96MB Low End VPS Review Part XX – Burst.Net Miami OpenVZ VPS
Burst.Net has been probably the oldest and largest budget VPS providers in the world. With 20 years of history and customer all over the world, it is certainly a giant in this VPS industry. As such, when they offered a first month free deal on LowEndBox.com, I decided to sign up for one and see how it works.
General Information and Set Up
First, I need to clarify that the offer presented on LowEndBox is not exactly a free month, you will need to pay one dollar of set up fee, which, according to the comment here, the fee is strictly for fraud prevention only. In any case, to take advantage of this offer, you will have to go to this page here and then select one of the VPS for a free month, under the category of Dedicated Servers:
I signed up with the lowest VPS #1 Plan with vePortal and here is the specifications I get:
Unlike many other providers, Burst still gives out 2 IPv4 addresses for each VPS.
Sign up page appears to be a highly customized WHMCS on secured HTTP login page. One interesting thing to note is that upon payment, you are asked to sign and fax a Paypal Authorization Form authorizing the payment made by Paypal, which I did not sign but received the VPS in any case.
The activation was not instant but was really close (Paypal confirmation came at 4:18PM and VPS information was received at 4:24PM). There was a lot of information in the welcome email, including your root username and password in plain text (which, unlike most of the other providers, you can not specify your root password during sign up), and the login details to vePortal. Strange enough, their vePortal’s URL is an IP address rather than FQDN, which is rather unprofessional.
Once login to the client account, you get a very detailed description of the product under My Products and Services:
However, unlike most of the VPS providers, Burst actually does not offer any control on WHMCS and all controls on VPS have to be done by logging in to the vePortal using the user name and password assigned.
After adding in two security exceptions due to the SSL certificate issues (since IP addresses were used instead of FQDN), you get logged into the vePortal page:
I have to say seeing vePortal for the first time makes me wondering why there were not more hosts using it. The graphical interface is really nice with all information on one page and most of the basic controls are just a click away as well.
There were quite a few OS templates available under Reload OS, unfortunately none of them stated that if the template is minimal or standard OS template:
Going through the tabs in vePortal, on the Backup Functions tab, you could create and restore and download backups:
Interestingly, although I did not sign up for any backup service, I did not receive any error message when initiate a backup. However, I did not get any backup file in the end, so I assume that is how Burst make you to pay for backups.
Under Info and Security, you can change root password and even enable TUN/TAP and enable the API automatically, you can also choose to secure directories which will make you lose /var/tmp and /tmp every time you reboot the VPS:
On Networking and Utilities tab, you could do things like changing host name, even doing ping tests and DNS Lookup with the external sites. There is also the functionality to enable veRemote, which is perhaps some sort of desktop VNC program:
Unfortunately, there is no IPv6 address included in the product and no instant rDNS available as well. UPDATE: According to Burst’s customer support, they could give up to 20 IPv6 address to each VPS for free.
Software Installation tab solved the mystery why Burst does not need the that many OS templates, because all the control panels could be set up automatically:
There is also a vePortal Exapps, which made me curious, but it really does not have anything when I clicked there.
Finally, the Preferences tab allows you to change the style and language:
Overall, I have to say that vePortal seems to be a lot nicer and more powerful than SolusVM, and other than set up rDNS and request more IP addresses (including IPv6 addresses), most of the common tasks could be done in the vePortal automatically, which is great.
Another area worth mentioning is their ticket system, which is separated from the WHMCS system and seems to use PerlDesk (although Firefox kept giving me warnings about PerlDesk when I tried to enter what appears to be their product page, so I did not include the link here). The interface does look a bit old-fashioned (last time PerlDesk seems to update their homepage was in 2007) but is nonetheless quite easy to use:
One big advantage I would like to mention about Burst’s helpdesk software is the ability to reply the ticket directly via email. Although I do not think it is a very difficult task, few VPS providers actually realise how useful it is to enable the functionalities to reply tickets directly via email so that the customers do not have to find the helpdesk URL, enter user name and password and then hit reply tickets, which could be frustrating once a while, particularly when the VPS is having a major issue.
