If you are/were searching for cheap VPS hosting for your projects or looking for ways to test your project code before submitting it to the live server, probably you have already heard of DigitalOcean (DO).
DigitalOcean is quite a young (2011) company located in New York.
So why are they different? The first and most important thing is that DO uses hourly billing, especially when it comes to testing the projects. This means that you can create a droplet (this is how a VPS is named there), use it for a couple of hours (e.g. 5 hours), and then you can simply destroy it. Using a droplet for such short duration will cost you just a few cents. That is extremely useful feature and I have already used it several times while testing my laravel projects. However, you must keep in mind that if you do not want to pay for a specific droplet, you must destroy it altogether instead of shutting it down. The reason is, if you shut the droplet down, technically you are still using the DO’s resources and you will be charged for it.
DO uses KVM for their VPS, and OpenVZ is not supported there.
There are 9 specifications available ranging from 512MB RAM + 20GB SSD + 1 CPU for $5/month to 64GB RAM + 640GB SSD + 20CPU for $640/month.
That being said, if you choose a cheap plan initially, you can easily upgrade your droplet from the Control Panel.
You can choose from 5 locations for your droplet namely New York, Amsterdam, San Francisco, Singapore, and London. Among all the said locations, Singapore is comparatively more interesting as there are not too many VPS providers available in Asia.
Host nodes use 1Gb/s connections so you won’t be limited even when you need to transfer large amount of data. With every VPS you get one static IPv4 address (not more than one). These VPSs also support IPv6.
For every droplet you can enable private networking, backup and SSH key. You can store your SSH keys in the Control Panel and assign any of the keys to your newly created droplet.
There is a wide number of OS distributions available including Ubuntu, FreeBSD, Fedora, Debian, CoreOS and Centos, and almost all the operating systems are available in both 32 and 64bit editions.
You can also install turnkey distros e.g. LAMP on Ubuntu 14.04 or LEMP on Ubuntu 14.04. Currently 19 turnkey distros (applications) are available.
Another interesting thing about DO is that it supports snapshots. This means you can create snapshots of your droplet and use them later to build new droplets that have predefined configuration as that of the snapshots. In other words, you can use your base droplet’s snapshots to create its replicas (mirrors) within a few seconds.
However, the key point here is, that you can create snapshots of your droplet only when it is shut down. Snapshots of a droplet cannot be created if it is in running state.
Note: Shutting down a droplet is not a big deal in test environment, whereas it can be a real pain in the neck when working on a live server.
I was able to create a droplet within 1 minute. They claim that your droplet will be created within 55 seconds. My average time was 52 seconds which is quite good, especially when you are creating multiple droplets for testing purposes. There was only one instance (while testing) when it took somewhere around 5 minutes to create a droplet. Apparently there was some unknown issue at the server side which even I am unable to guess at the moment.
There were lots of questions in the forums about droplet speed, network speed, latency and stability, DO support, etc. and therefore I decided to run a few tests and publish them here.
I created one small droplet at each location (i.e. 5 droplets at 5 locations), where every droplet had minimum available configuration, i.e. 512MB RAM + 20GB SSD + 1 CPU, and each droplet had Ubuntu x64 14.04 running as its operating system.
First I used script from freevps to test the download speed of the droplets and ran UnixBench to measure hardware performance.
I ran the download speed test twice, where the second test was performed around 40min after the first one. After the second test, I came up with some interesting results. (Take a look at the result from Singapore location).
Conclusion:
Pros:
- 5 locations.
- SSD drives in standard config.
- User friendly Control Panel.
- IPv6 (not very important for most users).
- Hourly billing.
- Very flexible (upgrade/downgrade).
- Ability to move between locations.
- Backups.
- Snapshots.
- SSH keys.
- DNS.
- Private network.
- 1Gb/s connection.
- API.
Cons:
- Only one IPv4.
- You can’t create snapshots when your system is running (and to be honest it’s not easy).
- No location in Australia and Japan (again…not very important for most users).
- You can’t use startup scripts.
- They claim its cloud but actually it isn’t. (I.e. no SAN storage, no fail-over, no redundancy, no firewall, etc.)
- The minimum RAM for VPS you can order is 512MB. (I would prefer to start from 128MB for tiny VPS, unfortunately for KVM VPS it’s probably impossible.)
- Upgrade/downgrade doesn’t work in all locations.
- You can’t create a droplet from your own ISO.
- No minimal distributions available.
- No openVZ available. (KVM uses more memory than OpenVZ even though it has far better isolation.)
- No public node uptime statistics.
