96MB Low End VPS Review Part 65–VPS Blast
Recently, there seems to be more and more providers offering either pure SSD-based VPS or SSD-cached + HDD based hybrid VPS. Although normally comes with very limited hard drive space, SSD-based VPS are promised to have amazing disk I/O speed and is suitable for I/O intensive applications such as databases. However, are those magical SSDs really as good as they have promised? Is the sacrifice in storage space really going to bring you the fast and stable disk I/O? Let’s take a look with VPSBlast.
Basic Information and Sign Up
As per their advertisement on WebhostingTalk, here is the resources you would be getting for 2.50USD per month:
Note that this plan is not offered directly on their homepage and if you order directly on their homepage, the smallest plan is 512MB of RAM and 10GB of SSD space for 10 USD per month:
Slightly below the slider bar, they have a pretty good introduction about what they offer:
Provision was instant and I received my new VPS information email the minute when I received my Paypal payment confirmation.
Note that their VPS information email is a little different than other providers, partly because instead of using SolusVM, they actually use KiwiVM control panel, which is a properitoty panel that they have developed in-house from scratch as per the comment made by Dan. I always give providers some extra points for developing their own control panels, since it shows the providers actually knows what they are doing and also indicates a certain level of commitment they have for the business since normally the R&D cost for control panels is quite high.
Once logged into their billing panel and click on Manage Server, you would be able to see a link that will lead you to the KiwiVM control panel:
You can also go to the Kiwi panel login page, which will ask you to login using your server’s IP address and password. Note that because of IP address is used as the user name rather than the actual user account, you will have to login separately for each VPS, instead of having every server under your name/email to be on one page. Furthermore, I am not sure what would be the password used to log into the server as it was never displayed anywhere, so I would assume the best way to login is just through the billing interface.
The main control panel for KiwiVM is actually very simplistic and clean, with everything on a single page, which is something I really like:
You can change the SSH port if necessary:
Note that if you have changed your SSH port within the VPS, the record won’t be updated in the control panel.
Basic actions can be performed on the control panel, including start, stop, reboot and it has a web console as well. However, the web console is a little different from the console in SolsuVM, which is essentially a Java-based SSH client. Instead, the web-shell in KiwiVM is a web-based page that allows you to run a single command at a time.
Once the command is run, you can see the output below:
You can click on the detailed statistics to see the detailed usage of your VPS resources:
As you can see, the number of graphs is almost overwhelming, and I am not sure if it would be necessary to have this much information available.
There is a pretty comprehensive list of operating systems available for reinstall:
arch-2010.05-i686-minimal arch-2010.05-x86_64-minimal arch-2012-11-19-x86_64-minimal-caner-version centos-4-i386-minimal centos-4-x86 centos-4-x86_64-minimal centos-4-x86_64 centos-5-x86 centos-5-x86_64 centos-6-x86 centos-6-x86_64 debian-3.1-x86 debian-4.0-x86 debian-4.0-x86_64 debian-5.0-x86 debian-5.0-x86_64 debian-6-turnkey-nginx-php-fastcgi_12.0-1_i386 debian-6.0-i386-minimal-beta debian-6.0-x86 debian-6.0-x86_64 fedora-10-x86 fedora-10-x86_64 fedora-11-x86 fedora-11-x86_64 fedora-12-x86 fedora-12-x86_64 fedora-13-x86 fedora-13-x86_64 fedora-14-x86 fedora-14-x86_64 fedora-15-x86 fedora-15-x86_64 gentoo-openvz-x86-2008.11.30 opensuse-11.0-i586-20081217 owl-3.0-release-i686 owl-3.0-release-x86_64 owl-current-20100323-i686 owl-current-20100323-x86_64 scientific-6-x86 scientific-6-x86_64 slackware-13.1-i386-minimal suse-11.2-x86 suse-11.2-x86_64 suse-11.3-x86 suse-11.3-x86_64 suse-11.4-x86 suse-11.4-x86_64 ubuntu-10.04-x86 ubuntu-10.04-x86_64 ubuntu-10.10-x86 ubuntu-10.10-x86_64 ubuntu-11.04-x86 ubuntu-11.04-x86_64 ubuntu-11.10-x86 ubuntu-11.10-x86_64 ubuntu-12.04-x86 ubuntu-12.04-x86_64 ubuntu-8.04-x86 ubuntu-8.04-x86_64 ubuntu-8.10-x86 ubuntu-8.10-x86_64 ubuntu-9.04-x86 ubuntu-9.04-x86_64 ubuntu-9.10-x86 ubuntu-9.10-x86_64
In fact, there are altogether 65 templates that covers almost all of the latest versions of major Linux distributions, there are also minimal and turnkey templates which might be useful as well.
