96MB Low End VPS Review Part XXXI – QualityServers Xen Blowout
It is again another lazy weekend, I know I know, I am back to school now, and with work piling up (never thought a graduate degree would be more demanding than a undergraduate one!), I really should study day and night and should spend more time in the library. However, for anyone who have been working for a few years, you know how difficult it is to give up a lazy weekend and do something you really want to do (in my case, writing on 96MB.com!). Therefore, when I was contacted by a member of LowEndTalk asking me to do a review on one of the QualityServers VPS that he bought recently, I used that as an excuse (don’t like to borrow things from others for long!) and decided to put in another review for the week.
Disclosure: The member on LET has offered me a logo design for 96MB.com as a token of appreciation. Although he has told me he won’t be “offended” if I do not use it, but I figure I should try to avoid of getting myself into more trouble. I will get it applied to 96MB once I get a chance to rest my fingers. Do note, however, I have independently collected all the data and even reloaded OS from the SolusVM to make sure the data are as accurate as possible.
General Information and Set Up
I have not received a confirmation to release the LET member’s name or even his LET user name, so I guess I will just address him as Mr. A for now. In any case, Mr. A saw this great offer on LET and for 25 GBP a year, you get the following:
There are quite a few OS distributions you can choose from during sign up:
Backup space is definitely not cheap though, so you may want to think twice before getting more:
Although IP address is pretty standard price at 12GBP per year.
So Mr. A signed up for the deal and since he offered me access to his VPS, WHMCS system and SolusVM Control Panel for a short period of time, I was able to see and confirm a few things. For one, the provision was instant and the VPS login details was sent the moment payment was completed:
The welcome email was sent with root password in plain text and SolusVM on a non-standard HTTPS port with assigned user name. However, what makes it different is the number of IP addresses you get:
As you can see, there are 1 main IP (IPv4 IP and 16 IPv6 IPs), which is quite a few, although given the fact that IPv6 are basically for free these days, I guess it is really the time invested on doing all these custom scripting that are worth more than IP addresses themselves.
Once logged into their WHMCS, I was hoping to see the controls and statistics of my VPS under My Service, however, here is what I see when I clicked there:
I guess QualityServers has really tried hard to save the processing power of their main node.
After logged into the SolusVM control panel using HTTPS connection on non-standard port, a pretty standard interface of SolusVM shows:
As usual, none of the backup methods are available, however, instant rDNS is available, for both IPV4 and IPV6, which is rather great:
And the instant rDNS for IPv6 appears to be working as well:
There are quite a few OS templates available, covering many of the CentOS-based templates, as well as Debian 5 and 6 (both 32 and 64 bit OS templates), among others:
However, one thing I have noticed is that it took unusually long for the OS template to be reinstalled. I had to wait for almost a full 5 minutes before the OS templates are reloaded and the VPS is boot up so that I can SSH into the box.
Test on the VPS
As mentioned above, the VPS package has 256MB of dedicated RAM and 256MB of SWAP, and 10GB of hard drive space, and I used Debian 6 32 bit for the testing purposes. The OS was reloaded before I run all these tests to make sure everything is clean and there is no issue with the set up.
