96MB Low End VPS Review Part 57–Atomic VPS
Recently, there has been quite a few cloud VPS hosting appeared on the 96MB Low End VPS reviews, and the review today is no exception. Atomic VPS is probably one of the “high-end” VPS providers as well, with their regular price start at 59.95USD per month for a 1GB RAM VPS. At this price level, how would the performance be? Let’s take a look.
Basic Information and Set Up
I have signed up for a cloudONE package during their open beta test stage for this particular review when I saw their posting on WebhostingTalk and here is the package specifications on their plan page:
At the moment they are running a promotion for 14.95 at the first month, which is really out of the “low end” VPS range as well:
The sign up page is pretty standard WHMCS page and in fact I was asked to enter my domain name as well, which is really something I hope they could get rid of.
You have quite a few options when it comes to operating systems:
You can even choose Windows 2008 Web Edition and CloudLinux if you are willing to pay a little extra.
For the control panel, there seems to be no discount and the price is pretty much the same as what you pay from any external vendor:
You can also get ClientExec for 15USD per month and R1Soft backup license for 10USD per month, which is not exactly cheap. I think AtomicVPS is not exactly catered towards the budget-conscious customers and I can only imagine a few dollars won’t make too much differences to those who are signing up for their services.
I am not sure if it was due to the fact that I signed up for the service during the “beta” stage but the provision of this VPS was not instant. My invoice payment confirmation has arrived at 8:20AM but I have not received the server details until 9:36AM.
The control panel is on standard HTTPS port and unlike most of the VPS providers who use SolusVM as the control panel, AtomicVPS uses the OnApp control panel, which is integrated with the WHMCS application (which they have nick-named the “Reactor” to echo their brand name).
Once logged in, the summary of the VPS is shown on the main page, as well as some basic actions to control the VPS. As you can see, the CPU priority is 20% and the port speed is only 10Mbps, which is quite slow:
Further down the page, you can actually see a log of all the activities you have performed on the VPS:
The next tab shows the hourly CPU usage graph, I prefer this kind of time series graph a lot more than just a number (as we see in some other control panels). To me, knowing how much the CPU has been used at different points of the day is definitely a lot more useful than just having a simple number:
The next tab showed two IP addresses that my VPS was assigned to, as well as the basic configuration information. Unfortunately, there is no place for you to set rDNS nor some control functionality for you to change the IP addresses easily:
Disks tab, as the name suggests, shows you the hard disk usage, there is also an automatic backup feature for the VPS, however I could not figure out how to turn it on as it is probably sold separately as an add-on.
Finally, for the backups tab, you can literally back up the entire hard drive with the click of one button, which is pretty good:
Test on the VPS
As mentioned above, the VPS comes with 0.5GHz of CPU, 1GB of RAM and 25GB of hard drive, it is located in Clifton, NJ in the FortressItx DC.
When the OS was first loaded, about 35MB of RAM was used:
free -m
total used free shared buffers cached
Mem: 1020 137 882 0 1 99
-/+ buffers/cache: 35 984
Swap: 0 0 0
Top output showing the processes running:
top - 02:26:16 up 24 min, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 50 total, 1 running, 49 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: 1045140k total, 141128k used, 904012k free, 1996k buffers
Swap: 0k total, 0k used, 0k free, 102476k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
832 snmp 20 0 8844 4072 2624 S 0.0 0.4 0:00.22 snmpd
994 root 20 0 5672 2972 1480 S 0.0 0.3 0:00.17 bash
991 root 20 0 8280 2868 2348 S 0.0 0.3 0:00.03 sshd
782 ntp 20 0 4684 1932 1508 S 0.0 0.2 0:00.08 ntpd
621 root 20 0 27336 1472 1036 S 0.0 0.1 0:00.02 rsyslogd
1028 root 20 0 2356 1116 896 R 0.0 0.1 0:00.00 top
911 root 20 0 5516 984 596 S 0.0 0.1 0:00.00 sshd
966 Debian-e 20 0 6536 948 632 S 0.0 0.1 0:00.00 exim4
474 statd 20 0 1960 792 664 S 0.0 0.1 0:00.00 rpc.statd
720 root 20 0 3808 788 620 S 0.0 0.1 0:00.00 cron
156 root 16 -4 2284 732 404 S 0.0 0.1 0:00.02 udevd
1 root 20 0 2056 724 624 S 0.0 0.1 0:00.14 init
192 root 18 -2 2280 648 320 S 0.0 0.1 0:00.00 udevd
200 root 18 -2 2280 648 320 S 0.0 0.1 0:00.00 udevd
663 root 20 0 1728 604 488 S 0.0 0.1 0:00.00 acpid
985 root 20 0 1728 556 476 S 0.0 0.1 0:00.00 getty
984 root 20 0 1728 552 476 S 0.0 0.1 0:00.00 getty
And htop output:
Interestingly, there is a process named snmped which seems to take quite a bit of memory, however it is definitely not one of the processes that I have commonly seen in Debian, turns out it is used to respond to SNMP requests, which I suspect has something to do with the way OnApp controls the individual VPS.