And finally, host name, root password, control panel password and contact email can all be changed from the interface:
Test on the VPS
As mentioned above, the VPS has 256MB of RAM, 5GB of SSD-based hard drive space, and 250GB of bandwidth on 1Gbit port (which seems a little low). The VPS is located in ColoCrossing datacenter in Buffalo, New York, although interestingly, the IP was geo-located to Niagara Falls, Canada, which could be useful for people who are in need of a Canadian IP address. I have installed Debian 6 minimal 32 bit for testing purposes.
When the OS was first loaded, 6MB of RAM was used, which indicates the OS template was optimized for smaller RAM usage:
free -m
total used free shared buffers cached
Mem: 256 14 241 0 0 8
-/+ buffers/cache: 6 249
Swap: 128 0 127
Top output showing the processes running:
top - 01:50:50 up 51 days, 6:42, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 18 total, 1 running, 17 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 262144k total, 15260k used, 246884k free, 0k buffers
Swap: 131072k total, 612k used, 130460k free, 8448k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
12635 root 20 0 8464 2944 2324 S 0 1.1 0:00.11 sshd
12638 root 20 0 2976 1632 1300 S 0 0.6 0:00.01 bash
12647 root 20 0 2328 1068 876 R 0 0.4 0:00.00 top
314 root 20 0 34464 716 472 S 0 0.3 0:14.43 rsyslogd
377 root 20 0 5488 316 212 S 0 0.1 0:12.94 sshd
370 root 20 0 2288 164 112 S 0 0.1 0:03.17 cron
1 root 20 0 2028 84 52 S 0 0.0 0:32.62 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd/1285
3 root 20 0 0 0 0 S 0 0.0 0:00.00 khelper/1285
4 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/0
5 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/1
6 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/2
7 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/3
8 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/4
9 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/5
10 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/6
11 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/7
Htop output:
About 170MB of hard drive space was used:
df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 5.0G 170M 4.9G 4% / tmpfs 128M 0 128M 0% /lib/init/rw tmpfs 128M 0 128M 0% /dev/shm
And the inodes usage is pretty standard:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/simfs 5120000 10544 5109456 1% / tmpfs 32768 4 32764 1% /lib/init/rw tmpfs 32768 1 32767 1% /dev/shm
When the full LNMP stack was loaded, RAM usage goes up to 22MB, which is still significantly lower than what normally would be:
free -m
total used free shared buffers cached
Mem: 256 252 3 0 0 229
-/+ buffers/cache: 22 233
Swap: 128 7 120
Top showing the processes running:
top - 10:16:05 up 51 days, 7:08, 1 user, load average: 0.34, 0.78, 0.58
Tasks: 28 total, 1 running, 27 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 262144k total, 258824k used, 3320k free, 0k buffers
Swap: 131072k total, 7572k used, 123500k free, 235328k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
19432 www 20 0 14996 10m 416 S 0 4.2 0:00.01 nginx
19413 mysql 20 0 33948 2024 2020 S 0 0.8 0:00.00 mysqld
19423 root 20 0 22624 1624 1404 S 0 0.6 0:00.00 php-cgi
19426 www 20 0 22624 1444 992 S 0 0.6 0:00.00 php-cgi
19427 www 20 0 22624 1444 992 S 0 0.6 0:00.00 php-cgi
19428 www 20 0 22624 1440 992 S 0 0.5 0:00.00 php-cgi
19425 www 20 0 22624 1388 992 S 0 0.5 0:00.00 php-cgi
19424 www 20 0 22624 1356 992 S 0 0.5 0:00.00 php-cgi
19457 root 20 0 2328 1092 884 R 0 0.4 0:00.00 top
12638 root 20 0 2976 1008 848 S 0 0.4 0:00.01 bash
19431 root 20 0 4784 724 276 S 0 0.3 0:00.00 nginx
314 root 20 0 34464 540 436 S 0 0.2 0:14.44 rsyslogd
12635 root 20 0 8464 524 408 S 0 0.2 0:01.32 sshd
19311 root 20 0 1744 460 456 S 0 0.2 0:00.00 mysqld_safe
370 root 20 0 2288 156 108 S 0 0.1 0:03.17 cron
377 root 20 0 5488 144 140 S 0 0.1 0:12.94 sshd
1 root 20 0 2028 96 76 S 0 0.0 0:32.63 init
2 root 20 0 0 0 0 S 0 0.0 0:00.00 kthreadd/1285
3 root 20 0 0 0 0 S 0 0.0 0:00.00 khelper/1285
4 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/0
5 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/1
6 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/2
7 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/3
8 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/4
9 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/5
10 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/6