After the fresh reload of the OS template, the VPS used only 10MB of memory:
free -m
total used free shared buffers cached
Mem: 247 21 225 0 0 10
-/+ buffers/cache: 10 237
Swap: 255 0 255
The corresponding top output shows the processes running:
top - 04:21:56 up 28 min, 1 user, load average: 0.06, 0.03, 0.00
Tasks: 70 total, 1 running, 69 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: 253692k total, 22548k used, 231144k free, 984k buffers
Swap: 262136k total, 0k used, 262136k free, 10936k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
567 root 20 0 8264 2884 2332 S 0 1.1 0:00.03 sshd
570 root 20 0 2972 1612 1284 S 0 0.6 0:00.00 bash
508 root 20 0 28340 1476 1032 S 0 0.6 0:00.01 rsyslogd
576 root 20 0 2340 1128 896 R 0 0.4 0:00.01 top
536 root 20 0 5500 980 584 S 0 0.4 0:00.00 sshd
529 root 20 0 2300 772 604 S 0 0.3 0:00.00 cron
162 root 16 -4 2264 752 416 S 0 0.3 0:00.02 udevd
1 root 20 0 2040 724 628 S 0 0.3 0:00.22 init
195 root 18 -2 2260 648 316 S 0 0.3 0:00.00 udevd
196 root 18 -2 2260 640 308 S 0 0.3 0:00.00 udevd
559 root 20 0 1712 556 476 S 0 0.2 0:00.00 getty
553 root 20 0 1712 552 476 S 0 0.2 0:00.00 getty
554 root 20 0 1712 552 476 S 0 0.2 0:00.00 getty
555 root 20 0 1712 552 476 S 0 0.2 0:00.00 getty
556 root 20 0 1712 552 476 S 0 0.2 0:00.00 getty
558 root 20 0 1712 552 476 S 0 0.2 0:00.00 getty
557 root 20 0 1712 548 476 S 0 0.2 0:00.00 getty
And as usual, htop output for the htop fans:
Close to 700MB of hard drive space were used for the OS template to be installed, which is pretty much the average:
df -h Filesystem Size Used Avail Use% Mounted on /dev/xvda1 9.9G 691M 8.7G 8% / tmpfs 124M 0 124M 0% /lib/init/rw udev 114M 24K 114M 1% /dev tmpfs 124M 4.0K 124M 1% /dev/shm
After the full LNMP stack was loaded, about 40MB of RAM were used, which is pretty good:
free -m
total used free shared buffers cached
Mem: 247 198 49 0 10 148
-/+ buffers/cache: 39 208
Swap: 255 0 255
And the corresponding top output for the stack:
top - 20:49:49 up 8:56, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 79 total, 1 running, 78 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni, 99.9%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 253692k total, 202908k used, 50784k free, 10304k buffers
Swap: 262136k total, 484k used, 261652k free, 152072k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
8943 www 20 0 14508 10m 412 S 0 4.2 0:00.01 nginx
21505 mysql 20 0 34816 5680 2996 S 0 2.2 0:00.00 mysqld
8934 root 20 0 22716 4564 1404 S 0 1.8 0:04.14 php-cgi
8935 www 20 0 22716 4168 1008 S 0 1.6 0:00.00 php-cgi
8936 www 20 0 22716 4168 1008 S 0 1.6 0:00.00 php-cgi
8937 www 20 0 22716 4168 1008 S 0 1.6 0:00.00 php-cgi
8938 www 20 0 22716 4168 1008 S 0 1.6 0:00.00 php-cgi
8939 www 20 0 22716 4168 1008 S 0 1.6 0:00.00 php-cgi
567 root 20 0 8404 1740 1152 S 0 0.7 0:02.32 sshd
570 root 20 0 2972 1424 1092 S 0 0.6 0:00.02 bash
508 root 20 0 28340 1204 772 S 0 0.5 0:00.02 rsyslogd
9246 root 20 0 2340 1140 900 R 0 0.4 0:00.01 top
536 root 20 0 5500 864 468 S 0 0.3 0:00.02 sshd
8942 root 20 0 4696 700 264 S 0 0.3 0:00.00 nginx
529 root 20 0 2300 628 460 S 0 0.2 0:00.03 cron
1 root 20 0 2040 588 544 S 0 0.2 0:00.48 init
21403 root 20 0 1756 552 464 S 0 0.2 0:00.01 mysqld_safe
And of course, htop output for the htop fans:
About 2.1GB of hard drive space were used, which is pretty standard for a Debian 6 installation:
df -h Filesystem Size Used Avail Use% Mounted on /dev/xvda1 9.9G 2.1G 7.4G 22% / tmpfs 124M 0 124M 0% /lib/init/rw udev 114M 24K 114M 1% /dev tmpfs 124M 4.0K 124M 1% /dev/shm
Uptime showed the VPS is very lightly used:
uptime 20:49:58 up 8:56, 1 user, load average: 0.00, 0.00, 0.00
which is shown by the output of vmstat as well:
vmstat procs -----------memory---------- ---swap-- -----io---- -system-- ----cpu---- r b swpd free buff cache si so bi bo in cs us sy id wa 0 0 484 50784 10312 152076 0 0 2 10 13 9 1 0 98 1
the Inode was set to pretty good values as well, which is something I like:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/xvda1 1310720 67927 1242793 6% / tmpfs 31711 4 31707 1% /lib/init/rw udev 29106 390 28716 2% /dev tmpfs 31711 3 31708 1% /dev/shm
CPUinfo shows that there are four cores available to the VPS, none of which are throttled, although for some interesting reason, the model of the CPU was not displayed properly:
cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Genuine Intel(R) CPU 000 @ 2.27GHz stepping : 0 cpu MHz : 2266.746 cache size : 12288 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu tsc msr pae cx8 cmov pat clflush mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est ssse3 sse4_1 sse4_2 popcnt hypervisor ida arat tpr_shadow vnmi flexpriority ept vpid bogomips : 4533.49 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Genuine Intel(R) CPU 000 @ 2.27GHz stepping : 0 cpu MHz : 2266.746 cache size : 12288 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu tsc msr pae cx8 cmov pat clflush mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est ssse3 sse4_1 sse4_2 popcnt hypervisor ida arat tpr_shadow vnmi flexpriority ept vpid bogomips : 4533.49 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 2 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Genuine Intel(R) CPU 000 @ 2.27GHz stepping : 0 cpu MHz : 2266.746 cache size : 12288 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu tsc msr pae cx8 cmov pat clflush mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est ssse3 sse4_1 sse4_2 popcnt hypervisor ida arat tpr_shadow vnmi flexpriority ept vpid bogomips : 4533.49 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: processor : 3 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Genuine Intel(R) CPU 000 @ 2.27GHz stepping : 0 cpu MHz : 2266.746 cache size : 12288 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 11 wp : yes flags : fpu tsc msr pae cx8 cmov pat clflush mmx fxsr sse sse2 ss ht nx constant_tsc nonstop_tsc aperfmperf pni vmx est ssse3 sse4_1 sse4_2 popcnt hypervisor ida arat tpr_shadow vnmi flexpriority ept vpid bogomips : 4533.49 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management:
Meminfo does not seems to show any overselling:
cat /proc/meminfo MemTotal: 253692 kB MemFree: 8964 kB Buffers: 8204 kB Cached: 201352 kB SwapCached: 132 kB Active: 42892 kB Inactive: 187408 kB Active(anon): 9320 kB Inactive(anon): 11472 kB Active(file): 33572 kB Inactive(file): 175936 kB Unevictable: 0 kB Mlocked: 0 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 253692 kB LowFree: 8964 kB SwapTotal: 262136 kB SwapFree: 261652 kB Dirty: 108 kB Writeback: 0 kB AnonPages: 20632 kB Mapped: 5448 kB Shmem: 36 kB Slab: 10048 kB SReclaimable: 6772 kB SUnreclaim: 3276 kB KernelStack: 680 kB PageTables: 812 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 388980 kB Committed_AS: 101364 kB VmallocTotal: 598008 kB VmallocUsed: 840 kB VmallocChunk: 597020 kB HardwareCorrupted: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 262144 kB DirectMap2M: 0 kB
Although I did not run vzfree since this is a Xen-based system.
Time sync results looks pretty OK too:
time sync real 0m0.153s user 0m0.004s sys 0m0.000s
The really weird part comes when running the dd tests to see how the I/O is like:
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, 124.466 s, 8.6 MB/s
As you can see, the dd output seems rather low.
Since it is an otherwise pretty decent VPS, I was surprised to see this results and quickly ran another dd test, which showed similar 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, 147.529 s, 7.3 MB/s
I thought it was just a random abuser on the node, so I decided to wait for a day and re-run the dd test, the dd ran better but is nonetheless very low:
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, 94.5905 s, 11.4 MB/s
As with many of the European VPS providers, network is something that really stand out, I was able to get more than 50MB/s on the download test from Cachefly:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2011-09-25 21:11:18-- 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 51.9M/s in 1.9s 2011-09-25 21:11:20 (51.9 MB/s) - `/dev/null' saved [104857600/104857600]
Upload, on the other hand, is not that great, but still decent. First is the results from my Nix Communications VPS from Montreal, Canada:
wget 91.227.124.118/100mb.test -O /dev/null --2011-09-26 08:23:12-- http://91.227.124.118/100mb.test Connecting to 91.227.124.118:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 912K/s in 1m 56s
This seems to be rather low, however given that many of my uploading tests using this VPS has consistently showing really disappointing results, perhaps the issue is not with the QualityServers’ side.