About 773MB of hard drive space was used, which was pretty standard for the VPS.
df -h Filesystem Size Used Avail Use% Mounted on /dev/xvda1 25G 773M 23G 4% / tmpfs 511M 0 511M 0% /lib/init/rw udev 493M 68K 493M 1% /dev tmpfs 511M 0 511M 0% /dev/shm
And the inodes are set to pretty standard values:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/xvda1 3276800 29089 3247711 1% / tmpfs 130642 5 130637 1% /lib/init/rw udev 126050 438 125612 1% /dev tmpfs 130642 1 130641 1% /dev/shm
Once the full LNMP stack was installed, the memory usage went all the way up to 96MB (which is perfect for this site!):
free -m
total used free shared buffers cached
Mem: 1020 1000 20 0 49 853
-/+ buffers/cache: 96 923
Swap: 0 0 0
Top showing the processes running:
top - 09:32:01 up 23:30, 1 user, load average: 0.00, 0.00, 0.00 Tasks: 60 total, 1 running, 59 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: 1045140k total, 1024960k used, 20180k free, 51184k buffers Swap: 0k total, 0k used, 0k free, 874372k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 2438 www 20 0 15012 11m 764 S 0.0 1.1 0:00.00 nginx 2418 mysql 20 0 33948 4644 2040 S 0.0 0.4 0:00.00 mysqld 2428 root 20 0 24132 4556 1412 S 0.0 0.4 0:11.28 php-cgi 2429 www 20 0 24132 4160 1016 S 0.0 0.4 0:00.00 php-cgi 2430 www 20 0 24132 4160 1016 S 0.0 0.4 0:00.00 php-cgi 2433 www 20 0 24132 4160 1016 S 0.0 0.4 0:00.00 php-cgi 2434 www 20 0 24132 4160 1016 S 0.0 0.4 0:00.00 php-cgi 2435 www 20 0 24132 4160 1016 S 0.0 0.4 0:00.00 php-cgi 832 snmp 20 0 8844 4072 2624 S 0.0 0.4 0:12.31 snmpd 2993 root 20 0 6096 3404 1468 S 0.0 0.3 0:00.18 bash 2989 root 20 0 8268 2880 2364 S 0.0 0.3 0:00.07 sshd 782 ntp 20 0 4684 1936 1508 S 0.0 0.2 0:04.23 ntpd 621 root 20 0 27336 1492 1036 S 0.0 0.1 0:00.03 rsyslogd 3014 root 20 0 2344 1132 904 R 0.0 0.1 0:00.00 top 911 root 20 0 5516 984 596 S 0.0 0.1 0:00.02 sshd 966 Debian-e 20 0 6536 952 636 S 0.0 0.1 0:00.02 exim4 474 statd 20 0 1960 792 664 S 0.0 0.1 0:00.00 rpc.statd
And the htop output ranked by memory usage:
Approximately 2GB of hard drive space was used to complete the installation:
df -h Filesystem Size Used Avail Use% Mounted on /dev/xvda1 25G 2.0G 22G 9% / tmpfs 511M 0 511M 0% /lib/init/rw udev 493M 68K 493M 1% /dev tmpfs 511M 0 511M 0% /dev/shm
And the inodes are set to pretty standard values:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/xvda1 3276800 77790 3199010 3% / tmpfs 130642 5 130637 1% /lib/init/rw udev 126050 438 125612 1% /dev tmpfs 130642 1 130641 1% /dev/shm
Vmstat shows the VPS is mostly idle with 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 0 19824 51204 874472 0 0 1 15 14 22 1 0 99 0
Uptime shows similar information:
uptime 09:33:40 up 23:31, 1 user, load average: 0.00, 0.00, 0.00
CPUInfo shows one E5620 core assigned to this VPS and is not throttled:
cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 44 model name : Intel(R) Xeon(R) CPU E5620 @ 2.40GHz stepping : 2 cpu MHz : 2399.998 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 de tsc msr pae cx8 sep cmov pat clflush mmx fxsr sse sse2 ss ht nx constant_tsc up nonstop_tsc pni ssse3 sse4_1 sse4_2 popcnt hypervisor bogomips : 4799.99 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management:
And meminfo:
cat /proc/meminfo MemTotal: 1045140 kB MemFree: 21668 kB Buffers: 26220 kB Cached: 918348 kB SwapCached: 0 kB Active: 92172 kB Inactive: 877988 kB Active(anon): 11404 kB Inactive(anon): 14264 kB Active(file): 80768 kB Inactive(file): 863724 kB Unevictable: 0 kB Mlocked: 0 kB HighTotal: 311304 kB HighFree: 624 kB LowTotal: 733836 kB LowFree: 21044 kB SwapTotal: 0 kB SwapFree: 0 kB Dirty: 0 kB Writeback: 0 kB AnonPages: 25588 kB Mapped: 10780 kB Shmem: 80 kB Slab: 28268 kB SReclaimable: 25600 kB SUnreclaim: 2668 kB KernelStack: 496 kB PageTables: 988 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 522568 kB Committed_AS: 109080 kB VmallocTotal: 122880 kB VmallocUsed: 2752 kB VmallocChunk: 118780 kB HardwareCorrupted: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 2048 kB DirectMap4k: 737272 kB DirectMap2M: 0 kB
