor ist z.B. in Kommazahlenberechnung VIEL schneller als der andere. Dafür ist der andere in Ganzzahlenberechnung viel schneller.
Welcher ist nun der schnellere? Man weiß es nicht...
Es kommt drauf an ob dein Programm (oder dein Benchmark) viele Ganzzahlberechnungen oder viele Kommazahlberechnungen macht.
Das Problem ist hier natürlich vereinfacht dargestellt, denn es gibt quasi hunderte solcher Spezialfunktionen. Und Caches, und die haben Zugriffszeiten, und, und, und...
Afaik sagt man eigentlich dass AMD mittlerweile besser ist als Intel...
Hier mal ein bisschen single threaded Kommazahlenberechnung zwischen G8:
perf stat sh -c 'echo "scale=5000; a(1)*4" | bc -l > /dev/null'
Performance counter stats for 'sh -c echo "scale=5000; a(1)*4" | bc -l > /dev/null':
21330.854074 task-clock (msec) # 1.000 CPUs utilized
19 context-switches # 0.001 K/sec
0 cpu-migrations # 0.000 K/sec
244 page-faults # 0.011 K/sec
62,782,990,547 cycles # 2.943 GHz
133,221,330,529 instructions # 2.12 insn per cycle
24,948,483,329 branches # 1169.596 M/sec
1,035,290,960 branch-misses # 4.15% of all branches
21.335457792 seconds time elapsed
Alles anzeigen
und G9:
Performance counter stats for 'sh -c echo "scale=5000; a(1)*4" | bc -l > /dev/null':
16.845,88 msec task-clock # 1,000 CPUs utilized
40 context-switches # 0,002 K/sec
0 cpu-migrations # 0,000 K/sec
250 page-faults # 0,015 K/sec
55.293.138.177 cycles # 3,282 GHz
1.006.607.785 stalled-cycles-frontend # 1,82% frontend cycles idle
11.578.898.880 stalled-cycles-backend # 20,94% backend cycles idle
129.659.377.689 instructions # 2,34 insn per cycle
# 0,09 stalled cycles per insn
24.707.143.036 branches # 1466,658 M/sec
0 branch-misses # 0,00% of all branches
16,849947441 seconds time elapsed
16,842666000 seconds user
0,003999000 seconds sys
Alles anzeigen
PS: Meine G8 haben noch Intel Xeon Gold 6140, nicht 6230...
VPS 2000 G9 - Test am Mittwoch um 13:15 - Läuft eigentlich nichts drauf, was CPU frisst.
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-02-10 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Mi 3. Jun 13:00:28 CEST 2020
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 4 @ 1996.249 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 15Gi
Swap : 0B
Disk : 315G
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4kb (IOPS) | 64kb (IOPS) ------ | --- ---- | ---- ----
Read | 162.50 MB/s (40.6k) | 1.82 GB/s (28.4k)
Write | 162.93 MB/s (40.7k) | 1.83 GB/s (28.6k)
Total | 325.44 MB/s (81.3k) | 3.65 GB/s (57.0k) | |
Block Size | 512kb (IOPS) | 1mb (IOPS) ------ | ----- ---- | --- ----
Read | 3.44 GB/s (6.7k) | 3.51 GB/s (3.4k)
Write | 3.62 GB/s (7.0k) | 3.75 GB/s (3.6k)
Total | 7.07 GB/s (13.8k) | 7.26 GB/s (7.0k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv S peed | | | Bouygues Telecom | Paris, FR (10G) | 977 Mbits/sec | 951 Mb its/sec
Online.net | Paris, FR (10G) | 947 Mbits/sec | 774 Mb its/sec
WorldStream | The Netherlands (10G) | 977 Mbits/sec | 908 Mb its/sec
wilhelm.tel | Hamburg, DE (10G) | 987 Mbits/sec | 939 Mb its/sec
Biznet | Bogor, Indonesia (1G) | 296 Mbits/sec | 73.5 M bits/sec
Hostkey | Moscow, RU (1G) | 912 Mbits/sec | 843 Mb its/sec
Velocity Online | Tallahassee, FL, US (10G) | 445 Mbits/sec | 818 Mb its/sec
Performing IPv4 iperf3 send test to Airstream Communications (Attempt #1 of 10). Performing IPv4 iperf3 recv test from Airstream Communications (Attempt #1 of 10 Airstream Communications | Eau Claire, WI, US (10G) | 227 Mbits/sec | 188 Mb its/sec
Hurricane Electric | Fremont, CA, US (10G) | busy | busy
iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | | |
Bouygues Telecom | Paris, FR (10G) | 960 Mbits/sec | 842 Mbits/sec
Online.net | Paris, FR (10G) | 786 Mbits/sec | 721 Mbits/sec
WorldStream | The Netherlands (10G) | 909 Mbits/sec | 817 Mbits/sec
wilhelm.tel | Hamburg, DE (10G) | 988 Mbits/sec | 887 Mbits/sec
Airstream Communications | Eau Claire, WI, US (10G) | 208 Mbits/sec | 290 Mbits/sec
Hurricane Electric | Fremont, CA, US (10G) | busy | busy