Test code used :
1 2 3 4 5 6 7 |
apt-get --yes --force-yes update apt-get --yes --force-yes install libx11-dev libgl1-mesa-dev libxext-dev perl perl-modules make gcc wget freevps.us/downloads/bench.sh -O - -o /dev/null|bash wget http://byte-unixbench.googlecode.com/files/UnixBench5.1.3.tgz tar xvf UnixBench5.1.3.tgz cd UnixBench ./Run |
Here are the results by location:
NewYork:
Speed test #1:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 1 min, Download speed from CacheFly: 78.8MB/s Download speed from Coloat, Atlanta GA: 55.2MB/s Download speed from Softlayer, Dallas, TX: 29.8MB/s Download speed from Linode, Tokyo, JP: 5.71MB/s Download speed from i3d.net, Rotterdam, NL: 11.4MB/s Download speed from Leaseweb, Haarlem, NL: 17.1MB/s Download speed from Softlayer, Singapore: 6.97MB/s Download speed from Softlayer, Seattle, WA: 22.2MB/s Download speed from Softlayer, San Jose, CA: 20.1MB/s Download speed from Softlayer, Washington, DC: 69.9MB/s I/O speed : 481 MB/s |
Speed test #2:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 45 min, Download speed from CacheFly: 106MB/s Download speed from Coloat, Atlanta GA: 32.2MB/s Download speed from Softlayer, Dallas, TX: 26.9MB/s Download speed from Linode, Tokyo, JP: 8.20MB/s Download speed from i3d.net, Rotterdam, NL: 12.1MB/s Download speed from Leaseweb, Haarlem, NL: 18.2MB/s Download speed from Softlayer, Singapore: 5.98MB/s Download speed from Softlayer, Seattle, WA: 20.8MB/s Download speed from Softlayer, San Jose, CA: 22.3MB/s Download speed from Softlayer, Washington, DC: 83.3MB/s I/O speed : 349 MB/s |
Unix Benchmark:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
======================================================================== BYTE UNIX Benchmarks (Version 5.1.3) System: tst1: GNU/Linux OS: GNU/Linux -- 3.13.0-43-generic -- #72-Ubuntu SMP Mon Dec 8 19:35:06 UTC 2014 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 07:49:25 up 2 min, 1 user, load average: 0.38, 0.18, 0.07; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Sun Feb 08 2015 07:49:25 - 08:17:35 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 27787237.4 lps (10.0 s, 7 samples) Double-Precision Whetstone 3655.2 MWIPS (9.9 s, 7 samples) Execl Throughput 3485.7 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 940147.0 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 265557.6 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1611958.7 KBps (30.0 s, 2 samples) Pipe Throughput 2090582.8 lps (10.0 s, 7 samples) Pipe-based Context Switching 287613.5 lps (10.0 s, 7 samples) Process Creation 10261.3 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 6611.7 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 848.9 lpm (60.0 s, 2 samples) System Call Overhead 3869554.5 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 27787237.4 2381.1 Double-Precision Whetstone 55.0 3655.2 664.6 Execl Throughput 43.0 3485.7 810.6 File Copy 1024 bufsize 2000 maxblocks 3960.0 940147.0 2374.1 File Copy 256 bufsize 500 maxblocks 1655.0 265557.6 1604.6 File Copy 4096 bufsize 8000 maxblocks 5800.0 1611958.7 2779.2 Pipe Throughput 12440.0 2090582.8 1680.5 Pipe-based Context Switching 4000.0 287613.5 719.0 Process Creation 126.0 10261.3 814.4 Shell Scripts (1 concurrent) 42.4 6611.7 1559.4 Shell Scripts (8 concurrent) 6.0 848.9 1414.9 System Call Overhead 15000.0 3869554.5 2579.7 ======== System Benchmarks Index Score 1434.7 |
San Francisco:
Speed test #1:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 1 min, Download speed from CacheFly: 106MB/s Download speed from Coloat, Atlanta GA: 19.8MB/s Download speed from Softlayer, Dallas, TX: 33.1MB/s Download speed from Linode, Tokyo, JP: 11.1MB/s Download speed from i3d.net, Rotterdam, NL: 6.07MB/s Download speed from Leaseweb, Haarlem, NL: 13.7MB/s Download speed from Softlayer, Singapore: 8.91MB/s Download speed from Softlayer, Seattle, WA: 60.4MB/s Download speed from Softlayer, San Jose, CA: 99.0MB/s Download speed from Softlayer, Washington, DC: 21.0MB/s I/O speed : 867 MB/s |