11 root 20 0 0 0 0 S 0 0.0 0:00.00 rpciod/1285/7
12 root 20 0 0 0 0 S 0 0.0 0:00.00 nfsiod/1285
Htop output:
About 1.6GB of hard drive space was used:
df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 5.0G 1.6G 3.5G 31% / tmpfs 128M 0 128M 0% /lib/init/rw tmpfs 128M 0 128M 0% /dev/shm
And inodes is again, pretty standard:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/simfs 5120000 63484 5056516 2% / tmpfs 32768 4 32764 1% /lib/init/rw tmpfs 32768 1 32767 1% /dev/shm
Uptime output shows the VPS is fairly stable:
uptime 10:03:28 up 72 days, 2:11, 1 user, load average: 0.00, 0.00, 0.00
VMstat shows there is no iowait:
vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 0 39840 242324 0 7720 0 0 1 0 0 3 0 0 100 0
CPUInfo shows there are two CPU cores assigned to this VPS, and the processor they used is E3-1270v2, which is a relatively new processor:
cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 58 model name : Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz stepping : 9 cpu MHz : 1750.500 cache size : 8192 KB physical id : 0 siblings : 8 core id : 0 cpu cores : 4 apicid : 0 initial apicid : 0 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 x2apic popcnt aes xsave avx f16c rdrand lahf_lm ida arat epb xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid fsgsbase smep erms bogomips : 6999.84 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 58 model name : Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz stepping : 9 cpu MHz : 800.000 cache size : 8192 KB physical id : 0 siblings : 8 core id : 1 cpu cores : 4 apicid : 2 initial apicid : 2 fpu : yes fpu_exception : yes cpuid level : 13 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm sse4_1 sse4_2 x2apic popcnt aes xsave avx f16c rdrand lahf_lm ida arat epb xsaveopt pln pts dts tpr_shadow vnmi flexpriority ept vpid fsgsbase smep erms bogomips : 6999.84 clflush size : 64 cache_alignment : 64 address sizes : 36 bits physical, 48 bits virtual power management:
Meminfo shows nothing too interesting:
cat /proc/meminfo MemTotal: 262144 kB MemFree: 628 kB Cached: 255712 kB Active: 93884 kB Inactive: 162236 kB Active(anon): 132 kB Inactive(anon): 276 kB Active(file): 93752 kB Inactive(file): 161960 kB Unevictable: 0 kB Mlocked: 0 kB SwapTotal: 131072 kB SwapFree: 107188 kB Dirty: 72220 kB Writeback: 0 kB AnonPages: 408 kB Shmem: 5128 kB Slab: 5384 kB SReclaimable: 2884 kB SUnreclaim: 2500 kB
Beancounters, as with the most recent versions of OpenVZ kernels, seem to show some relatively large values:
cat /proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
1285: kmemsize 5532169 14413824 112721920 123994112 0
lockedpages 0 0 27520 27520 0
privvmpages 26066 74977 9223372036854775807 9223372036854775807 0
shmpages 1282 1938 16384 16384 0
dummy 0 0 0 0 0
numproc 33 105 200 200 0
physpages 64910 65537 0 65536 0
vmguarpages 0 0 65536 9223372036854775807 0
oomguarpages 6073 6819 65536 9223372036854775807 0
numtcpsock 6 88 1670 1670 0
numflock 1 3 512 563 0
numpty 1 8 32 32 0
numsiginfo 0 27 256 256 0
tcpsndbuf 124392 1077184 8550400 12825600 0
tcprcvbuf 98304 8981000 8550400 12825600 178097
othersockbuf 6936 28224 4275200 8550400 0
dgramrcvbuf 0 4624 427520 427520 0
numothersock 36 43 1670 1670 0
dcachesize 2970375 7864320 7149382 7864320 0
numfile 313 525 10240 10240 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 24 24 384 384 0
Note that the failcnt for tcprcvbuf is not 0, which is uncommon since most of the time fail count should be 0 for all items.
As a result, the output of VZFree is also showing some weird numbers:
vzfree
Total Used Free
Kernel: 107.50M 5.28M 102.22M
Allocate: 36028797018963968.00M 101.82M 36028797018963868.00M (256M Guaranteed)
Commit: 256.00M 29.00M 227.00M (23.3% of Allocated)
Swap: -229.75M (-968.5% of Committed)
Time sync results:
time sync real 0m0.002s user 0m0.000s sys 0m0.001s
For the SSD-based VPS, the most interesting part is the disk I/O and VPSBlast was showing some inconsistent but pretty good results:
dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 10.9609 s, 98.0 MB/s
As you can see, this is not a lot better than a regular RAID 10 hard drive configuration.
And when I ran the I/O test again a while later, here is what I see:
dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync 16384+0 records in 16384+0 records out 1073741824 bytes (1.1 GB) copied, 4.24318 s, 253 MB/s
This is closer to what the disk I/O of a SSD configuration would offer.