As well as time sync:
time sync real 0m0.021s user 0m0.004s sys 0m0.000s
As with any other cloud providers whose storage device is not on the same physical node, disk I/O from Atomic VPS did not impress me too much:
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, 14.0765 s, 76.3 MB/s
Testing again showed slightly worse 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, 15.3326 s, 70.0 MB/s
Fortunately, ioping is more or less stable:
ioping -c 10 . 4096 bytes from . (ext3 /dev/xvda1): request=1 time=0.3 ms 4096 bytes from . (ext3 /dev/xvda1): request=2 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=3 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=4 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=5 time=0.5 ms 4096 bytes from . (ext3 /dev/xvda1): request=6 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=7 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=8 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=9 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=10 time=0.4 ms --- . (ext3 /dev/xvda1) ioping statistics --- 10 requests completed in 9005.5 ms, 2513 iops, 9.8 mb/s min/avg/max/mdev = 0.3/0.4/0.5/0.0 ms
Testing again showed slightly worse results, but is still pretty good:
ioping -c 10 . 4096 bytes from . (ext3 /dev/xvda1): request=1 time=0.3 ms 4096 bytes from . (ext3 /dev/xvda1): request=2 time=0.9 ms 4096 bytes from . (ext3 /dev/xvda1): request=3 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=4 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=5 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=6 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=7 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=8 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=9 time=0.4 ms 4096 bytes from . (ext3 /dev/xvda1): request=10 time=1.6 ms --- . (ext3 /dev/xvda1) ioping statistics --- 10 requests completed in 9007.0 ms, 1782 iops, 7.0 mb/s min/avg/max/mdev = 0.3/0.6/1.6/0.4 ms
Network speed is normally a strength of some of these “cloud” providers that I have reviewed before, unfortunately this does not seem to be the case with Atomic VPS. However, given the fact that the port speed is only 10mbps, the file downloaded from Cachefly is actually close to the port speed:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2012-07-20 09:37:54-- 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: a€?/dev/nulla€ 100%[======================================>] 104,857,600 1.35M/s in 73s 2012-07-20 09:39:07 (1.36 MB/s) - a€?/dev/nulla€
Trying again had exactly the same results, I think it really seems to me that the port speed is capped but is dedicated:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2012-07-20 09:41:14-- 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: a€?/dev/nulla€ 100%[======================================>] 104,857,600 1.35M/s in 73s 2012-07-20 09:42:27 (1.36 MB/s) - a€?/dev/nulla€
The upload tests yields pretty much the same speed as the download tests, which suggests that the port is fully symetrical. First, the test file was downloaded from this VPS to my Quickweb VPS in Chicago, IL:
wget 69.72.xxx.xxx/100mb.test -O /dev/null --2012-07-20 10:52:39-- http://69.72.xxx.xxx/100mb.test Connecting to 69.72.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 1.12M/s in 90s 2012-07-20 10:54:09 (1.11 MB/s) - `/dev/null' saved [104857600/104857600]
This is actually the test that yielded the fastest speed as the physical location between Chicago and Clifton is the closest.