Speed test #2:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 43 min, Download speed from CacheFly: 104MB/s Download speed from Coloat, Atlanta GA: 19.3MB/s Download speed from Softlayer, Dallas, TX: 34.2MB/s Download speed from Linode, Tokyo, JP: 17.5MB/s Download speed from i3d.net, Rotterdam, NL: 5.91MB/s Download speed from Leaseweb, Haarlem, NL: 15.3MB/s Download speed from Softlayer, Singapore: 9.82MB/s Download speed from Softlayer, Seattle, WA: 62.3MB/s Download speed from Softlayer, San Jose, CA: 99.9MB/s Download speed from Softlayer, Washington, DC: 23.1MB/s I/O speed : 820 MB/s |
Unix Benchmark:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
======================================================================== BYTE UNIX Benchmarks (Version 5.1.3) System: sanfran: GNU/Linux OS: GNU/Linux -- 3.13.0-43-generic -- #72-Ubuntu SMP Mon Dec 8 19:35:06 UTC 2014 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 07:38:04 up 4:07, 1 user, load average: 0.09, 0.05, 0.06; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Mon Feb 09 2015 07:38:04 - 08:06:12 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 26088634.7 lps (10.0 s, 7 samples) Double-Precision Whetstone 3411.3 MWIPS (9.9 s, 7 samples) Execl Throughput 3927.5 lps (29.9 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 952261.9 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 268531.6 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1569243.1 KBps (30.0 s, 2 samples) Pipe Throughput 1921340.1 lps (10.0 s, 7 samples) Pipe-based Context Switching 295306.4 lps (10.0 s, 7 samples) Process Creation 12799.4 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 6949.2 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 896.3 lpm (60.0 s, 2 samples) System Call Overhead 3607547.2 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 26088634.7 2235.5 Double-Precision Whetstone 55.0 3411.3 620.2 Execl Throughput 43.0 3927.5 913.4 File Copy 1024 bufsize 2000 maxblocks 3960.0 952261.9 2404.7 File Copy 256 bufsize 500 maxblocks 1655.0 268531.6 1622.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 1569243.1 2705.6 Pipe Throughput 12440.0 1921340.1 1544.5 Pipe-based Context Switching 4000.0 295306.4 738.3 Process Creation 126.0 12799.4 1015.8 Shell Scripts (1 concurrent) 42.4 6949.2 1639.0 Shell Scripts (8 concurrent) 6.0 896.3 1493.9 System Call Overhead 15000.0 3607547.2 2405.0 ======== System Benchmarks Index Score 1456.6 |
Amsterdam:
Speed test #1:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 1 min, Download speed from CacheFly: 77.4MB/s Download speed from Coloat, Atlanta GA: 15.2MB/s Download speed from Softlayer, Dallas, TX: 14.9MB/s Download speed from Linode, Tokyo, JP: 6.54MB/s Download speed from i3d.net, Rotterdam, NL: 98.2MB/s Download speed from Leaseweb, Haarlem, NL: 94.9MB/s Download speed from Softlayer, Singapore: 5.02MB/s Download speed from Softlayer, Seattle, WA: 11.0MB/s Download speed from Softlayer, San Jose, CA: 10.3MB/s Download speed from Softlayer, Washington, DC: 22.0MB/s I/O speed : 340 MB/s |
Speed test #2:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 46 min, Download speed from CacheFly: 52.9MB/s Download speed from Coloat, Atlanta GA: 14.8MB/s Download speed from Softlayer, Dallas, TX: 14.3MB/s Download speed from Linode, Tokyo, JP: 7.81MB/s Download speed from i3d.net, Rotterdam, NL: 104MB/s Download speed from Leaseweb, Haarlem, NL: 36.7MB/s Download speed from Softlayer, Singapore: 5.25MB/s Download speed from Softlayer, Seattle, WA: 12.3MB/s Download speed from Softlayer, San Jose, CA: 11.7MB/s Download speed from Softlayer, Washington, DC: 20.9MB/s I/O speed : 328 MB/s |
Unix Benchmark:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
======================================================================== BYTE UNIX Benchmarks (Version 5.1.3) System: ts1: GNU/Linux OS: GNU/Linux -- 3.13.0-43-generic -- #72-Ubuntu SMP Mon Dec 8 19:35:06 UTC 2014 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 07:37:52 up 4:09, 1 user, load average: 0.34, 0.11, 0.07; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Mon Feb 09 2015 07:37:52 - 08:05:51 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 27347544.5 lps (10.0 s, 7 samples) Double-Precision Whetstone 3629.8 MWIPS (9.5 s, 7 samples) Execl Throughput 4292.5 lps (30.0 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 