Benchmark DigitalOcean Singapore

Karena hari ini hari terakhir saldo/credit bonus DigitalOcean saya berlaku – tidak banyak sih cuma sisa sekitar 10$an lah – daripada mubazir saya akan mendata performa masing – masing jenis Droplet dengan datacenter di Singapore karena relevan sekali dengan pengguna yang ada di Indonesia. Yah… semoga ada manfaatnya kalau pas milih – milih VPS yang bagus. 🙂

Supaya standar, disini saya menggunakan CentOS 7.2 dan telah melakukan update ke versi terbarunya saat ini sebelum melakukan benchmark. Jadi masih kosongan ya isinya, belum diinstall apa – apa kecuali default template CentOS dari DigitalOcean. Dan ini bukan review lho ya. 🙂

Droplet 512MB SG

Langsung saja ini adalah hasil tes bandwidth dan I/O memanfaatkan Bench.sh:

----------------------------------------------------------------------
CPU model            : Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz
Number of cores      : 1
CPU frequency        : 2399.998 MHz
Total amount of ram  : 489 MB
Total amount of swap : 0 MB
System uptime        : 0days, 0:9:11
OS                   : CentOS Linux 7
Arch                 : x86_64 (64 Bit)
Kernel               : 3.10.0-327.10.1.el7.x86_64
----------------------------------------------------------------------
Node Name			IPv4 address		Download Speed
CacheFly			205.234.175.175		31.0MB/s
Linode, Tokyo, JP		106.187.96.148		20.0MB/s
Linode, Singapore, SG		139.162.23.4		103MB/s
Linode, London, UK		176.58.107.39		9.37MB/s
Linode, Frankfurt, DE		139.162.130.8		8.09MB/s
Linode, Fremont, CA		50.116.14.9		9.03MB/s
Softlayer, Dallas, TX		173.192.68.18		9.25MB/s
Softlayer, Seattle, WA		67.228.112.250		14.5MB/s
Softlayer, Frankfurt, DE	159.122.69.4		10.8MB/s
Softlayer, Singapore, SG	119.81.28.170		71.6MB/s
Softlayer, HongKong, CN		119.81.130.170		47.6MB/s
----------------------------------------------------------------------
I/O speed(1st run) : 436 MB/s
I/O speed(2nd run) : 544 MB/s
I/O speed(3rd run) : 627 MB/s
Average I/O: 535.667 MB/s

Kemudian dilanjutkan dengan hasil dari UnixBench

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com


1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

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

   System: centos-512mb-sgp1-01: GNU/Linux
   OS: GNU/Linux -- 3.10.0-327.10.1.el7.x86_64 -- #1 SMP Tue Feb 16 17:03:50 UTC 2016
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz (4800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   14:46:13 up 15 min,  1 user,  load average: 0.49, 0.28, 0.19; runlevel 3

------------------------------------------------------------------------
Benchmark Run: Fri Apr 29 2016 14:46:13 - 15:14:13
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       28460322.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3649.4 MWIPS (8.9 s, 7 samples)
Execl Throughput                               3354.1 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        731404.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          212570.2 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1436474.2 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1430750.9 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 255961.2 lps   (10.0 s, 7 samples)
Process Creation                               9825.9 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   4427.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    589.5 lpm   (60.0 s, 2 samples)
System Call Overhead                        2056859.9 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   28460322.5   2438.8
Double-Precision Whetstone                       55.0       3649.4    663.5
Execl Throughput                                 43.0       3354.1    780.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     731404.4   1847.0
File Copy 256 bufsize 500 maxblocks            1655.0     212570.2   1284.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1436474.2   2476.7
Pipe Throughput                               12440.0    1430750.9   1150.1
Pipe-based Context Switching                   4000.0     255961.2    639.9
Process Creation                                126.0       9825.9    779.8
Shell Scripts (1 concurrent)                     42.4       4427.2   1044.2
Shell Scripts (8 concurrent)                      6.0        589.5    982.5
System Call Overhead                          15000.0    2056859.9   1371.2
                                                                   ========
System Benchmarks Index Score                                        1160.8

Sekalian menambahkan dari Geekbench, cuma link saja hasilnya: http://browser.primatelabs.com/geekbench3/6294295

Droplet 1GB SG

Sama, diawali dari Bench.sh:

