Benchmark KVM MINI - webh - najtańszy VPS w ofercie webh.pl

Najtańszy serwer KVM MINI w ofercie firmy webh.pl.

Parametry serwera:

  • Model KVM MINI
  • CPU 1x
  • 2 GB ram
  • 10 Gib SSD
  • 1 Gbit sieć
  • 1 adres IPv4
  • DC Łódź
  • OS Centos 7
  • 500 GB transferu miesięcznie

UnixBench

#    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
#    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
#    #  # #  #  #    ##            #####   #####   # #  #  #       ######
#    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
#    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
 ####   #    #  #  #    #          #####   ######  #    #   ####   #    #
========================================================================
   BYTE UNIX Benchmarks (Version 5.1.3)

   System: sebastian.webh.me: GNU/Linux
   OS: GNU/Linux -- 3.10.0-514.6.1.el7.x86_64 -- #1 SMP Wed Jan 18 13:06:36 UTC 2017
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel Xeon E312xx (Sandy Bridge) (5187.5 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
   16:47:07 up 67 days,  3:11,  1 user,  load average: 0.82, 0.24, 0.12; runlevel 2017-02-12

------------------------------------------------------------------------
Benchmark Run: Thu Apr 20 2017 16:47:07 - 17:15:17
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       30254358.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3626.4 MWIPS (10.0 s, 7 samples)
Execl Throughput                               3388.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        905156.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          255272.0 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1604134.3 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1438555.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 215208.9 lps   (10.0 s, 7 samples)
Process Creation                               9075.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4542.7 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    603.2 lpm   (60.0 s, 2 samples)
System Call Overhead                        2163304.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   30254358.9   2592.5
Double-Precision Whetstone                       55.0       3626.4    659.3
Execl Throughput                                 43.0       3388.7    788.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     905156.3   2285.7
File Copy 256 bufsize 500 maxblocks            1655.0     255272.0   1542.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1604134.3   2765.7
Pipe Throughput                               12440.0    1438555.1   1156.4
Pipe-based Context Switching                   4000.0     215208.9    538.0
Process Creation                                126.0       9075.8    720.3
Shell Scripts (1 concurrent)                     42.4       4542.7   1071.4
Shell Scripts (8 concurrent)                      6.0        603.2   1005.3
System Call Overhead                          15000.0    2163304.3   1442.2
                                                                   ========
System Benchmarks Index Score                                        1202.5

Dysk

dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  

Widzimy tutaj że dysk jest bardzo wydajny. Dodatkowo pragnę zauważyć że wyniki utrzymują się przez cały okres wynajmu serwera.

[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.363637 s, 738 MB/s
[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.384941 s, 697 MB/s
[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.437506 s, 614 MB/s
[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.413606 s, 649 MB/s
[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.39436 s, 681 MB/s
[root@sebastian UnixBench]# dd bs=1M count=256 if=/dev/zero of=test conv=fdatasync  
256+0 records in
256+0 records out
268435456 bytes (268 MB) copied, 0.46398 s, 579 MB/s

Ram

sysbench --test=memory --memory-block-size=1M --memory-total-size=10G run

Również pamięć ram jest bardzo szybka.

Operations performed: 10240 ( 8036.62 ops/sec)

10240.00 MB transferred (8036.62 MB/sec)


Test execution summary:
    total time:                          1.2742s
    total number of events:              10240
    total time taken by event execution: 1.2722
    per-request statistics:
         min:                                  0.11ms
         avg:                                  0.12ms
         max:                                 27.41ms
         approx.  95 percentile:               0.13ms

Threads fairness:
    events (avg/stddev):           10240.0000/0.00
    execution time (avg/stddev):   1.2722/0.00

Internet

Bardzo przyzwoite wyniki, niskie pingi rekompensują 500GB miesięczny limit transferu.

speedtest-cli --share  

http://www.speedtest.net/result/6235677708.png

speedtest-cli --share --server 3574  

http://www.speedtest.net/result/6235678901.png

speedtest-cli --share --server 5134  

http://www.speedtest.net/result/6235679726.png

speedtest-cli --share --server 8123  

http://www.speedtest.net/result/6235680848.png

speedtest-cli --share --server 10799 

http://www.speedtest.net/result/6235681819.png

Podsumowanie

Niska ilość pamięci ram oraz powierzchni na przechowywanie danych rekompensuje wydajność podzespołów. Możliwości zastosowania takie VPSa jest wiele, od prostego hostingu po podpięcia środowiska do budowania i dostarczania oprogramowania jak np. gitlab-ci.