1035625.9 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 289868.0 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1659416.8 KBps (30.0 s, 2 samples) Pipe Throughput 2074686.8 lps (10.0 s, 7 samples) Pipe-based Context Switching 320408.9 lps (10.0 s, 7 samples) Process Creation 13277.8 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 7418.9 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 976.7 lpm (60.0 s, 2 samples) System Call Overhead 3867414.6 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 27347544.5 2343.4 Double-Precision Whetstone 55.0 3629.8 660.0 Execl Throughput 43.0 4292.5 998.2 File Copy 1024 bufsize 2000 maxblocks 3960.0 1035625.9 2615.2 File Copy 256 bufsize 500 maxblocks 1655.0 289868.0 1751.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 1659416.8 2861.1 Pipe Throughput 12440.0 2074686.8 1667.8 Pipe-based Context Switching 4000.0 320408.9 801.0 Process Creation 126.0 13277.8 1053.8 Shell Scripts (1 concurrent) 42.4 7418.9 1749.7 Shell Scripts (8 concurrent) 6.0 976.7 1627.8 System Call Overhead 15000.0 3867414.6 2578.3 ======== System Benchmarks Index Score 1560.9 |
London:
Speed test #1:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 2 min, Download speed from CacheFly: 68.2MB/s Download speed from Coloat, Atlanta GA: 15.0MB/s Download speed from Softlayer, Dallas, TX: 12.9MB/s Download speed from Linode, Tokyo, JP: 7.68MB/s Download speed from i3d.net, Rotterdam, NL: 71.5MB/s Download speed from Leaseweb, Haarlem, NL: 12.8MB/s Download speed from Softlayer, Singapore: 4.97MB/s Download speed from Softlayer, Seattle, WA: 11.6MB/s Download speed from Softlayer, San Jose, CA: 6.16MB/s Download speed from Softlayer, Washington, DC: 17.5MB/s I/O speed : 265 MB/s |
Speed test #2:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 38 min, Download speed from CacheFly: 40.1MB/s Download speed from Coloat, Atlanta GA: 16.7MB/s Download speed from Softlayer, Dallas, TX: 13.3MB/s Download speed from Linode, Tokyo, JP: 6.54MB/s Download speed from i3d.net, Rotterdam, NL: 70.3MB/s Download speed from Leaseweb, Haarlem, NL: 12.9MB/s Download speed from Softlayer, Singapore: 4.76MB/s Download speed from Softlayer, Seattle, WA: 10.7MB/s Download speed from Softlayer, San Jose, CA: 9.87MB/s Download speed from Softlayer, Washington, DC: 19.5MB/s I/O speed : 257 MB/s |
Unix Benchmark:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
======================================================================== BYTE UNIX Benchmarks (Version 5.1.3) System: london: GNU/Linux OS: GNU/Linux -- 3.13.0-43-generic -- #72-Ubuntu SMP Mon Dec 8 19:35:06 UTC 2014 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 07:38:12 up 4:00, 1 user, load average: 0.19, 0.07, 0.06; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Mon Feb 09 2015 07:38:12 - 08:05:32 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 26319042.2 lps (10.0 s, 7 samples) Double-Precision Whetstone 3395.4 MWIPS (6.4 s, 7 samples) Execl Throughput 3716.2 lps (29.8 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 887150.5 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 267223.2 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1537151.9 KBps (30.0 s, 2 samples) Pipe Throughput 1910514.1 lps (10.0 s, 7 samples) Pipe-based Context Switching 294738.5 lps (10.0 s, 7 samples) Process Creation 12740.4 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 6471.4 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 895.0 lpm (60.0 s, 2 samples) System Call Overhead 3706963.7 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 26319042.2 2255.3 Double-Precision Whetstone 55.0 3395.4 617.3 Execl Throughput 43.0 3716.2 864.2 File Copy 1024 bufsize 2000 maxblocks 3960.0 887150.5 2240.3 File Copy 256 bufsize 500 maxblocks 1655.0 267223.2 1614.6 File Copy 4096 bufsize 8000 maxblocks 5800.0 1537151.9 2650.3 Pipe Throughput 12440.0 1910514.1 1535.8 Pipe-based Context Switching 4000.0 294738.5 736.8 Process Creation 126.0 12740.4 1011.1 Shell Scripts (1 concurrent) 42.4 6471.4 1526.3 Shell Scripts (8 concurrent) 6.0 895.0 1491.7 System Call Overhead 15000.0 3706963.7 2471.3 ======== System Benchmarks Index Score 1431.9 |
Singapore:
Speed test #1:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 1 min, Download speed from CacheFly: 13.9MB/s Download speed from Coloat, Atlanta GA: 7.00MB/s Download speed from Softlayer, Dallas, TX: 10.2MB/s Download speed from Linode, Tokyo, JP: 17.8MB/s Download speed from i3d.net, Rotterdam, NL: 2.26MB/s Download speed from Leaseweb, Haarlem, NL: 6.41MB/s Download speed from Softlayer, Singapore: 88.3MB/s Download speed from Softlayer, Seattle, WA: 9.48MB/s Download speed from Softlayer, San Jose, CA: 10.7MB/s Download speed from Softlayer, Washington, DC: 8.09MB/s I/O speed : 91.9 MB/s |
Speed test #2:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 |
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 : 490 MB Total amount of swap : 0 MB System uptime : 40 min, Download speed from CacheFly: 14.1MB/s Download speed from Coloat, Atlanta GA: 6.90MB/s Download speed from Softlayer, Dallas, TX: 8.88MB/s Download speed from Linode, Tokyo, JP: 16.1MB/s Download speed from i3d.net, Rotterdam, NL: 2.96MB/s Download speed from Leaseweb, Haarlem, NL: 5.70MB/s Download speed from Softlayer, Singapore: 43.2MB/s Download speed from Softlayer, Seattle, WA: 10.5MB/s Download speed from Softlayer, San Jose, CA: 6.83MB/s Download speed from Softlayer, Washington, DC: 7.36MB/s I/O speed : 264 MB/s |
Unix Benchmark:
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 |
======================================================================== BYTE UNIX Benchmarks (Version 5.1.3) System: singapore: GNU/Linux OS: GNU/Linux -- 3.13.0-43-generic -- #72-Ubuntu SMP Mon Dec 8 19:35:06 UTC 2014 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 07:39:13 up 4:04, 1 user, load average: 0.25, 0.08, 0.06; runlevel 2 ------------------------------------------------------------------------ Benchmark Run: Mon Feb 09 2015 07:39:13 - 08:06:38 1 CPU in system; running 1 parallel copy of tests Dhrystone 2 using register variables 24722567.6 lps (10.0 s, 7 samples) Double-Precision Whetstone 3346.5 MWIPS (6.9 s, 7 samples) Execl Throughput 3960.0 lps (29.7 s, 2 samples) File Copy 1024 bufsize 2000 maxblocks 915753.1 KBps (30.0 s, 2 samples) File Copy 256 bufsize 500 maxblocks 282095.4 KBps (30.0 s, 2 samples) File Copy 4096 bufsize 8000 maxblocks 1582557.7 KBps (30.0 s, 2 samples) Pipe Throughput 2018965.0 lps (10.0 s, 7 samples) Pipe-based Context Switching 286340.6 lps (10.0 s, 7 samples) Process Creation 11479.8 lps (30.0 s, 2 samples) Shell Scripts (1 concurrent) 6760.1 lpm (60.0 s, 2 samples) Shell Scripts (8 concurrent) 918.6 lpm (60.0 s, 2 samples) System Call Overhead 3164131.2 lps (10.0 s, 7 samples) System Benchmarks Index Values BASELINE RESULT INDEX Dhrystone 2 using register variables 116700.0 24722567.6 2118.5 Double-Precision Whetstone 55.0 3346.5 608.5 Execl Throughput 43.0 3960.0 920.9 File Copy 1024 bufsize 2000 maxblocks 3960.0 915753.1 2312.5 File Copy 256 bufsize 500 maxblocks 1655.0 282095.4 1704.5 File Copy 4096 bufsize 8000 maxblocks 5800.0 1582557.7 2728.5 Pipe Throughput 12440.0 2018965.0 1623.0 Pipe-based Context Switching 4000.0 286340.6 715.9 Process Creation 126.0 11479.8 911.1 Shell Scripts (1 concurrent) 42.4 6760.1 1594.4 Shell Scripts (8 concurrent) 6.0 918.6 1531.1 System Call Overhead 15000.0 3164131.2 2109.4 ======== System Benchmarks Index Score 1424.1 |
As you can see in the graph below, the best I/O speed was achieved in San Francisco, and Singapore location was quite unstable.
It seems that CacheFly can serve users from North America far better than those in Asia.
Talking about uptime, I found no issue so far. None of my test droplets experienced any downtime.
So, would I recommend DigitalOcean ? My answer is Yes! Why ? Because DO is cheap and offers good service/price ratio. For $5/month you can get 512MB RAM and 20GB fast SSD space which, after some fine tuning, can easily handle a few thousands of daily visitors on your WordPress powered site(s). In addition to this, if you need VPS for testing purposes, you won’t find any better option than hourly billing (offered by DO). Simply, create, test and destroy 🙂
Click this link to get $10 credit when you sign up.
If you found this review useful please share it. Thank you !