Tests on the VPS
As mentioned above, the VPS that I have received has 512MB of RAM and 20 GB of hard drive space, with OpenVZ virtualization. The VPS is in Miami, Florida and I have put up Debian 6 32 bit for testing.
When the OS was initially loaded, it took about 315MB of memory, which is definitely really high:
free -m
total used free shared buffers cached
Mem: 512 315 196 0 0 0
-/+ buffers/cache: 315 196
Swap: 0 0 0
Top output shows that the only memory monster that was running was Apache2:
top - 15:45:43 up 1 day, 19:19, 1 user, load average: 0.09, 0.03, 0.01
Tasks: 12 total, 2 running, 10 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: 524800k total, 323812k used, 200988k free, 0k buffers
Swap: 0k total, 0k used, 0k free, 0k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
32530 www-data 19 0 275m 2900 1376 S 0.0 0.6 0:00.00 apache2
6015 root 18 0 8224 2752 2244 R 0.0 0.5 0:00.07 sshd
32515 root 18 0 5520 2740 1444 S 0.0 0.5 0:05.34 apache2
6018 root 15 0 4012 2712 1332 S 0.0 0.5 0:00.16 bash
32529 www-data 19 0 5292 1852 560 S 0.0 0.4 0:00.00 apache2
27774 root 17 0 33436 1408 1044 S 0.0 0.3 0:00.12 rsyslogd
6035 root 15 0 2324 1060 872 R 0.0 0.2 0:00.00 top
27786 root 15 0 5484 964 576 S 0.0 0.2 0:00.15 sshd
27843 root 15 0 2284 760 596 S 0.0 0.1 0:00.18 cron
1 root 15 0 2024 680 588 S 0.0 0.1 0:02.31 init
27706 daemon 25 0 1800 472 388 S 0.0 0.1 0:00.00 portmap
27819 daemon 18 0 2152 416 292 S 0.0 0.1 0:00.00 atd
And the default installation took about 465MB of disk space:
df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 20G 465M 20G 3% / tmpfs 257M 0 257M 0% /lib/init/rw tmpfs 257M 0 257M 0% /dev/shm
After the LNMP stack was put up and apache2 was removed, the memory went down to 90MB:
free -m
total used free shared buffers cached
Mem: 512 90 422 0 0 0
-/+ buffers/cache: 90 422
Swap: 0 0 0
And the corresponding top output:
top - 00:42:14 up 1 day, 20:16, 1 user, load average: 0.02, 0.51, 0.78
Tasks: 19 total, 2 running, 17 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.3%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 524800k total, 92404k used, 432396k free, 0k buffers
Swap: 0k total, 0k used, 0k free, 0k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
27945 www 19 0 14492 10m 412 S 0.0 2.0 0:00.00 nginx
13722 mysql 19 0 34792 5828 3152 S 0.0 1.1 0:00.00 mysqld
27936 root 18 0 22700 4536 1380 S 0.0 0.9 0:00.11 php-cgi
27937 www 18 0 22700 4156 1000 S 0.0 0.8 0:00.00 php-cgi
27938 www 18 0 22700 4156 1000 S 0.0 0.8 0:00.00 php-cgi
27939 www 18 0 22700 4156 1000 S 0.0 0.8 0:00.00 php-cgi
27940 www 18 0 22700 4156 1000 S 0.0 0.8 0:00.00 php-cgi
27941 www 18 0 22700 4156 1000 S 0.0 0.8 0:00.00 php-cgi
6015 root 16 0 8360 2784 2244 R 0.0 0.5 0:02.29 sshd
6018 root 15 0 4012 2720 1340 S 0.0 0.5 0:00.17 bash
27774 root 17 0 33436 1396 1044 S 0.0 0.3 0:00.12 rsyslogd
13619 root 25 0 2672 1204 1000 S 0.0 0.2 0:00.00 mysqld_safe
27968 root 15 0 2324 1068 876 R 0.0 0.2 0:00.00 top
27786 root 15 0 5484 964 576 S 0.0 0.2 0:00.15 sshd
27843 root 15 0 2284 760 596 S 0.0 0.1 0:00.19 cron
27944 root 19 0 4680 704 264 S 0.0 0.1 0:00.00 nginx
1 root 15 0 2024 680 588 S 0.0 0.1 0:02.38 init
And used less than 2GB of hard drive space:
df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 20G 1.8G 19G 9% / tmpfs 257M 0 257M 0% /lib/init/rw tmpfs 257M 0 257M 0% /dev/shm
vmstat shows the VPS is very lightly used:
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 432096 0 0 0 0 3 15 0 122 1 0 99 0