Next is the result from my BuyVM VPS in San Jose, CA:
wget 91.227.124.118/100mb.test -O /dev/null --2011-09-25 20:18:43-- http://91.227.124.118/100mb.test Connecting to 91.227.124.118:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 1.30M/s in 74s 2011-09-25 20:19:57 (1.36 MB/s) - `/dev/null' saved [104857600/104857600]
As you can see, this time the result is a lot better.
Finally, my QuickWeb VPS in London, UK:
wget 91.227.124.118/100mb.test -O /dev/null --2011-09-25 20:17:20-- http://91.227.124.118/100mb.test Connecting to 91.227.124.118:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 4.38M/s in 19s 2011-09-25 20:17:39 (5.29 MB/s) - `/dev/null' saved [104857600/104857600]
Being geographically the closest, there is little wonder this is the best results among the three tests.
Even though the disk I/O is almost terrible (at least from what I can see in the dd test), the benchmarks are not too bad – that is, if it runs, for some reason, initially when I ran UnixBench, I had a few errors:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
4 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
4 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
4 x Execl Throughput 1 2 3
4 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
4 x File Copy 256 bufsize 500 maxblocks 1 2 3
4 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
4 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
4 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
4 x Process Creation 1 2 3
4 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
4 x Shell Scripts (1 concurrent) 1 2
**********************************************
Run: "Shell Scripts (1 concurrent)": /root/UnixBench/pgms/tst.sh: 18: srt: not found; aborting
Tried again and it went one test further:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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
4 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
4 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
4 x Execl Throughput 1 2 3
4 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
4 x File Copy 256 bufsize 500 maxblocks 1 2 3
4 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
4 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
4 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
4 x Process Creation 1 2 3
4 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
4 x Shell Scripts (1 concurrent) 1 2 3
4 x Shell Scripts (8 concurrent) 1 2 3
**********************************************
Run: "Shell Scripts (8 concurrent)": /root/UnixBench/pgms/tst.sh: 19: grp: not found; aborting
However, being a stubborn person, I gave this VPS yet another chance and finally it managed to run through and give me a 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
4 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
4 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
4 x Execl Throughput 1 2 3
4 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
4 x File Copy 256 bufsize 500 maxblocks 1 2 3
4 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
4 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
4 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
4 x Process Creation 1 2 3
4 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
4 x Shell Scripts (1 concurrent) 1 2 3
4 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: bob.infinitymedia.org.uk: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-686-bigmem -- #1 SMP Wed Jan 12 04:40:25 UTC 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Genuine Intel(R) CPU 000 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, MMX, Physical Address Ext, Intel virtualization
CPU 1: Genuine Intel(R) CPU 000 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, MMX, Physical Address Ext, Intel virtualization
CPU 2: Genuine Intel(R) CPU 000 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, MMX, Physical Address Ext, Intel virtualization
CPU 3: Genuine Intel(R) CPU 000 @ 2.27GHz (4533.5 bogomips)