----------------------------------------------------------------------
CPU model            : Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz
Number of cores      : 1
CPU frequency        : 2399.998 MHz
Total amount of ram  : 993 MB
Total amount of swap : 0 MB
System uptime        : 0days, 0:8:45
OS                   : CentOS Linux 7
Arch                 : x86_64 (64 Bit)
Kernel               : 3.10.0-327.10.1.el7.x86_64
----------------------------------------------------------------------
Node Name			IPv4 address		Download Speed
CacheFly			205.234.175.175		31.3MB/s
Linode, Tokyo, JP		106.187.96.148		16.6MB/s
Linode, Singapore, SG		139.162.23.4		102MB/s
Linode, London, UK		176.58.107.39		10.4MB/s
Linode, Frankfurt, DE		139.162.130.8		5.23MB/s
Linode, Fremont, CA		50.116.14.9		5.82MB/s
Softlayer, Dallas, TX		173.192.68.18		8.76MB/s
Softlayer, Seattle, WA		67.228.112.250		13.4MB/s
Softlayer, Frankfurt, DE	159.122.69.4		10.5MB/s
Softlayer, Singapore, SG	119.81.28.170		63.9MB/s
Softlayer, HongKong, CN		119.81.130.170		46.2MB/s
----------------------------------------------------------------------
I/O speed(1st run) : 362 MB/s
I/O speed(2nd run) : 424 MB/s
I/O speed(3rd run) : 640 MB/s
Average I/O: 475.333 MB/s

Disusul oleh UnixBench:

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com


1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

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

   System: centos-1gb-sgp1-01: GNU/Linux
   OS: GNU/Linux -- 3.10.0-327.10.1.el7.x86_64 -- #1 SMP Tue Feb 16 17:03:50 UTC 2016
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz (4800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   14:46:22 up 15 min,  1 user,  load average: 0.37, 0.27, 0.20; runlevel 3

------------------------------------------------------------------------
Benchmark Run: Fri Apr 29 2016 14:46:22 - 15:14:09
1 CPU in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       21160236.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2798.4 MWIPS (8.3 s, 7 samples)
Execl Throughput                               2385.8 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        510835.6 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          145214.1 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1052013.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1084853.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 185328.3 lps   (10.0 s, 7 samples)
Process Creation                               7048.5 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3184.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    430.9 lpm   (60.1 s, 2 samples)
System Call Overhead                        1508884.2 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   21160236.3   1813.2
Double-Precision Whetstone                       55.0       2798.4    508.8
Execl Throughput                                 43.0       2385.8    554.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     510835.6   1290.0
File Copy 256 bufsize 500 maxblocks            1655.0     145214.1    877.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1052013.0   1813.8
Pipe Throughput                               12440.0    1084853.0    872.1
Pipe-based Context Switching                   4000.0     185328.3    463.3
Process Creation                                126.0       7048.5    559.4
Shell Scripts (1 concurrent)                     42.4       3184.9    751.2
Shell Scripts (8 concurrent)                      6.0        430.9    718.2
System Call Overhead                          15000.0    1508884.2   1005.9
                                                                   ========
System Benchmarks Index Score                                         843.1

Dan juga Geekbench: http://browser.primatelabs.com/geekbench3/6294339

Droplet 2GB SG

Hasil tes Bench.sh:

----------------------------------------------------------------------
Node Name			IPv4 address		Download Speed
CacheFly			205.234.175.175		31.2MB/s
Linode, Tokyo, JP		106.187.96.148		19.0MB/s
Linode, Singapore, SG		139.162.23.4		102MB/s
Linode, London, UK		176.58.107.39		9.49MB/s
Linode, Frankfurt, DE		139.162.130.8		4.67MB/s
Linode, Fremont, CA		50.116.14.9		9.03MB/s
Softlayer, Dallas, TX		173.192.68.18		8.46MB/s
Softlayer, Seattle, WA		67.228.112.250		12.0MB/s
Softlayer, Frankfurt, DE	159.122.69.4		8.59MB/s
Softlayer, Singapore, SG	119.81.28.170		65.8MB/s
Softlayer, HongKong, CN		119.81.130.170		48.9MB/s
----------------------------------------------------------------------
I/O speed(1st run) : 180 MB/s
I/O speed(2nd run) : 594 MB/s
I/O speed(3rd run) : 485 MB/s
Average I/O: 419.667 MB/s

Setelah itu UnixBench:

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com


1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

2 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

2 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

2 x Execl Throughput  1 2 3

2 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

2 x File Copy 256 bufsize 500 maxblocks  1 2 3

2 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

2 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

2 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

2 x Process Creation  1 2 3

2 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

2 x Shell Scripts (1 concurrent)  1 2 3

2 x Shell Scripts (8 concurrent)  1 2 3

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

   System: centos-2gb-sgp1-01: GNU/Linux
   OS: GNU/Linux -- 3.10.0-327.10.1.el7.x86_64 -- #1 SMP Tue Feb 16 17:03:50 UTC 2016
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz (4800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   CPU 1: Intel(R) Xeon(R) CPU E5-2630L v2 @ 2.40GHz (4800.0 bogomips)
          x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   14:46:31 up 15 min,  1 user,  load average: 0.12, 0.15, 0.13; runlevel 3

------------------------------------------------------------------------
Benchmark Run: Fri Apr 29 2016 14:46:31 - 15:14:15
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       23057194.5 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3010.2 MWIPS (8.4 s, 7 samples)
Execl Throughput                               1874.3 lps   (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        519052.7 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          160759.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1167811.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1191127.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 186920.3 lps   (10.0 s, 7 samples)
Process Creation                               3257.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3093.8 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    984.5 lpm   (60.0 s, 2 samples)
System Call Overhead                        1513172.5 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   23057194.5   1975.8
Double-Precision Whetstone                       55.0       3010.2    547.3
Execl Throughput                                 43.0       1874.3    435.9
File Copy 1024 bufsize 2000 maxblocks          3960.0     519052.7   1310.7
File Copy 256 bufsize 500 maxblocks            1655.0     160759.7    971.4
File Copy 4096 bufsize 8000 maxblocks          5800.0    1167811.9   2013.5
Pipe Throughput                               12440.0    1191127.0    957.5
Pipe-based Context Switching                   4000.0     186920.3    467.3
Process Creation                                126.0       3257.0    258.5
Shell Scripts (1 concurrent)                     42.4       3093.8    729.7
Shell Scripts (8 concurrent)                      6.0        984.5   1640.8
System Call Overhead                          15000.0    1513172.5   1008.8
                                                                   ========
System Benchmarks Index Score                                         862.2

------------------------------------------------------------------------
Benchmark Run: Fri Apr 29 2016 15:14:15 - 15:41:45
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       53498182.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     6976.7 MWIPS (6.3 s, 7 samples)
Execl Throughput                               5523.2 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        915786.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          249299.3 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2316129.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                             2512010.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 402676.2 lps   (10.0 s, 7 samples)
Process Creation                              14073.0 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   8723.4 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1183.5 lpm   (60.1 s, 2 samples)
System Call Overhead                        2877403.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   53498182.0   4584.2
Double-Precision Whetstone                       55.0       6976.7   1268.5
Execl Throughput                                 43.0       5523.2   1284.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     915786.1   2312.6
File Copy 256 bufsize 500 maxblocks            1655.0     249299.3   1506.3
File Copy 4096 bufsize 8000 maxblocks          5800.0    2316129.0   3993.3
Pipe Throughput                               12440.0    2512010.0   2019.3
Pipe-based Context Switching                   4000.0     402676.2   1006.7
Process Creation                                126.0      14073.0   1116.9
Shell Scripts (1 concurrent)                     42.4       8723.4   2057.4
Shell Scripts (8 concurrent)                      6.0       1183.5   1972.6
System Call Overhead                          15000.0    2877403.8   1918.3
                                                                   ========
System Benchmarks Index Score                                        1870.7

Terakhir adalah Geekbench: http://browser.primatelabs.com/geekbench3/6294343

Kesimpulannya? Kalau secara bandwidth dan kecepatan akses SSDnya sih antara Droplet kurang lebih setara. Tapi yang jadi pembedanya adalah kemampuan CPUnya dan besar RAM karena inilah yang akan berpengaruh terhadap performa VPSnya, jadi semakin banyak CPU corenya jelas semakin baik.

3 Comments

  1. wuih..josss tp mahal.. 😀
    review vultr dong mas…
    #aq pengen incip trialnya, tp syg gak ada paypal..ehhe

    1. Saya mau coba Vultr memang mas, tapi PayPalnya harus ada kartu kredit baru boleh ngisi creditnya. Haha. 😀 Lah saya belum punya kartu kredit. Semoga tahun ini bisa.

Tinggalkan Balasan ke ahmad fatah Batalkan balasan

Alamat email Anda tidak akan dipublikasikan. Ruas yang wajib ditandai *