And similar results in the output of uptime:
uptime 00:43:05 up 1 day, 20:17, 1 user, load average: 0.00, 0.43, 0.73
Beancounters shows some pretty decent values:
cat /proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
4006187: kmemsize 2955335 3827431 51200000 51200000 0
lockedpages 0 0 2048 2048 0
privvmpages 23243 130666 131200 262200 1
shmpages 642 729 128000 128000 0
dummy 0 0 0 0 0
numproc 22 60 500 500 0
physpages 7053 36749 0 200000000 0
vmguarpages 0 0 512000 512000 0
oomguarpages 7053 36749 512000 512000 0
numtcpsock 6 9 4096 4096 0
numflock 3 9 2048 2048 0
numpty 1 2 32 32 0
numsiginfo 0 3 1024 1024 0
tcpsndbuf 221128 203624 20480000 20480000 0
tcprcvbuf 98304 4759864 20480000 20480000 0
othersockbuf 11640 26024 20480000 20480000 0
dgramrcvbuf 0 8752 10240000 10240000 0
numothersock 11 13 2048 2048 0
dcachesize 214092 226847 10240000 10240000 0
numfile 591 1086 50000 50000 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 24 24 2048 2048 0
Although I would question if the Inode value is set to a bit low:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/simfs 400000 68961 331039 18% / tmpfs 65600 5 65595 1% /lib/init/rw tmpfs 65600 1 65599 1% /dev/shm
Disk I/O veries quite a bit, the first time when I ran dd test, the I/O was really good at about 80MB/s:
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, 13.217 s, 81.2 MB/s
Second time running it after a while shows the results is not all that impressive:
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, 23.8376 s, 45.0 MB/s
CPU info shows that the CPU given is indeed one core, however unlike advertised, the clock speed is actually 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 : 2394.071 cache size : 12288 KB physical id : 1 siblings : 8 core id : 0 cpu cores : 4 apicid : 32 fpu : yes fpu_exception : yes cpuid level : 11 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 syscall nx pdpe1gb rdtscp lm constant_tsc ida nonstop_tsc arat pni monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr sse4_1 sse4_2 popcnt lahf_lm bogomips : 4788.14 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management: [8]
vzfree shows no sign of overselling and interestingly enough, the burst RAM is 2GB in size, which is 4 times of the dedicated RAM:
vzfree
Total Used Free
Kernel: 48.83M 2.83M 46.00M
Allocate: 512.50M 90.79M 421.71M (2000M Guaranteed)
Commit: 2000.00M 30.38M 1969.62M (30.4% of Allocated)
Swap: 0.00M (0.0% of Committed)
And this is again confirmed by meminfo output:
cat /proc/meminfo MemTotal: 524800 kB MemFree: 431660 kB Buffers: 0 kB Cached: 0 kB SwapCached: 0 kB Active: 0 kB Inactive: 0 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 524800 kB LowFree: 431660 kB SwapTotal: 0 kB SwapFree: 0 kB Dirty: 98540 kB Writeback: 0 kB AnonPages: 0 kB Mapped: 0 kB Slab: 0 kB PageTables: 0 kB NFS_Unstable: 0 kB Bounce: 0 kB CommitLimit: 0 kB Committed_AS: 0 kB VmallocTotal: 0 kB VmallocUsed: 0 kB VmallocChunk: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 Hugepagesize: 2048 kB
Downloading from Cachefly was pretty good, I was able to achieve over 10MBps:
wget cachefly.cachefly.net/100mb.test --2011-08-12 00:45:43-- 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: `100mb.test' 100%[======================================>] 104,857,600 11.0M/s in 8.0s 2011-08-12 00:45:51 (12.5 MB/s) - `100mb.test' saved [104857600/104857600]
Considering that it is on shared 100Mbit port, it is a pretty good speed.