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value |
Single Core | 847
Multi Core | 2045
Full Test | https://browser.geekbench.com/v5/cpu/2393440
Alles anzeigen
VPS 200 G8 - 08 vor der Neuinstallation (auf Jessi)
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-09-21 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fr 16. Okt 21:07:05 CEST 2020
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2593.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 2.0G
Swap : 3.7G
Disk : 6.1G
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 8.86 MB/s (2.2k) | 57.64 MB/s (900)
Write | 8.90 MB/s (2.2k) | 58.17 MB/s (908)
Total | 17.76 MB/s (4.4k) | 115.81 MB/s (1.8k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 152.67 MB/s (298) | 216.41 MB/s (211)
Write | 160.78 MB/s (314) | 230.83 MB/s (225)
Total | 313.45 MB/s (612) | 447.24 MB/s (436)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 665 Mbits/sec | 331 Mbits/sec
Online.net | Paris, FR (10G) | 800 Mbits/sec | 408 Mbits/sec
WorldStream | The Netherlands (10G) | 647 Mbits/sec | 439 Mbits/sec
Wifx | Zurich, CH (10G) | 734 Mbits/sec | busy
Biznet | Jakarta, Indonesia (1G) | 535 Mbits/sec | 54.8 Mbits/sec
Clouvider | NYC, NY, US (10G) | 695 Mbits/sec | 341 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 749 Mbits/sec | 182 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 683 Mbits/sec | 24.0 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | busy | busy
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 302
Multi Core | 307
Full Test | https://browser.geekbench.com/v5/cpu/4205671
Alles anzeigen
Nach der Neuinstallation (buster)
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-09-21 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fri 16 Oct 2020 09:29:39 PM CEST
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2593.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 1.9Gi
Swap : 0B
Disk : 20G
Generating fio test file...^C
** Aborting YABS. Cleaning up files...
root@TOR-Piratenpartei08:~# passwd
New password:
Retype new password:
passwd: password updated successfully
root@TOR-Piratenpartei08:~# curl -sL yabs.sh | bash
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-09-21 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fri 16 Oct 2020 09:29:54 PM CEST
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2593.748 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 1.9Gi
Swap : 0B
Disk : 20G
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 8.45 MB/s (2.1k) | 57.44 MB/s (897)
Write | 8.50 MB/s (2.1k) | 58.00 MB/s (906)
Total | 16.95 MB/s (4.2k) | 115.44 MB/s (1.8k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 159.55 MB/s (311) | 200.47 MB/s (195)
Write | 168.02 MB/s (328) | 213.82 MB/s (208)
Total | 327.57 MB/s (639) | 414.29 MB/s (403)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | 631 Mbits/sec | 359 Mbits/sec
Online.net | Paris, FR (10G) | 669 Mbits/sec | 415 Mbits/sec
WorldStream | The Netherlands (10G) | 574 Mbits/sec | 434 Mbits/sec
Wifx | Zurich, CH (10G) | busy | busy
Biznet | Jakarta, Indonesia (1G) | 571 Mbits/sec | 133 Mbits/sec
Clouvider | NYC, NY, US (10G) | 799 Mbits/sec | 234 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 721 Mbits/sec | 189 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 755 Mbits/sec | 67.4 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | busy | busy
iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed
| | |
Clouvider | London, UK (10G) | busy | busy
Online.net | Paris, FR (10G) | busy | busy
WorldStream | The Netherlands (10G) | 693 Mbits/sec | 406 Mbits/sec
Wifx | Zurich, CH (10G) | 763 Mbits/sec | busy
Clouvider | NYC, NY, US (10G) | 792 Mbits/sec | 150 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 751 Mbits/sec | 249 Mbits/sec
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 310
Multi Core | 313
Full Test | https://browser.geekbench.com/v5/cpu/4206140
Alles anzeigen
Maximal 10.000 Zeichen pro Nachricht... dann halt zwei.