IOPing results are pretty inconsistent as well:
ioping -c 10 . 4096 bytes from . (simfs /vz/private/1285): request=1 time=0.0 ms 4096 bytes from . (simfs /vz/private/1285): request=2 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=3 time=0.5 ms 4096 bytes from . (simfs /vz/private/1285): request=4 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=5 time=0.1 ms 4096 bytes from . (simfs /vz/private/1285): request=6 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=7 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=8 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=9 time=0.5 ms 4096 bytes from . (simfs /vz/private/1285): request=10 time=0.5 ms --- . (simfs /vz/private/1285) ioping statistics --- 10 requests completed in 9009.1 ms, 2181 iops, 8.5 mb/s min/avg/max/mdev = 0.0/0.5/0.6/0.2 ms
Normally for SSD I would expect the ioping to be around 0.1ms range, but the second time I ran it, the results are even less consistent:
ioping -c 10 . 4096 bytes from . (simfs /vz/private/1285): request=1 time=0.0 ms 4096 bytes from . (simfs /vz/private/1285): request=2 time=14.1 ms 4096 bytes from . (simfs /vz/private/1285): request=3 time=0.5 ms 4096 bytes from . (simfs /vz/private/1285): request=4 time=1.1 ms 4096 bytes from . (simfs /vz/private/1285): request=5 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=6 time=0.2 ms 4096 bytes from . (simfs /vz/private/1285): request=7 time=11.6 ms 4096 bytes from . (simfs /vz/private/1285): request=8 time=0.6 ms 4096 bytes from . (simfs /vz/private/1285): request=9 time=15.2 ms 4096 bytes from . (simfs /vz/private/1285): request=10 time=0.6 ms --- . (simfs /vz/private/1285) ioping statistics --- 10 requests completed in 9047.6 ms, 225 iops, 0.9 mb/s min/avg/max/mdev = 0.0/4.5/15.2/6.1 ms
As you can see, this results is not even as good as some of the HDD-based configurations.
The network on the other hand, is a lot more impressive. The server was on 1Gbit port, so when I tried to download the Cachefly test file, it was completed really quickly.
wget cachefly.cachefly.net/100mb.test -O /dev/null --2012-12-11 11:44:12-- http://cachefly.cachefly.net/100mb.test Resolving cachefly.cachefly.net... 205.234.175.175 Connecting to cachefly.cachefly.net|205.234.175.175|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[==========================================================================================>] 104,857,600 24.5M/s in 4.2s 2012-12-11 11:44:16 (23.9 MB/s) - `/dev/null' saved [104857600/104857600]
The testing for second time was showing even better results:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2013-04-30 08:05:30-- http://cachefly.cachefly.net/100mb.test Resolving cachefly.cachefly.net... 205.234.175.175 Connecting to cachefly.cachefly.net|205.234.175.175|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 40.6M/s in 2.5s 2013-04-30 08:05:33 (40.6 MB/s) - `/dev/null' saved [104857600/104857600]
For the 1Gbit port, I have normally use aget to test the multi-threaded download of a 1GB file, and this time it is no exception.
The first test was done on the ThinkBroadBand 1GB test file:
aget -n 10 http://ipv4.download.thinkbroadband.com/1GB.zipAttempting to read log file aget-1GB.zip.log for resuming download job... Couldn't find log file for this download, starting a clean job... Head-Request Connection established Downloading /1GB.zip (1073741824 bytes) from site ipv4.download.thinkbroadband.com(80.249.99.148:80). Number of Threads: 10 .. [4% completed] ..... [9% completed] ....... [14% completed] .......... [19% completed] ............ [24% completed] ............... [29% completed] ................. [34% completed] .................... [39% completed] ...................... [44% completed] ......................... [49% completed] ........................... [54% completed] .............................. [59% completed] ................................ [64% completed] ................................... [69% completed] ..................................... [74% completed] ........................................ [79% completed] .......................................... [84% completed] ............................................. [89% completed] ............................................... [94% completed] ............................................... [94% completed] .................................................. [99% completed] .................................................. [100% completed] Download completed, job completed in 89 seconds. (11781 Kb/sec) Shutting down...
Testing again:
aget -n 10 http://ipv4.download.thinkbroadband.com/1GB.zipAttempting to read log file aget-1GB.zip.log for resuming download job... Couldn't find log file for this download, starting a clean job... Head-Request Connection established Downloading /1GB.zip (1073741824 bytes) from site ipv4.download.thinkbroadband.com(80.249.99.148:80). Number of Threads: 10 .. [4% completed] ..... [9% completed] ....... [14% completed] .......... .......... .......... .......... [19% completed] [19% completed] [19% completed] [19% completed] ............ [24% completed] ............... [29% completed] ................. ................. [34% completed] [34% completed] .................... [39% completed] ...................... ...................... [44% completed] [44% completed] ......................... [49% completed] ........................... [54% completed] .............................. [59% completed] ................................ ................................ [64% completed] [64% completed] ................................... ................................... [69% completed] [69% completed] ..................................... [74% completed] ........................................ [79% completed] .......................................... [84% completed] ............................................. [89% completed] ............................................... [94% completed] .................................................. [99% completed] .................................................. [100% completed] Download completed, job completed in 89 seconds. (11781 Kb/sec) Shutting down...