As for upload speed, it varies greatly based on location, with the east coast the worst, west coast in the middle, and surprisingly, UK location had the best upload speed:
First from my Nix Communication’s VPS in Montreal, Canada:
wget 184.22.120.20/100mb.test -O /dev/null --2011-08-16 02:25:41-- http://184.22.120.20/100mb.test Connecting to 184.22.120.20:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 938K/s in 69s 2011-08-16 02:26:51 (1.44 MB/s) - `/dev/null' saved [104857600/104857600]
Note that I have added a parameter there to redirect the wget output to /dev/null to avoid any disk I/O issue.
With my BuyVM VPS in San Jose, CA (west coast):
wget 184.22.120.20/100mb.test -O /dev/null --2011-08-15 14:24:22-- http://184.22.120.20/100mb.test Connecting to 184.22.120.20:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 6.42M/s in 18s 2011-08-15 14:24:40 (5.58 MB/s) - `/dev/null' saved [104857600/104857600]
And finally with Quickweb VPS in London, UK:
wget 184.22.120.20/100mb.test -O /dev/null --2011-08-15 14:28:21-- http://184.22.120.20/100mb.test Connecting to 184.22.120.20:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 12.4M/s in 10s 2011-08-15 14:28:32 (9.71 MB/s) - `/dev/null' saved [104857600/104857600]
It is interesting that despite the specifications seems decent, the UnixBench score does not turned out to be very great, which I assume was due to the fluctuation in disk I/O:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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: burst: GNU/Linux
OS: GNU/Linux -- 2.6.18-238.5.1.el5.028stab085.3 -- #1 SMP Mon Mar 21 20:05:12 MSK 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4788.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
06:43:45 up 2 days, 2:17, 1 user, load average: 0.09, 0.68, 0.98; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Fri Aug 12 2011 06:43:45 - 07:12:11
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 12818214.4 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2582.9 MWIPS (9.9 s, 7 samples)
Execl Throughput 3765.0 lps (29.7 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 492302.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 132234.9 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1101314.2 KBps (30.0 s, 2 samples)
Pipe Throughput 878621.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 260134.1 lps (10.0 s, 7 samples)
Process Creation 11679.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4256.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 583.4 lpm (60.1 s, 2 samples)
System Call Overhead 677236.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12818214.4 1098.4
Double-Precision Whetstone 55.0 2582.9 469.6
Execl Throughput 43.0 3765.0 875.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 492302.7 1243.2
File Copy 256 bufsize 500 maxblocks 1655.0 132234.9 799.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 1101314.2 1898.8
Pipe Throughput 12440.0 878621.0 706.3
Pipe-based Context Switching 4000.0 260134.1 650.3
Process Creation 126.0 11679.2 926.9
Shell Scripts (1 concurrent) 42.4 4256.3 1003.8
Shell Scripts (8 concurrent) 6.0 583.4 972.3
System Call Overhead 15000.0 677236.2 451.5
========
System Benchmarks Index Score 858.3
Running it three days later, however, does not seem to give a better 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
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: burst: GNU/Linux
OS: GNU/Linux -- 2.6.18-238.5.1.el5.028stab085.3 -- #1 SMP Mon Mar 21 20:05:12 MSK 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: Intel(R) Xeon(R) CPU E5620 @ 2.40GHz (4788.1 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
01:51:39 up 5 days, 21:25, 1 user, load average: 0.00, 0.00, 0.00; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Tue Aug 16 2011 01:51:39 - 02:20:06