VPS200 G8 - 09 vor der Neuinstallation (jessie)
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-09-21 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fri Oct 16 21:20:28 CEST 2020
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2399.996 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 2.0G
Swap : 3.8G
Disk : 16G
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS) ------ | --- ---- | ---- ----
Read | 41.30 MB/s (10.3k) | 470.13 MB/s (7.3k)
Write | 41.38 MB/s (10.3k) | 472.61 MB/s (7.3k)
Total | 82.68 MB/s (20.6k) | 942.75 MB/s (14.7k) | |
Block Size | 512k (IOPS) | 1m (IOPS) ------ | --- ---- | ---- ----
Read | 904.72 MB/s (1.7k) | 878.22 MB/s (857)
Write | 952.80 MB/s (1.8k) | 936.70 MB/s (914)
Total | 1.85 GB/s (3.6k) | 1.81 GB/s (1.7k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | | |
Clouvider | London, UK (10G) | 561 Mbits/sec | 457 Mbits/sec
Online.net | Paris, FR (10G) | 537 Mbits/sec | 389 Mbits/sec
WorldStream | The Netherlands (10G) | 441 Mbits/sec | 509 Mbits/sec
Wifx | Zurich, CH (10G) | 516 Mbits/sec | busy
Biznet | Jakarta, Indonesia (1G) | busy | 14.2 Mbits/sec
Clouvider | NYC, NY, US (10G) | 451 Mbits/sec | 195 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 515 Mbits/sec | 155 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 452 Mbits/sec | 137 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | busy | busy
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value |
Single Core | 420
Multi Core | 401
Full Test | https://browser.geekbench.com/v5/cpu/4205869
root@v22018122160778285:~# lsb_release -a
No LSB modules are available.
Distributor ID: Debian
Description: Debian GNU/Linux 8.11 (jessie)
Release: 8.11
Codename: jessie
Alles anzeigen
Nach der Reinstallation (buster)
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2020-09-21 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Fri 16 Oct 2020 09:44:45 PM CEST
Basic System Information:
---------------------------------
Processor : QEMU Virtual CPU version 2.5+
CPU cores : 1 @ 2399.996 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 1.9Gi
Swap : 0B
Disk : 20G
fio Disk Speed Tests (Mixed R/W 50/50):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS) ------ | --- ---- | ---- ----
Read | 38.77 MB/s (9.6k) | 509.04 MB/s (7.9k)
Write | 38.86 MB/s (9.7k) | 511.71 MB/s (7.9k)
Total | 77.64 MB/s (19.4k) | 1.02 GB/s (15.9k) | |
Block Size | 512k (IOPS) | 1m (IOPS) ------ | --- ---- | ---- ----
Read | 1.20 GB/s (2.3k) | 1.17 GB/s (1.1k)
Write | 1.27 GB/s (2.4k) | 1.25 GB/s (1.2k)
Total | 2.47 GB/s (4.8k) | 2.43 GB/s (2.3k)
iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | | |
Clouvider | London, UK (10G) | 533 Mbits/sec | 355 Mbits/sec
Online.net | Paris, FR (10G) | 672 Mbits/sec | 368 Mbits/sec
WorldStream | The Netherlands (10G) | 476 Mbits/sec | 446 Mbits/sec
Wifx | Zurich, CH (10G) | busy | busy
Biznet | Jakarta, Indonesia (1G) | 602 Mbits/sec | 100 Mbits/sec
Clouvider | NYC, NY, US (10G) | 737 Mbits/sec | 293 Mbits/sec
Velocity Online | Tallahassee, FL, US (10G) | 699 Mbits/sec | 233 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 653 Mbits/sec | 325 Mbits/sec
Iveloz Telecom | Sao Paulo, BR (2G) | busy | busy
iperf3 Network Speed Tests (IPv6):
---------------------------------
Provider | Location (Link) | Send Speed | Recv Speed | | |
Clouvider | London, UK (10G) | 672 Mbits/sec | 323 Mbits/sec
Online.net | Paris, FR (10G) | 705 Mbits/sec | 319 Mbits/sec
WorldStream | The Netherlands (10G) | 663 Mbits/sec | 383 Mbits/sec
Wifx | Zurich, CH (10G) | 695 Mbits/sec | busy
Clouvider | NYC, NY, US (10G) | 713 Mbits/sec | 150 Mbits/sec
Clouvider | Los Angeles, CA, US (10G) | 678 Mbits/sec | 211 Mbits/sec
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value |
Single Core | 395
Multi Core | 386
Full Test | https://browser.geekbench.com/v5/cpu/4206182
Alles anzeigen
In den nächsten Tagen werde ich noch einige andere Server nachtragen.
Erstaunlich sind doch die sehr unterschiedlichen R/W-Werte...
Neu erstellte Beiträge unterliegen der Moderation und werden erst sichtbar, wenn sie durch einen Moderator geprüft und freigeschaltet wurden.
Die letzte Antwort auf dieses Thema liegt mehr als 365 Tage zurück. Das Thema ist womöglich bereits veraltet. Bitte erstellen Sie ggf. ein neues Thema.