Hyper-Threading, MMX, Physical Address Ext, Intel virtualization
06:56:49 up 19:03, 2 users, load average: 0.00, 0.00, 0.00; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Mon Sep 26 2011 06:56:49 - 07:25:04
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 9665509.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 1755.5 MWIPS (9.9 s, 7 samples)
Execl Throughput 1282.8 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 322124.4 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 88341.6 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 800716.5 KBps (30.0 s, 2 samples)
Pipe Throughput 538702.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 64481.7 lps (10.0 s, 7 samples)
Process Creation 2597.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2921.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 901.3 lpm (60.0 s, 2 samples)
System Call Overhead 379676.9 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 9665509.8 828.2
Double-Precision Whetstone 55.0 1755.5 319.2
Execl Throughput 43.0 1282.8 298.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 322124.4 813.4
File Copy 256 bufsize 500 maxblocks 1655.0 88341.6 533.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 800716.5 1380.5
Pipe Throughput 12440.0 538702.1 433.0
Pipe-based Context Switching 4000.0 64481.7 161.2
Process Creation 126.0 2597.5 206.1
Shell Scripts (1 concurrent) 42.4 2921.6 689.1
Shell Scripts (8 concurrent) 6.0 901.3 1502.1
System Call Overhead 15000.0 379676.9 253.1
========
System Benchmarks Index Score 487.1
------------------------------------------------------------------------
Benchmark Run: Mon Sep 26 2011 07:25:04 - 07:53:25
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 44579134.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8489.2 MWIPS (10.6 s, 7 samples)
Execl Throughput 3766.3 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 407964.3 KBps (30.2 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 104805.3 KBps (30.4 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1182506.3 KBps (30.0 s, 2 samples)
Pipe Throughput 2436391.7 lps (10.0 s, 7 samples)
Pipe-based Context Switching 197604.5 lps (10.0 s, 7 samples)
Process Creation 4712.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 7816.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1120.5 lpm (60.1 s, 2 samples)
System Call Overhead 1624944.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 44579134.3 3820.0
Double-Precision Whetstone 55.0 8489.2 1543.5
Execl Throughput 43.0 3766.3 875.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 407964.3 1030.2
File Copy 256 bufsize 500 maxblocks 1655.0 104805.3 633.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 1182506.3 2038.8
Pipe Throughput 12440.0 2436391.7 1958.5
Pipe-based Context Switching 4000.0 197604.5 494.0
Process Creation 126.0 4712.2 374.0
Shell Scripts (1 concurrent) 42.4 7816.6 1843.5
Shell Scripts (8 concurrent) 6.0 1120.5 1867.5
System Call Overhead 15000.0 1624944.4 1083.3
========
System Benchmarks Index Score 1203.9
For a small VPS like this, I would say the score is pretty OK, however I am curious as to why it failed several times before finally managed to run.
Next, the GeekBench scores, which, failed as well with GeekBench 2.2. Seriously, I am thinking for some reason, v2.2 is really VPS-unfriendly and I think most of the VPS that I have reviewed for the past few weeks have failed with GeekBench 2.2.
Geekbench 2.2.0 : http://www.primatelabs.ca/geekbench/
Geekbench is in tryout mode
Geekbench is limited to running 32-bit benchmarks while in tryout mode.
Please purchase Geekbench to remove all limitations found in tryout mode.
If you would like to purchase Geekbench you can do so online:
http://store.primatelabs.ca/
If you have already purchased Geekbench, enter your email address and license
key from your email receipt with the following command line:
dist/Geekbench-2.2.0-Linux/geekbench_x86_32 -r
Run Geekbench as root to gather accurate system information.
