[Linux] 오라클 클라우드 VM.Standard.A1.Flex sysbench

오라클 클라우드 VM 인스턴스에서 sysbench 한번 돌려봤습니다.

구성: VM.Standard.A1.Flex

OCPU 개수: 4

네트워크 대역폭(Gbps): 4

메모리(GB): 24

 

[root@rocky ~]# sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 run
sysbench 1.0.20 (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:  1260.57

General statistics:
    total time:                          7.9314s
    total number of events:              10000

Latency (ms):
         min:                                    0.77
         avg:                                    0.79
         max:                                    1.01
         95th percentile:                        0.81
         sum:                                 7929.53

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   7.9295/0.00

Thread 1개로 돌린 결과입니다.
1260.57

Thread 4개로 돌린 결과입니다.
5117.90

[root@rocky ~]# sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 --num-threads=4 run
WARNING: --num-threads is deprecated, use --threads instead
sysbench 1.0.20 (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:  5117.90

General statistics:
    total time:                          1.9524s
    total number of events:              10000

Latency (ms):
         min:                                    0.77
         avg:                                    0.78
         max:                                    1.07
         95th percentile:                        0.80
         sum:                                 7800.25

Threads fairness:
    events (avg/stddev):           2500.0000/23.27
    execution time (avg/stddev):   1.9501/0.00

비교 대상으로 인텔 i5-6400 CPU 환경에서 Proxmox 에서도 실행해봤습니다.

root@proxmox:~# sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 run
sysbench 1.0.20 (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:   422.36

General statistics:
    total time:                          23.6749s
    total number of events:              10000

Latency (ms):
         min:                                    2.36
         avg:                                    2.37
         max:                                    3.15
         95th percentile:                        2.39
         sum:                                23673.21

Threads fairness:
    events (avg/stddev):           10000.0000/0.00
    execution time (avg/stddev):   23.6732/0.00

root@proxmox:~# sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 --num-threads=4 run
WARNING: --num-threads is deprecated, use --threads instead
sysbench 1.0.20 (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:  1585.18

General statistics:
    total time:                          6.3068s
    total number of events:              10000

Latency (ms):
         min:                                    2.51
         avg:                                    2.52
         max:                                    8.32
         95th percentile:                        2.52
         sum:                                25221.86

Threads fairness:
    events (avg/stddev):           2500.0000/4.30
    execution time (avg/stddev):   6.3055/0.00

Thread 1개 : 422.36
Thread 4개 : 1585.18

좀 의외의 결과인 것 같네요~