Indeed the speed does not exactly show the speed of a 1Gbit port, however it is very consistent and it is almost known that ThinkBroadband itself, which is based in London, UK, is not very fast to start with.
The other test file that I have used is a 1GB test file from DimeNOC/HostDime, and the test file is located in Orlando, FL.
aget -n 10 http://72.29.70.131/1gbfile.tgzAttempting to read log file aget-1gbfile.tgz.log for resuming download job... Couldn't find log file for this download, starting a clean job... Head-Request Connection established Downloading /1gbfile.tgz (1023410176 bytes) from site 72.29.70.131(72.29.70.131:80). Number of Threads: 10 .. [4% completed] ..... [9% completed] ....... ....... [14% completed] [14% completed] .......... [19% completed] ............ [24% completed] ............... [29% completed] ................. ................. [34% completed] [34% completed] .................... [39% completed] ...................... [44% completed] ......................... [49% completed] ............................ .......................... [54% completed] [54% completed] .............................. [59% completed] ................................. ............................... [64% completed] [64% completed] ................................... [69% completed] ..................................... ..................................... [74% completed] [74% completed] ........................................ [79% completed] .......................................... .......................................... [84% completed] [84% completed] ............................................. [89% completed] ............................................... [94% completed] .................................................. [99% completed] .................................................. [100% completed] Download completed, job completed in 26 seconds. (38439 Kb/sec) Shutting down...
As you can see, the speed is now a lot better.
Testing again:
aget -n 10 http://72.29.70.131/1gbfile.tgzAttempting to read log file aget-1gbfile.tgz.log for resuming download job... Couldn't find log file for this download, starting a clean job... Head-Request Connection established Downloading /1gbfile.tgz (1023410176 bytes) from site 72.29.70.131(72.29.70.131:80). Number of Threads: 10 .. .. [4% completed] [4% completed] ..... [9% completed] ....... [14% completed] .......... [19% completed] ............ [24% completed] ............... [29% completed] ................. [34% completed] .................... ..................... ................... [39% completed] [39% completed] [39% completed] ...................... [44% completed] ......................... ......................... [49% completed] [49% completed] ........................... [54% completed] .............................. [59% completed] ................................ ................................ ................................ [64% completed] [64% completed] [64% completed] .................................... .................................. [69% completed] [69% completed] ..................................... [74% completed] ......................................... ....................................... [79% completed] [79% completed] .......................................... .......................................... [84% completed] [84% completed] ............................................. ............................................. [89% completed] [89% completed] ............................................... [94% completed] .................................................. [99% completed] .................................................. [100% completed] Download completed, job completed in 27 seconds. (37015 Kb/sec) Shutting down...
Again, as both servers are along the east coast, the download speed is much better.
For the upload speed tests, the first is on my Quickweb test VPS at Chicago, IL, which is on the east coast:
wget http://192.227.xxx.xxx/100mb.test -O /dev/null --2013-04-30 00:53:39-- http://192.227.xxx.xxx/100mb.test Connecting to 192.227.xxx.xxx:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 4.39M/s in 23s 2013-04-30 00:54:02 (4.33 MB/s) - `/dev/null' saved [104857600/104857600]
From my ChicagoVPS in Los Angeles, CA, which is on the west coast:
wget http://192.227.xxx.xxx/100mb.test -O /dev/null --2013-04-30 08:48:33-- http://192.227.xxx.xxx/100mb.test Connecting to 192.227.xxx.xxx:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 799K/s in 2m 13s 2013-04-30 08:50:47 (768 KB/s) - `/dev/null' saved [104857600/104857600]
Finally, the VPS from Prometeus from Milan, Italy, which is in Europe:
wget http://192.227.xxx.xxx/100mb.test -O /dev/null --2013-04-29 20:56:07-- http://192.227.xxx.xxx/100mb.test Connecting to 192.227.xxx.xxx:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 11.3M/s in 12s 2013-04-29 20:56:20 (8.33 MB/s) - `/dev/null' saved [104857600/104857600]
As you can see, the VPS seems to have the best download speed with a European VPS, which is not exactly the closest spot physically. However, doing a traceroute indicated to me that the particularly routing was highly optimized:
6 nyk-bb1-link.telia.net (80.91.253.122) 88.592 ms nyk-bb1-link.telia.net (213.155.135.5) 88.647 ms nyk-bb1-link.telia.net (80.91.251.96) 88.772 ms 7 buf-b1-link.telia.net (80.91.246.36) 98.020 ms 97.731 ms 98.157 ms 8 giglinx-ic-155660-buf-b1.c.telia.net (213.248.96.42) 112.883 ms 113.366 ms 112.527 ms 9 host.colocrossing.com (198.23.xxx.xxx) 105.883 ms 107.056 ms 106.560 ms 10 v29.xxx.xxx (198.245.xxx.xxx) 105.027 ms 105.241 ms 105.230 ms 11 192.227.xxx.xxx (192.227.xxx.xxx) 98.174 ms 98.049 ms 98.164 ms
Finally, benchmarks. With a SSD storage and two CPU cores I was hoping for some impressive Unixbench score, however the first time I ran through UnixBench test, this did not exactly happen:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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: xxxxxx: GNU/Linux
OS: GNU/Linux -- 2.6.32-042stab062.2 -- #1 SMP Wed Oct 10 18:28:35 MSK 2012
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz (6385.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz (6385.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
10:24:54 up 51 days, 7:16, 1 user, load average: 0.00, 0.12, 0.31; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Tue Dec 11 2012 10:24:54 - 10:53:02
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 20676550.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3354.1 MWIPS (10.0 s, 7 samples)
Execl Throughput 2497.2 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 716569.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 189359.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1675263.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1158210.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 64648.6 lps (10.0 s, 7 samples)
Process Creation 6555.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3503.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 575.6 lpm (60.1 s, 2 samples)
System Call Overhead 876835.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 20676550.5 1771.8
Double-Precision Whetstone 55.0 3354.1 609.8
Execl Throughput 43.0 2497.2 580.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 716569.6 1809.5
File Copy 256 bufsize 500 maxblocks 1655.0 189359.3 1144.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 1675263.0 2888.4
Pipe Throughput 12440.0 1158210.6 931.0
Pipe-based Context Switching 4000.0 64648.6 161.6
Process Creation 126.0 6555.9 520.3
Shell Scripts (1 concurrent) 42.4 3503.9 826.4
Shell Scripts (8 concurrent) 6.0 575.6 959.3
System Call Overhead 15000.0 876835.6 584.6
========
System Benchmarks Index Score 846.9
------------------------------------------------------------------------
Benchmark Run: Tue Dec 11 2012 10:53:02 - 11:23:25
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 18966959.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 6210.6 MWIPS (10.1 s, 7 samples)
Execl Throughput 2841.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 559216.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 149599.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1224017.4 KBps (30.0 s, 2 samples)
Pipe Throughput 1055048.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 65205.3 lps (10.0 s, 7 samples)
Process Creation 6553.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3568.8 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 555.0 lpm (60.2 s, 2 samples)
System Call Overhead 773853.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 18966959.7 1625.3
Double-Precision Whetstone 55.0 6210.6 1129.2
Execl Throughput 43.0 2841.0 660.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 559216.4 1412.2
File Copy 256 bufsize 500 maxblocks 1655.0 149599.7 903.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 1224017.4 2110.4
Pipe Throughput 12440.0 1055048.0 848.1
Pipe-based Context Switching 4000.0 65205.3 163.0
Process Creation 126.0 6553.1 520.1
Shell Scripts (1 concurrent) 42.4 3568.8 841.7
Shell Scripts (8 concurrent) 6.0 555.0 924.9
System Call Overhead 15000.0 773853.0 515.9
========
System Benchmarks Index Score 821.4
Testing again showed much better results, almost doubling the previous score:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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: xxxxxxx: GNU/Linux
OS: GNU/Linux -- 2.6.32-042stab075.2 -- #1 SMP Tue Mar 5 15:21:53 MSK 2013
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz (6999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E3-1270 V2 @ 3.50GHz (6999.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
11:01:31 up 11:14, 1 user, load average: 0.00, 0.00, 0.00; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Sun Mar 31 2013 11:01:31 - 11:29:35
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 26341754.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3859.5 MWIPS (10.0 s, 7 samples)
Execl Throughput 7853.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 862060.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 222662.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2295277.6 KBps (30.0 s, 2 samples)
