Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

sysbench benchmark results for $7 and under DEDICATED servers

I'm interested to see how these low end servers compare on CPU benchmarks, I was surprised when my Raspberry pi 4 home server has a better CPU benchmark than KS-2E with an Atom N2800. Make sure to put ``` at the start and and of your results so they format properly.

Benchmark command (install sysbench package first): sysbench --test=cpu --cpu-max-prime=20000 --num-threads=4 run

Kimsufi KS-2E (Intel(R) Atom(TM) CPU N2800 @ 1.86GHz):

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:                          90.1258s
    total number of events:              10000
    total time taken by event execution: 360.4430
    per-request statistics:
         min:                                 23.51ms
         avg:                                 36.04ms
         max:                                154.11ms
         approx.  95 percentile:              58.16ms

Threads fairness:
    events (avg/stddev):           2500.0000/40.84
    execution time (avg/stddev):   90.1107/0.01

Raspberry Pi 4 B 2GB (Broadcom BCM2711B0 quad-core A72 (ARMv8-A) 64-bit @ 1.5GHz):

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:                          62.6571s
    total number of events:              10000
    total time taken by event execution: 250.5635
    per-request statistics:
         min:                                 24.42ms
         avg:                                 25.06ms
         max:                                 72.70ms
         approx.  95 percentile:              25.08ms

Threads fairness:
    events (avg/stddev):           2500.0000/2.92
    execution time (avg/stddev):   62.6409/0.01

Comments

  • SoYouStart ARM box (I only did two threads since there are only two cores):

    Test execution summary:
        total time:                          288.6728s
        total number of events:              10000
        total time taken by event execution: 577.2823
        per-request statistics:
             min:                                 54.14ms
             avg:                                 57.73ms
             max:                                143.43ms
             approx.  95 percentile:              75.39ms
    
    Threads fairness:
        events (avg/stddev):           5000.0000/28.00
        execution time (avg/stddev):   288.6412/0.02
    
    
    Thanked by 1Creativity
  • hmm, tried to install sysbench but the deps seem broken on debian 10.2..

  • I think the hardware performance is one part of the story. You run into fewer issues administering the server with Intel CPUs in general compared to ARM.

    Thanked by 1Creativity
  • poisson said: administering the server

    Agree. ARM still need more time to become more popular in the server space.

Sign In or Register to comment.