Next is the test VPS located in Los Angeles, CA, which is along the west coast.
wget 69.72.xxx.xxx/100mb.test -O /dev/null --2012-07-21 13:07:58-- http://69.72.xxx.xxx/100mb.test Connecting to 69.72.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 859K/s in 2m 1s 2012-07-21 13:09:59 (849 KB/s) - `/dev/null' saved [104857600/104857600]
And finally, with my XenVZ VPS in Maidenhead, UK:
wget 69.72.xxx.xxx/100mb.test -O /dev/null --2012-07-22 01:28:42-- http://69.72.xxx.xxx/100mb.test Connecting to 69.72.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 942K/s in 1m 53s 2012-07-22 01:30:35 (906 KB/s) - `/dev/null' saved [104857600/104857600]
For the Unixbench, I do not think I could realistically expect a lot out of this VPS since although the RAM amount is quite generous, CPU is actually very limited, and in the end, it is indeed the case when the Unixbench score is just a little more than 400 points:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: xxxxxx: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-xen-686 -- #1 SMP Mon Mar 26 10:32:49 UTC 2012
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.0 bogomips)
Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT
09:43:49 up 23:42, 1 user, load average: 0.08, 0.04, 0.01; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Fri Jul 20 2012 09:43:49 - 10:12:07
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 14979187.1 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2577.1 MWIPS (10.0 s, 7 samples)
Execl Throughput 1597.1 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 297705.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 49194.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 688827.4 KBps (30.0 s, 2 samples)
Pipe Throughput 104470.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 32702.4 lps (10.0 s, 7 samples)
Process Creation 2719.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2974.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 388.4 lpm (60.1 s, 2 samples)
System Call Overhead 675009.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 14979187.1 1283.6
Double-Precision Whetstone 55.0 2577.1 468.6
Execl Throughput 43.0 1597.1 371.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 297705.0 751.8
File Copy 256 bufsize 500 maxblocks 1655.0 49194.0 297.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 688827.4 1187.6
Pipe Throughput 12440.0 104470.8 84.0
Pipe-based Context Switching 4000.0 32702.4 81.8
Process Creation 126.0 2719.4 215.8
Shell Scripts (1 concurrent) 42.4 2974.6 701.6
Shell Scripts (8 concurrent) 6.0 388.4 647.4
System Call Overhead 15000.0 675009.7 450.0
========
System Benchmarks Index Score 402.3
Testing again yielded even lower results:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: xxxxxx: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-xen-686 -- #1 SMP Mon Mar 26 10:32:49 UTC 2012
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4800.0 bogomips)
Hyper-Threading, MMX, Physical Address Ext, SYSENTER/SYSEXIT
10:18:00 up 1 day, 16 min, 1 user, load average: 0.02, 1.28, 1.39; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Fri Jul 20 2012 10:18:00 - 10:46:17
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 14945471.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2571.6 MWIPS (10.1 s, 7 samples)
Execl Throughput 1591.7 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 186784.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 22771.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 674981.8 KBps (30.0 s, 2 samples)
Pipe Throughput 105373.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 32720.3 lps (10.0 s, 7 samples)
Process Creation 2794.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2923.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 391.8 lpm (60.1 s, 2 samples)
System Call Overhead 667181.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 14945471.3 1280.7
Double-Precision Whetstone 55.0 2571.6 467.6
Execl Throughput 43.0 1591.7 370.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 186784.6 471.7
File Copy 256 bufsize 500 maxblocks 1655.0 22771.5 137.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 674981.8 1163.8
Pipe Throughput 12440.0 105373.8 84.7