Pipe Throughput 1332087.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 310172.4 lps (10.0 s, 7 samples)
Process Creation 20782.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 9780.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1253.7 lpm (60.0 s, 2 samples)
System Call Overhead 1016407.3 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 26341754.8 2257.2
Double-Precision Whetstone 55.0 3859.5 701.7
Execl Throughput 43.0 7853.0 1826.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 862060.2 2176.9
File Copy 256 bufsize 500 maxblocks 1655.0 222662.7 1345.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 2295277.6 3957.4
Pipe Throughput 12440.0 1332087.4 1070.8
Pipe-based Context Switching 4000.0 310172.4 775.4
Process Creation 126.0 20782.0 1649.4
Shell Scripts (1 concurrent) 42.4 9780.9 2306.8
Shell Scripts (8 concurrent) 6.0 1253.7 2089.6
System Call Overhead 15000.0 1016407.3 677.6
========
System Benchmarks Index Score 1518.2
------------------------------------------------------------------------
Benchmark Run: Sun Mar 31 2013 11:29:35 - 11:59:45
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 26048461.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 7680.0 MWIPS (10.0 s, 7 samples)
Execl Throughput 7694.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 859269.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 223645.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2327457.2 KBps (30.0 s, 2 samples)
Pipe Throughput 1318553.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 297629.4 lps (10.0 s, 7 samples)
Process Creation 20485.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 9562.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1214.4 lpm (60.0 s, 2 samples)
System Call Overhead 1002323.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 26048461.1 2232.1
Double-Precision Whetstone 55.0 7680.0 1396.4
Execl Throughput 43.0 7694.0 1789.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 859269.0 2169.9
File Copy 256 bufsize 500 maxblocks 1655.0 223645.7 1351.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 2327457.2 4012.9
Pipe Throughput 12440.0 1318553.7 1059.9
Pipe-based Context Switching 4000.0 297629.4 744.1
Process Creation 126.0 20485.0 1625.8
Shell Scripts (1 concurrent) 42.4 9562.4 2255.3
Shell Scripts (8 concurrent) 6.0 1214.4 2024.0
System Call Overhead 15000.0 1002323.5 668.2
========
System Benchmarks Index Score 1587.7
For Geekbench, we could see the processors are throttled to 800MHz, and the results are a lot more consistent than what was showing in the UnixBench results:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.32-042stab062.2 i686
Model: Linux PC (Intel(R) Xeon(R) CPU E31230 @ 3.20GHz)
Motherboard: Unknown Motherboard
Processor: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz
Processor ID: GenuineIntel Family 6 Model 42 Stepping 7
Logical Processors: 2
Physical Processors: 1
Processor Frequency: 800 MHz
L1 Instruction Cache: 0.00 B
L1 Data Cache: 0.00 B
L2 Cache: 256 KB
L3 Cache: 0.00 B
Bus Frequency: 0.00 Hz
Memory: 256 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz
Processor Cores: 2
Integer
Blowfish
single-threaded scalar 2241 ||||||||
multi-threaded scalar 4298 |||||||||||||||||
Text Compress
single-threaded scalar 2950 |||||||||||
multi-threaded scalar 2997 |||||||||||
Text Decompress
single-threaded scalar 3208 ||||||||||||
multi-threaded scalar 3791 |||||||||||||||
Image Compress
single-threaded scalar 2502 ||||||||||
multi-threaded scalar 4787 |||||||||||||||||||
Image Decompress
single-threaded scalar 2472 |||||||||
multi-threaded scalar 4895 |||||||||||||||||||
Lua
single-threaded scalar 4101 ||||||||||||||||
multi-threaded scalar 4326 |||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2760 |||||||||||
multi-threaded scalar 3121 ||||||||||||
Dot Product
single-threaded scalar 4656 ||||||||||||||||||
multi-threaded scalar 5955 |||||||||||||||||||||||
single-threaded vector 5647 ||||||||||||||||||||||
multi-threaded vector 12630 ||||||||||||||||||||||||||||||||||||||||||||||||||
LU Decomposition
single-threaded scalar 3589 ||||||||||||||
multi-threaded scalar 5853 |||||||||||||||||||||||
Primality Test
single-threaded scalar 4881 |||||||||||||||||||
multi-threaded scalar 3859 |||||||||||||||
Sharpen Image
single-threaded scalar 11442 |||||||||||||||||||||||||||||||||||||||||||||
multi-threaded scalar 20328 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 8476 |||||||||||||||||||||||||||||||||
multi-threaded scalar 10189 ||||||||||||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 6238 ||||||||||||||||||||||||
Write Sequential
single-threaded scalar 6234 ||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 5059 ||||||||||||||||||||
Stdlib Write
single-threaded scalar 4101 ||||||||||||||||
Stdlib Copy
single-threaded scalar 7837 |||||||||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 3163 ||||||||||||
single-threaded vector 3992 |||||||||||||||
Stream Scale
single-threaded scalar 3337 |||||||||||||
single-threaded vector 3903 |||||||||||||||
