96MB Low End VPS Review Part XVIII – Last Minute Review of BuyVM
It has been a while since my last post about BuyVM and several things have happened since. Over all of them, the biggest change was that they are no longer with Hurricane Electric and instead moved all their servers to Coresite.com, as indicated here. How are the pony VPS doing now after months of improvements? 96MB decided to take a quick peek just before the new stock opens again in a few hours’ time.
Note that this is a not a full review since many thing have not changed since my last review, instead, I will only focus on things that are changed.
OS Selections
Several OS templates have been added to the BuyVM’s list of OS selections since the last time I reviewed them, most notably Debian 6, CentOS 6 and CPanel DNS-only template, which according to Francisco, would use very minimal amount of memory.
Here is a list of the OS templates that are available to the VPS users at the moment:
Tests on the VPS
The testing VPS that I used here is one of their famous 15 per year, 128MB VPS on node 32 and I loaded it up with the Ubuntu 10.04 minimal template. The data centre, as mentioned above, has been moved from Fremont to San Jose in California in west coast US.
With the OS fresh install, the VPS uses merely 15MB of RAM, which is good for Ubuntu since Ubuntu itself is a little heavier than Debian in terms of memory footprint:
free -m
total used free shared buffers cached
Mem: 256 15 240 0 0 0
-/+ buffers/cache: 15 240
Swap: 0 0 0
The top output is really short, which shows most of the default installations have been removed:
top - 07:18:17 up 13 min, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 11 total, 2 running, 9 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 262144k total, 16208k used, 245936k free, 0k buffers
Swap: 0k total, 0k used, 0k free, 0k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1604 root 18 0 8280 2812 S2240 R 0.0 1.1 0:00.09 sshd
1593 root 18 0 5156 2324 1072 S 0.0 0.9 0:00.04 apache2
1401 root 18 0 5472 2116 1712 S 0.0 0.8 0:00.01 sshd
1594 www-data 22 0 5156 1744 484 S 0.0 0.7 0:00.00 apache2
1576 root 15 0 11064 1676 528 S 0.0 0.6 0:00.01 sendmail-mta
1615 root 15 0 2984 1660 1320 S 0.0 0.6 0:00.00 bash
1 root 15 0 2472 1396 1160 S 0.0 0.5 0:00.23 init
1627 root 15 0 2344 1076 888 R 0.0 0.4 0:00.00 top
1211 root 18 0 2304 896 712 S 0.0 0.3 0:00.00 cron
1548 root 25 0 2424 860 688 S 0.0 0.3 0:00.00 xinetd
1232 syslog 15 0 1916 672 536 S 0.0 0.3 0:00.00 syslogd
Note that Apache2 is still installed, which I am sure it will save a little memory here if that could be removed as well.
After the LNMP stack is installed, the memory usage basically takes about 60MB:
free -m
total used free shared buffers cached
Mem: 256 61 194 0 0 0
-/+ buffers/cache: 61 194
Swap: 0 0 0
And the top output shows the entire LNMP stack
top - 20:26:38 up 9:21, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 19 total, 2 running, 17 sleeping, 0 stopped, 0 zombie
Cpu(s): 0.0%us, 0.4%sy, 0.0%ni, 99.6%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 262144k total, 64348k used, 197796k free, 0k buffers
Swap: 0k total, 0k used, 0k free, 0k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
19608 www 15 0 14484 10m 776 S 0.0 4.2 0:00.01 nginx
19594 mysql 19 0 34872 5812 3124 S 0.0 2.2 0:00.02 mysqld
19600 www 18 0 22756 4908 1584 S 0.0 1.9 0:00.00 php-cgi
19601 www 18 0 22756 4908 1584 S 0.0 1.9 0:00.00 php-cgi
19602 www 18 0 22756 4908 1584 S 0.0 1.9 0:00.00 php-cgi
19603 www 18 0 22756 4908 1584 S 0.0 1.9 0:00.00 php-cgi
19604 www 18 0 22756 4908 1584 S 0.0 1.9 0:00.00 php-cgi
19599 root 18 0 22756 4692 1396 S 0.0 1.8 0:06.83 php-cgi
1604 root 18 0 8416 2868 2260 R 0.0 1.1 0:05.38 sshd
1401 root 18 0 5472 2116 1712 S 0.0 0.8 0:00.01 sshd
1576 root 15 0 11064 1676 528 S 0.0 0.6 0:00.63 sendmail-mta
1615 root 15 0 2984 1668 1324 S 0.0 0.6 0:00.02 bash
1 root 15 0 2472 1396 1160 S 0.0 0.5 0:00.23 init
20380 root 15 0 2344 1088 888 R 0.0 0.4 0:00.00 top
1211 root 15 0 2304 904 720 S 0.0 0.3 0:00.07 cron
1548 root 25 0 2424 860 688 S 0.0 0.3 0:00.00 xinetd
19607 root 18 0 4672 700 252 S 0.0 0.3 0:00.00 nginx
vmstats shows the node is not heavily 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 198020 0 0 0 0 18 79 0 422 4 1 94 0
which is confirmed again by the output of uptime:
uptime 06:14:05 up 19:09, 1 user, load average: 0.05, 0.02, 0.00
Beancounters value are set to pretty high:
cat /proc/user_beancounters
Version: 2.5
uid resource held maxheld barrier limit failcnt
11183: kmemsize 2749601 9523915 2147483646 2147483646 0
lockedpages 0 448 999999 999999 0
privvmpages 15948 53515 65536 65536 0
shmpages 642 673 32768 32768 0
dummy 0 0 0 0 0
numproc 20 48 999999 999999 0
physpages 7459 31632 0 2147483647 0
vmguarpages 0 0 32768 2147483647 0
oomguarpages 7459 31632 32768 2147483647 0
numtcpsock 7 16 7999992 7999992 0
numflock 2 8 999999 999999 0
numpty 1 2 500000 500000 0
numsiginfo 0 7 999999 999999 0
tcpsndbuf 122528 5876968 214748160 396774400 0
tcprcvbuf 114688 12253224 214748160 396774400 0
othersockbuf 13968 57184 214748160 396774400 0
dgramrcvbuf 0 8472 214748160 396774400 0