Pipe-based Context Switching 4000.0 32720.3 81.8
Process Creation 126.0 2794.2 221.8
Shell Scripts (1 concurrent) 42.4 2923.4 689.5
Shell Scripts (8 concurrent) 6.0 391.8 653.0
System Call Overhead 15000.0 667181.1 444.8
========
System Benchmarks Index Score 362.5
Geekbench score, is probably more relatively impressive in this case, at least is generally in line with many of the VPS that I have reviewed:
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-xen-686 i686
Model: Linux PC (Intel(R) Xeon(R) CPU E5620 @ 2.40GHz)
Motherboard: Unknown Motherboard
Processor: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
Processor ID: GenuineIntel Family 6 Model 44 Stepping 2
Logical Processors: 1
Physical Processors: 1
Processor Frequency: 2.40 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: 1021 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
Processor Cores: 1
Integer
Blowfish
single-threaded scalar 1774 |||||||
multi-threaded scalar 1890 |||||||
Text Compress
single-threaded scalar 1997 |||||||
multi-threaded scalar 1938 |||||||
Text Decompress
single-threaded scalar 2043 ||||||||
multi-threaded scalar 2040 ||||||||
Image Compress
single-threaded scalar 1672 ||||||
multi-threaded scalar 1635 ||||||
Image Decompress
single-threaded scalar 1446 |||||
multi-threaded scalar 1502 ||||||
Lua
single-threaded scalar 3110 ||||||||||||
multi-threaded scalar 3148 ||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2120 ||||||||
multi-threaded scalar 2155 ||||||||
Dot Product
single-threaded scalar 3452 |||||||||||||
multi-threaded scalar 3629 ||||||||||||||
single-threaded vector 4101 ||||||||||||||||
multi-threaded vector 4720 ||||||||||||||||||
LU Decomposition
single-threaded scalar 2389 |||||||||
multi-threaded scalar 2441 |||||||||
Primality Test
single-threaded scalar 3486 |||||||||||||
multi-threaded scalar 2790 |||||||||||
Sharpen Image
single-threaded scalar 5671 ||||||||||||||||||||||
multi-threaded scalar 5737 ||||||||||||||||||||||
Blur Image
single-threaded scalar 5279 |||||||||||||||||||||
multi-threaded scalar 5310 |||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 4874 |||||||||||||||||||
Write Sequential
single-threaded scalar 7615 ||||||||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3344 |||||||||||||
Stdlib Write
single-threaded scalar 3255 |||||||||||||
Stdlib Copy
single-threaded scalar 6305 |||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 5460 |||||||||||||||||||||
single-threaded vector 8228 ||||||||||||||||||||||||||||||||
Stream Scale
single-threaded scalar 4240 ||||||||||||||||
single-threaded vector 8681 ||||||||||||||||||||||||||||||||||
Stream Add
single-threaded scalar 5197 ||||||||||||||||||||
single-threaded vector 7873 |||||||||||||||||||||||||||||||
Stream Triad
single-threaded scalar 5550 ||||||||||||||||||||||
single-threaded vector 5696 ||||||||||||||||||||||
Integer Score: 2016 ||||||||
Floating Point Score: 3805 |||||||||||||||
Memory Score: 5078 ||||||||||||||||||||
Stream Score: 6365 |||||||||||||||||||||||||
Overall Geekbench Score: 3689 ||||||||||||||
Repeat the tests and the score is more or less identical:
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-xen-686 i686
Model: Linux PC (Intel(R) Xeon(R) CPU E5620 @ 2.40GHz)
Motherboard: Unknown Motherboard
Processor: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
Processor ID: GenuineIntel Family 6 Model 44 Stepping 2
Logical Processors: 1
Physical Processors: 1
Processor Frequency: 2.40 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: 1021 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz
Processor Cores: 1
Integer
Blowfish
single-threaded scalar 1778 |||||||
multi-threaded scalar 1886 |||||||
Text Compress
single-threaded scalar 1994 |||||||
multi-threaded scalar 1930 |||||||
Text Decompress
single-threaded scalar 2024 ||||||||
multi-threaded scalar 2083 ||||||||
Image Compress
single-threaded scalar 1673 ||||||
multi-threaded scalar 1639 ||||||
Image Decompress
single-threaded scalar 1391 |||||
multi-threaded scalar 1428 |||||
Lua
single-threaded scalar 3140 ||||||||||||
multi-threaded scalar 3140 ||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2098 ||||||||
multi-threaded scalar 2146 ||||||||