Floating point exception
However, I did have some luck using the older version 2.1 though, and the tests actually produced some decent scores:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.32-5-686-bigmem i686
Model: Linux PC (Genuine Intel(R) CPU 000 @ 2.27GHz)
Motherboard: Unknown Motherboard
Processor: Genuine Intel(R) CPU 000 @ 2.27GHz
Processor ID: GenuineIntel Family 6 Model 44 Stepping 0
Logical Processors: 4
Physical Processors: 1
Processor Frequency: 2.27 GHz
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: 248 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Genuine Intel(R) CPU 000 @ 2.27GHz
Processor Cores: 4
Integer
Blowfish
single-threaded scalar 1101 ||||
multi-threaded scalar 5289 |||||||||||||||||||||
Text Compress
single-threaded scalar 1459 |||||
multi-threaded scalar 6252 |||||||||||||||||||||||||
Text Decompress
single-threaded scalar 1761 |||||||
multi-threaded scalar 6873 |||||||||||||||||||||||||||
Image Compress
single-threaded scalar 1469 |||||
multi-threaded scalar 3983 |||||||||||||||
Image Decompress
single-threaded scalar 1099 ||||
multi-threaded scalar 3343 |||||||||||||
Lua
single-threaded scalar 1969 |||||||
multi-threaded scalar 5891 |||||||||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1804 |||||||
multi-threaded scalar 8137 ||||||||||||||||||||||||||||||||
Dot Product
single-threaded scalar 2904 |||||||||||
multi-threaded scalar 13381 |||||||||||||||||||||||||||||||||||||||||||||||||||||
single-threaded vector 3903 |||||||||||||||
multi-threaded vector 11872 |||||||||||||||||||||||||||||||||||||||||||||||
LU Decomposition
single-threaded scalar 2283 |||||||||
multi-threaded scalar 6101 ||||||||||||||||||||||||
Primality Test
single-threaded scalar 2651 ||||||||||
multi-threaded scalar 9277 |||||||||||||||||||||||||||||||||||||
Sharpen Image
single-threaded scalar 5357 |||||||||||||||||||||
multi-threaded scalar 14390 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 4877 |||||||||||||||||||
multi-threaded scalar 16961 |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 3585 ||||||||||||||
Write Sequential
single-threaded scalar 6164 ||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3571 ||||||||||||||
Stdlib Write
single-threaded scalar 2928 |||||||||||
Stdlib Copy
single-threaded scalar 6482 |||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 4498 |||||||||||||||||
single-threaded vector 5578 ||||||||||||||||||||||
Stream Scale
single-threaded scalar 2751 |||||||||||
single-threaded vector 3781 |||||||||||||||
Stream Add
single-threaded scalar 2976 |||||||||||
single-threaded vector 4427 |||||||||||||||||
Stream Triad
single-threaded scalar 4320 |||||||||||||||||
single-threaded vector 2618 ||||||||||
Integer Score: 3374 |||||||||||||
Floating Point Score: 7421 |||||||||||||||||||||||||||||
Memory Score: 4546 ||||||||||||||||||
Stream Score: 3868 |||||||||||||||
Overall Geekbench Score: 5074 ||||||||||||||||||||
Trying again showed pretty decent score too:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.32-5-686-bigmem i686
Model: Linux PC (Genuine Intel(R) CPU 000 @ 2.27GHz)
Motherboard: Unknown Motherboard
Processor: Genuine Intel(R) CPU 000 @ 2.27GHz
Processor ID: GenuineIntel Family 6 Model 44 Stepping 0
Logical Processors: 4
Physical Processors: 1
Processor Frequency: 2.27 GHz
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: 248 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Genuine Intel(R) CPU 000 @ 2.27GHz
Processor Cores: 4
Integer
Blowfish
single-threaded scalar 1511 ||||||
multi-threaded scalar 7015 ||||||||||||||||||||||||||||
Text Compress
single-threaded scalar 1540 ||||||
multi-threaded scalar 5901 |||||||||||||||||||||||
Text Decompress
single-threaded scalar 1364 |||||
multi-threaded scalar 5844 |||||||||||||||||||||||
Image Compress
single-threaded scalar 1407 |||||
multi-threaded scalar 3998 |||||||||||||||
Image Decompress
single-threaded scalar 847 |||
multi-threaded scalar 3141 ||||||||||||
Lua
single-threaded scalar 2115 ||||||||
multi-threaded scalar 5906 |||||||||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1331 |||||
multi-threaded scalar 5862 |||||||||||||||||||||||
Dot Product
single-threaded scalar 2580 ||||||||||
multi-threaded scalar 9001 ||||||||||||||||||||||||||||||||||||
single-threaded vector 2565 ||||||||||
multi-threaded vector 11263 |||||||||||||||||||||||||||||||||||||||||||||
LU Decomposition
single-threaded scalar 1506 ||||||
multi-threaded scalar 6059 ||||||||||||||||||||||||
Primality Test
single-threaded scalar 2142 ||||||||
multi-threaded scalar 6705 ||||||||||||||||||||||||||
Sharpen Image
single-threaded scalar 3647 ||||||||||||||
multi-threaded scalar 13317 |||||||||||||||||||||||||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 3326 |||||||||||||
multi-threaded scalar 12633 ||||||||||||||||||||||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 3465 |||||||||||||
Write Sequential
single-threaded scalar 5515 ||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 2296 |||||||||
Stdlib Write
single-threaded scalar 2783 |||||||||||
Stdlib Copy
single-threaded scalar 5580 ||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 3559 ||||||||||||||
single-threaded vector 4711 ||||||||||||||||||
Stream Scale
single-threaded scalar 2485 |||||||||
single-threaded vector 4922 |||||||||||||||||||
Stream Add
single-threaded scalar 3341 |||||||||||||
single-threaded vector 4436 |||||||||||||||||
Stream Triad
single-threaded scalar 3076 ||||||||||||
single-threaded vector 3635 ||||||||||||||
Integer Score: 3382 |||||||||||||
Floating Point Score: 5852 |||||||||||||||||||||||
Memory Score: 3927 |||||||||||||||
Stream Score: 3770 |||||||||||||||
Overall Geekbench Score: 4394 |||||||||||||||||
It really surprised me, especially knowing that UnixBench and GeekBench sometime penalize heavily if a VPS has a low disk I/O and yet, despite the low disk I/O, both benchmark scores are pretty good for such a small VPS. I was thinking if there is any chance such issue was caused by the time issue in the 2.6.18 kernel, but it does not seem to be the case.