Stream Add
single-threaded scalar 3203 ||||||||||||
single-threaded vector 3685 ||||||||||||||
Stream Triad
single-threaded scalar 3454 |||||||||||||
single-threaded vector 2730 ||||||||||
Integer Score: 3547 ||||||||||||||
Floating Point Score: 7384 |||||||||||||||||||||||||||||
Memory Score: 5893 |||||||||||||||||||||||
Stream Score: 3433 |||||||||||||
Overall Geekbench Score: 5347 |||||||||||||||||||||
Testing again showed more or less similar results:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.32-042stab062.2 i686
Model: Linux PC (Intel(R) Xeon(R) CPU E31230 @ 3.20GHz)
Motherboard: Unknown Motherboard
Processor: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz
Processor ID: GenuineIntel Family 6 Model 42 Stepping 7
Logical Processors: 2
Physical Processors: 1
Processor Frequency: 800 MHz
L1 Instruction Cache: 0.00 B
L1 Data Cache: 0.00 B
L2 Cache: 256 KB
L3 Cache: 0.00 B
Bus Frequency: 0.00 Hz
Memory: 256 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel(R) Xeon(R) CPU E31230 @ 3.20GHz
Processor Cores: 2
Integer
Blowfish
single-threaded scalar 2226 ||||||||
multi-threaded scalar 4428 |||||||||||||||||
Text Compress
single-threaded scalar 2967 |||||||||||
multi-threaded scalar 2986 |||||||||||
Text Decompress
single-threaded scalar 3170 ||||||||||||
multi-threaded scalar 3903 |||||||||||||||
Image Compress
single-threaded scalar 2087 ||||||||
multi-threaded scalar 4727 ||||||||||||||||||
Image Decompress
single-threaded scalar 2482 |||||||||
multi-threaded scalar 4904 |||||||||||||||||||
Lua
single-threaded scalar 4108 ||||||||||||||||
multi-threaded scalar 3724 ||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2838 |||||||||||
multi-threaded scalar 2987 |||||||||||
Dot Product
single-threaded scalar 4551 ||||||||||||||||||
multi-threaded scalar 6293 |||||||||||||||||||||||||
single-threaded vector 5603 ||||||||||||||||||||||
multi-threaded vector 11007 ||||||||||||||||||||||||||||||||||||||||||||
LU Decomposition
single-threaded scalar 3262 |||||||||||||
multi-threaded scalar 4832 |||||||||||||||||||
Primality Test
single-threaded scalar 4906 |||||||||||||||||||
multi-threaded scalar 4073 ||||||||||||||||
Sharpen Image
single-threaded scalar 10008 ||||||||||||||||||||||||||||||||||||||||
multi-threaded scalar 18261 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 7940 |||||||||||||||||||||||||||||||
multi-threaded scalar 9925 |||||||||||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 5910 |||||||||||||||||||||||
Write Sequential
single-threaded scalar 5883 |||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 5094 ||||||||||||||||||||
Stdlib Write
single-threaded scalar 4023 ||||||||||||||||
Stdlib Copy
single-threaded scalar 8033 ||||||||||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 3240 ||||||||||||
single-threaded vector 3696 ||||||||||||||
Stream Scale
single-threaded scalar 3839 |||||||||||||||
single-threaded vector 3783 |||||||||||||||
Stream Add
single-threaded scalar 3192 ||||||||||||
single-threaded vector 3685 ||||||||||||||
Stream Triad
single-threaded scalar 3487 |||||||||||||
single-threaded vector 2692 ||||||||||
Integer Score: 3476 |||||||||||||
Floating Point Score: 6891 |||||||||||||||||||||||||||
Memory Score: 5788 |||||||||||||||||||||||
Stream Score: 3451 |||||||||||||
Overall Geekbench Score: 5131 ||||||||||||||||||||
Customer Service and Support
In order to test out the level of customer support, I have logged in two tickets, one is for a straight-forward “yes/no” type of question and the second one is to add an OS template, which I thought would take more time to do so. Both tickets are logged at about 9:30pm on a Saturday evening and the first question was responded in three minutes and the second one, was responded in four minutes, and the OS template was already added. I have to say such a quick speed in response is probably something that is truly impressive.
Conclusion
It is really difficult for me to come to a solid conclusion on VPSBlast, on one hand, the KiwiVM panel was impressive, network speed was great and the support is amazingly fast. However, the inconsistent disk I/O and UnixBench score is really what stopped me from saying this VPS as the “perfect” provider, particularly since most of the people are getting SSD VPS for the sake of getting good disk I/O and that is, for the most part, why people are paying the premium of the SSD VPS versus the traditional HDD-based configuration. Furthermore, the CPU cores are throttled, and considering the fact that most of the programs running in low end VPS are single-threaded, it is probably better to have a single unthrottled core than two throttled CPU cores. Nonetheless, overall, the experience with VPSBlast is positive and if you are looking for a SSD VPS with fast network and fast support, it definitely worth a try.
Please approve my new banner I uploaded on buysellads.com. You still have not approved and it seems to be taking a very long time. Thank you.
@Chad Done, sorry for the wait, if you have any issue with advertisement in future please email me admin AT 96mb DOT com.