Dot Product
single-threaded scalar 3443 |||||||||||||
multi-threaded scalar 3631 ||||||||||||||
single-threaded vector 4097 ||||||||||||||||
multi-threaded vector 4726 ||||||||||||||||||
LU Decomposition
single-threaded scalar 2376 |||||||||
multi-threaded scalar 2416 |||||||||
Primality Test
single-threaded scalar 3472 |||||||||||||
multi-threaded scalar 2778 |||||||||||
Sharpen Image
single-threaded scalar 5386 |||||||||||||||||||||
multi-threaded scalar 5443 |||||||||||||||||||||
Blur Image
single-threaded scalar 5212 ||||||||||||||||||||
multi-threaded scalar 5237 ||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 5231 ||||||||||||||||||||
Write Sequential
single-threaded scalar 7664 ||||||||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3317 |||||||||||||
Stdlib Write
single-threaded scalar 3249 ||||||||||||
Stdlib Copy
single-threaded scalar 6182 ||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 5472 |||||||||||||||||||||
single-threaded vector 7929 |||||||||||||||||||||||||||||||
Stream Scale
single-threaded scalar 4175 ||||||||||||||||
single-threaded vector 7637 ||||||||||||||||||||||||||||||
Stream Add
single-threaded scalar 4138 ||||||||||||||||
single-threaded vector 6061 ||||||||||||||||||||||||
Stream Triad
single-threaded scalar 4349 |||||||||||||||||
single-threaded vector 5797 |||||||||||||||||||||||
Integer Score: 2008 ||||||||
Floating Point Score: 3747 ||||||||||||||
Memory Score: 5128 ||||||||||||||||||||
Stream Score: 5694 ||||||||||||||||||||||
Overall Geekbench Score: 3609 ||||||||||||||
Customer Service and Support
Compare to the VPS itself, I was a lot more impressed with the support from Atomic VPS. When I first signed up for their services, there was no Debian 6 x86 OS templates and I sent a ticket at 8:28AM, the ticket was responded at 8:33AM saying they are adding the templates and rather than just give me a Debian template, they actually sent me another note 11 minutes later asked me if I want a plain Debian template or one with Vyatta pre-installed, which I think is something great. Finally, by 9:24AM, the template was added and I happily loaded the OS template without any issue.
Conclusion
Quoted from John from Atomic VPS:
we’re really excited about our Cloud platform and we’ve spared no expense to make sure that the term “cloud” was actually functional instead of just being marketable. So that was our goal while working to get AtomicVPS launched.
As such, the performance of the smallest node, although is very expensive compare to the “low-end” VPS standard, is rather “average” at best. For most of the readers of 96MB.com, the service is rather unimpressive since we can probably get the budget services with better performance at less than a quarter of the cost. However, I can definitely see, with the professional and fast response from the client support team, that their service would appeal to customers who needs stability for mission-critical services.
do they know who you are when you submitted ticket? thats impressive response indeed!
@jcaleb Not at all, I will never contact a provider to identify myself prior to the review, in fact, that is why this website is whois protected, so that I can use my genuine personal information to sign up, review them, post the review, without having the providers know who I am at all in order to give the best indication of their services.
then that is indeed an impressive service. i had a similar experience with prometeus with their kvm, when i asked for additional iso be provided.
@jcaleb: Prometeus is normally pretty quick in replying but is never within a few minutes for me, I guess you got really lucky!
I PM’ed uncle in LET thats why. In tickets, its usually within 2 hours. But when they do answer, they also have solved your problem also. So it does not go back and forth.
@jcaleb: LOL, you probably just shared one of the best kept secrets on LET, now Uncle Sal’s going to get quite a few PMs every day!
On a more serious note, although what you have suggested is good, but I won’t take the response time for LET PM into account since not every provider is active on LET and I do not think providers should be punished in the review if they are not active on LET.