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 12680222.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2569.3 MWIPS (10.0 s, 7 samples)
Execl Throughput 3738.0 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 481810.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 132079.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1079214.5 KBps (30.0 s, 2 samples)
Pipe Throughput 868606.3 lps (10.0 s, 7 samples)
Pipe-based Context Switching 260336.9 lps (10.0 s, 7 samples)
Process Creation 11679.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4067.1 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 570.9 lpm (60.1 s, 2 samples)
System Call Overhead 668336.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 12680222.9 1086.6
Double-Precision Whetstone 55.0 2569.3 467.1
Execl Throughput 43.0 3738.0 869.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 481810.7 1216.7
File Copy 256 bufsize 500 maxblocks 1655.0 132079.4 798.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 1079214.5 1860.7
Pipe Throughput 12440.0 868606.3 698.2
Pipe-based Context Switching 4000.0 260336.9 650.8
Process Creation 126.0 11679.5 926.9
Shell Scripts (1 concurrent) 42.4 4067.1 959.2
Shell Scripts (8 concurrent) 6.0 570.9 951.5
System Call Overhead 15000.0 668336.1 445.6
========
System Benchmarks Index Score 847.1
GeekBench score is pretty good though:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.18-238.5.1.el5.028stab085.3 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: 2.39 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: 23.5 GB
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 1794 |||||||
multi-threaded scalar 1880 |||||||
Text Compress
single-threaded scalar 2019 ||||||||
multi-threaded scalar 1941 |||||||
Text Decompress
single-threaded scalar 2010 ||||||||
multi-threaded scalar 2038 ||||||||
Image Compress
single-threaded scalar 1674 ||||||
multi-threaded scalar 1642 ||||||
Image Decompress
single-threaded scalar 1548 ||||||
multi-threaded scalar 1630 ||||||
Lua
single-threaded scalar 3104 ||||||||||||
multi-threaded scalar 3130 ||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2066 ||||||||
multi-threaded scalar 2182 ||||||||
Dot Product
single-threaded scalar 3450 |||||||||||||
multi-threaded scalar 3639 ||||||||||||||
single-threaded vector 4159 ||||||||||||||||
multi-threaded vector 4484 |||||||||||||||||
LU Decomposition
single-threaded scalar 2462 |||||||||
multi-threaded scalar 2433 |||||||||
Primality Test
single-threaded scalar 3397 |||||||||||||
multi-threaded scalar 2784 |||||||||||
Sharpen Image
single-threaded scalar 5871 |||||||||||||||||||||||
multi-threaded scalar 5719 ||||||||||||||||||||||
Blur Image
single-threaded scalar 5272 |||||||||||||||||||||
multi-threaded scalar 5328 |||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 4413 |||||||||||||||||
Write Sequential
single-threaded scalar 7860 |||||||||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3725 ||||||||||||||
Stdlib Write
single-threaded scalar 3236 ||||||||||||
Stdlib Copy
single-threaded scalar 6527 ||||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 5415 |||||||||||||||||||||
single-threaded vector 7532 ||||||||||||||||||||||||||||||
Stream Scale
single-threaded scalar 4073 ||||||||||||||||
single-threaded vector 7164 ||||||||||||||||||||||||||||
Stream Add
single-threaded scalar 4746 ||||||||||||||||||
single-threaded vector 7104 ||||||||||||||||||||||||||||
Stream Triad
single-threaded scalar 4957 |||||||||||||||||||
single-threaded vector 4880 |||||||||||||||||||
Integer Score: 2034 ||||||||
Floating Point Score: 3803 |||||||||||||||
Memory Score: 5152 ||||||||||||||||||||
Stream Score: 5733 ||||||||||||||||||||||
Overall Geekbench Score: 3646 ||||||||||||||
And testing again 3 days later shows pretty consistent results:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.18-238.5.1.el5.028stab085.3 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: 2.39 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: 23.5 GB