Customer Service and Support
As per my previous review on Quality Servers, Kayako Helpdesk continued to be integrated with the WHMCS system and when the user clicks on Support in the WHMCS system, they would be redirected to the Kayako Helpdesk. However, the WHMCS account does not work by default in the Kayako system. You will be assigned a new password when you try to create a ticket without sign in for the first time, based on your email address. After that, you will have to sign in to see the ticket replies or create a new ticket.
Same with the last time, QualityServers continued to have quite a few departments to handle different types of tickets:
One pretty annoying feature here is that you will have to supply IP addresses and root password when submitting the tickets, although I guess you can always just type in some random numbers for IP and root password:
Once the ticket is created, you can log back in to see the current status of the ticket as well as providing feedbacks to the satisfaction levels.
James seems to be the only one responding to all of the technical tickets, and I think sometimes it could take a while for him to reply although I think for a budget box, it is well within the acceptable range:
The first ticket was for an rDNS request, which was sent in at 02:47AM and did not get a reply until 03:09AM, about 12 hours later (they are all in UK time I believe). Note: I have received the permission from Mr. A to publish the screenshots of the tickets here.
The next one is a more bizarre one, after noticed that the disk I/O is somewhat low, a ticket was sent to James, and this time he replied relatively fast, within 3 minute:
And from the link he sent, here is what I can see:
I have to agree with him on that since the benchmarks score is pretty good, therefore, I decided to ask James what would be a good way to test the VPS disk I/O and to my surprise (again), here is the response:
OK, so we do not care about the disk I/O as long as it is good enough for “normal use”. Fair enough, how about ioping? Is that another bad way to test the disk I/O?
Well, I guess not, but I would be surprised if I am getting a dd output of less than 10MB/s because people from LEB are trying to trash his servers.
In the end, I never get a concrete solution/explanation to the low I/O, and basically have to have the faith that the VPS is fast even though the dd tests and iopings are low.
Conclusion
There are indeed many great aspects of the QualityServers VPS, for example, network is fast, ticket response is acceptable and for a low end VPS, the benchmarks are actually pretty decent. However, the disk I/O is pretty disappointing, at least from the output of the dd tests, which James could not care less about. Although I have to agree with James that the VPS is relatively fast and I did not have to wait for 10 minutes just to log in via SSH (which happened to some of the VPS showing comparable dd results), the fact that I have no clue why it is so is something really puzzled me and would prevent me from considering using it as a production-grade VPS despite the cheap price.
Yep, that mysterious Mr. A is me 🙂 Mr. H to be precist 😛
“but I figure I should try to avoid of getting myself into more trouble” <– Nah, I'm not that kinda person!
@Humza: All right, I can confirm Mr. A is Mr. H 🙂 Thanks a lot for lending me the VPS for the review 🙂
No worries 🙂
Pingback: 96MB Low End VPS Review Part XXXII – EDIS | | 96MB.com96MB.com