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 1765 |||||||
multi-threaded scalar 1922 |||||||
Text Compress
single-threaded scalar 1965 |||||||
multi-threaded scalar 1901 |||||||
Text Decompress
single-threaded scalar 1982 |||||||
multi-threaded scalar 2031 ||||||||
Image Compress
single-threaded scalar 1703 ||||||
multi-threaded scalar 1663 ||||||
Image Decompress
single-threaded scalar 1578 ||||||
multi-threaded scalar 1605 ||||||
Lua
single-threaded scalar 3048 ||||||||||||
multi-threaded scalar 3108 ||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 2125 ||||||||
multi-threaded scalar 2183 ||||||||
Dot Product
single-threaded scalar 3429 |||||||||||||
multi-threaded scalar 3705 ||||||||||||||
single-threaded vector 4141 ||||||||||||||||
multi-threaded vector 4608 ||||||||||||||||||
LU Decomposition
single-threaded scalar 2376 |||||||||
multi-threaded scalar 2403 |||||||||
Primality Test
single-threaded scalar 3440 |||||||||||||
multi-threaded scalar 2767 |||||||||||
Sharpen Image
single-threaded scalar 5652 ||||||||||||||||||||||
multi-threaded scalar 5870 |||||||||||||||||||||||
Blur Image
single-threaded scalar 5346 |||||||||||||||||||||
multi-threaded scalar 5135 ||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 4372 |||||||||||||||||
Write Sequential
single-threaded scalar 7891 |||||||||||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3852 |||||||||||||||
Stdlib Write
single-threaded scalar 3298 |||||||||||||
Stdlib Copy
single-threaded scalar 6220 ||||||||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 5656 ||||||||||||||||||||||
single-threaded vector 7646 ||||||||||||||||||||||||||||||
Stream Scale
single-threaded scalar 4300 |||||||||||||||||
single-threaded vector 6562 ||||||||||||||||||||||||||
Stream Add
single-threaded scalar 4849 |||||||||||||||||||
single-threaded vector 5933 |||||||||||||||||||||||
Stream Triad
single-threaded scalar 4720 ||||||||||||||||||
single-threaded vector 4743 ||||||||||||||||||
Integer Score: 2022 ||||||||
Floating Point Score: 3798 |||||||||||||||
Memory Score: 5126 ||||||||||||||||||||
Stream Score: 5551 ||||||||||||||||||||||
Overall Geekbench Score: 3617 ||||||||||||||
Overall, the VPS has pretty decent hardware specifications and other than the disk I/O speed and perhaps the Inode was a little too low, the resources are pretty good for the price.
Customer Service and Support
Despite having a huge customer base and a somewhat outdated customer service system, the customer service level is actually and surprisingly pretty good. My first technical support request was replied in about half an hour (ticket was submitted at 2:31pm and reply was received at 3:12pm, which was the “slowest” one among all. The reply for that ticket was answered in 6 minutes and another tickets that I have put in at 8:00pm was answered at 8:05pm.
Billing department, although quite frankly Jennifer sometimes appears to be quite blunt in response, was nonetheless pretty efficient. My only ticket with billing was raised at 2:02pm and response was received at 4:14pm.
Conclusion
Being perhaps one of the oldest companies in VPS service definitely has its advantages, economies of scale enabled Burst to offer a pretty good specification with at a very decent price. Although disk I/O may fluctuate a lot and the Inode set to a bit low value, which made me probably would think twice before putting any production platform on it, it is nonetheless a very good VPS for non-production uses. The vePortal control panel is definitely impressive and the customer service is pretty fast, furthermore, since Burst has been in business for 20 years, knowing that it is unlikely to be in the deadpool soon is definitely something that would give them a few more points.
Have you tested to send email to Yahoo mail, with that VPS?
Nope, I did not when I did the test, but I decided to give it a try just now and it was OK:
X-Originating-IP: [184.22.120.20]
Authentication-Results: mta1005.mail.cnb.yahoo.com from=96mb.com; domainkeys=neutral (no sig); from=96mb.com; dkim=neutral (no sig)
Received: from 127.0.0.1 (EHLO burst) (184.22.120.20)
by mta1005.mail.cnb.yahoo.com with SMTP; Tue, 16 Aug 2011 11:16:59 +0800
Received: by burst (Postfix, from userid 1001)
id 7BFB2CE0688; Tue, 16 Aug 2011 11:17:57 +0800 (CST)
Thanks for the review.
I would like to touch on a few points you mentioned:
• Every node has a FQDN. If you need one you only have to send in a request.
• Inodes – We start them at a decent level and will raise them upon request at no charge/issue.
• Template Names for OS – As we do not offer minimal installs, there’s no need for template names.
•Backups – The ability to backup for free comes with every VPS. The reason you probably did not see anything was because your backup was not complete. You should have a directory listing available by now.You can also DL your backup and store it off-site as we do not do backups of your data.
•IPV6 – It should have set them up automatically for you (Some older nodes in Scranton do not, but all Miami nodes should have that). If you can provide the first four numbers of your VEID, or open a ticket and ask that engineers look into it (And mention this link) we can look into that.
•Jennifer – She’s extremely pleasant, but sometimes when there’s a lot of tickets and issues (And you can imagine how many we have with the free month offer), she has to tuck the charm into a back-pocket for a wee bit. Either way she is always helpful and thorough.
•Benchmarks – IMHO running benchmarks on VPS is akin to running performance tests on a Yugo (http://en.wikipedia.org/wiki/Zastava_Koral). 🙂
I hope this cleared up a few points. Feel free to ask us any questions any time.
” interestingly enough, the burst RAM is 2GB in size, which is 4 times of the dedicated RAM”
The burst ram is set to 512mb and the guaranteed ram set to 2GB. Since OpenVZ primarily limits allocation by the burst ram you will never be able to allocate more than 512mb.
@dmmcintyre3: but you will still be able to use more than 512MB correct? Even though you perhaps can’t do that 24×7?
no, 512mb is all you can allocate.
@dmmcintyre3: So you can NEVER go over 512MB? Basically it is like you have a server with only 512MB of RAM set up and nothing else for memory?
512mb is the limit.
@dmmcintyre3: Interesting, all along my understanding is like 512MB is the “dedicated”, i.e, like your own little room where nobody else is allowed to enter, and 2G is like the “living room” where you could dump your stuff there for a while but not for too long. So in this case, what is the use of that 2GB then?
the 2000MB is what the “guarrenteed” ram is set to, but since the burst ram is set to 512mb you can’t use more than 512mb. I’m not sure why burst.net/veportal does it like that.
@dmmcintyre3: So this is like saying you have a room of 200sq ft but because you put up a wall with only 51.2 sq ft of space in there and only allows you to use that much space? That is so weird…
Well, it’s not EXACTLY like that.
Let’s take your private room analogy:
Let’s say you’re a slob and you fill your room to the rafters with crap. The person who is renting you the room has two options:
1) Let the room explode and try to clean up the mess.
2) Open the door a little and let some crap spill out into the living room so the room doesn’t explode and you don’t get paged at 2AM because some fool has chosen to take their one bedroom and act like it’s a mansion.
We choose option 2 as we hate getting paged 🙂
Basically, the 2G limit is like an overflow valve. If your server is gonna blow up, we vent it out a little. But it’s by no means BURSTABLE memory. It’s why we don’t advertize it. Now, our PREMIUM VPS’s does indeed have burstable memory, and can be utilized as such.
Hope this clears it up.
@BurstNET_CSM: So basically the idea is that this is like for emergency use only but won’t normally make available, correct? That is a rather interesting set up 🙂
Pingback: 96MB Low End VPS Review Part 67–Virpus Xen - 96MB.com