numothersock 15 20 7999992 7999992 0
dcachesize 0 0 2147483646 2147483646 0
numfile 470 1527 23999976 23999976 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
dummy 0 0 0 0 0
numiptent 24 24 256 256 0
Vzfree shows no signs of overselling memory:
vzfree
Total Used Free
Kernel: 2048.00M 2.62M 2045.38M
Allocate: 256.00M 62.28M 193.72M (128M Guaranteed)
Commit: 128.00M 31.74M 96.26M (46.8% of Allocated)
Swap: 0.00M (0.0% of Committed)
whic is re-affirmed by meminfo output:
cat /proc/meminfo MemTotal: 262144 kB MemFree: 198640 kB Buffers: 0 kB Cached: 0 kB SwapCached: 0 kB Active: 0 kB Inactive: 0 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 262144 kB LowFree: 198640 kB SwapTotal: 0 kB SwapFree: 0 kB Dirty: 520 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
As with last time, the CPU is assigned to this VPS is still just one core, and remains to be the same E5520 CPU:
cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 26 model name : Intel(R) Xeon(R) CPU E5520 @ 2.27GHz stepping : 5 cpu MHz : 2261.452 cache size : 4096 KB physical id : 0 siblings : 4 core id : 0 cpu cores : 4 apicid : 0 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 mmx fxsr sse sse2 ss ht syscall nx rdtscp lm constant_tsc pni ssse3 cx16 sse4_1 sse4_2 popcnt lahf_lm bogomips : 4522.90 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management:
Hard drive space is not oversold,as indicated by the df output:
df -h Filesystem Size Used Avail Use% Mounted on /dev/simfs 15G 1.6G 14G 11% /
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/simfs 7864320 67925 7796395 1% /
However, unfortunately, the disk I/O speed is getting pretty unstable:
I was getting a pretty good results running it for the first time:
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, 12.9436 s, 83.0 MB/s
Running it the next morning shows the speed is pretty bad:
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, 22.769 s, 47.2 MB/s
However, runnning it while I was typing this article right now shows the results has improved significantly:
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, 7.60879 s, 141 MB/s
Nonetheless, I think even the worst I/O among the three tests is more than sufficient for a general-purpose server and is already way better than what I have got a few weeks ago, right after the move:
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, 165.596 s, 6.5 MB/s
Network was one of the major issues cited by BuyVM on their decision to move away from Hurricance Electric, and Coresite seems to offer a better download speed:
wget cachefly.cachefly.net/100mb.test --2011-08-08 20:29:30-- http://cachefly.cachefly.net/100mb.test Resolving cachefly.cachefly.net... 205.234.175.175 Connecting to cachefly.cachefly.net|205.234.175.175|:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `100mb.test' 100%[======================================>] 104,857,600 28.0M/s in 3.6s 2011-08-08 20:29:33 (28.0 MB/s) - `100mb.test' saved [104857600/104857600]
Unfortunately upload speed has not improved a lot, and although the best among my three test VPS, the results from Nix Communications VPS in Montreal, Canada, gives a respectable 5.6MB/s:
wget 205.185.123.50/100mb.test --2011-08-08 21:30:20-- http://205.185.123.50/100mb.test Connecting to 205.185.123.50:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `100mb.test' 100%[======================================>] 104,857,600 4.50M/s in 20s 2011-08-08 21:30:41 (5.08 MB/s) - `100mb.test' saved [104857600/104857600]
QuickWeb VPS in London, UK, gives only 1.63MB/s:
wget 205.185.123.50/100mb.test --2011-08-08 12:30:17-- http://205.185.123.50/100mb.test Connecting to 205.185.123.50:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `100mb.test' 100%[======================================>] 104,857,600 851K/s in 61s 2011-08-08 12:31:19 (1.63 MB/s) - `100mb.test' saved [104857600/104857600]
And for the slowest speed goes with my test VPS with Inception Hosting in Netherlands, which seems to show that Coresite is also having trouble in providing fast trans-Atlantic connection:
wget 205.185.123.50/100mb.test --2011-08-08 13:23:05-- http://205.185.123.50/100mb.test Connecting to 205.185.123.50:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `100mb.test' 100%[======================================>] 104,857,600 739K/s in 3m 19s 2011-08-08 13:26:24 (516 KB/s) - `100mb.test' saved [104857600/104857600]
Finally, for the benchmarks, given the improvements in the VPS, including the upgrade in CPU, I am not surprised that the VPS did pretty well during the UnixBench run:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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: *****: GNU/Linux
OS: GNU/Linux -- 2.6.18-238.12.1.el5.pony6-1 -- #3 SMP Fri Jun 3 16:37:31 PDT 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 E5520 @ 2.27GHz (4522.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
23:17:02 up 12:11, 1 user, load average: 0.21, 0.07, 0.02; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Mon Aug 08 2011 23:17:02 - 23:45:05
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 9732199.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2229.3 MWIPS (9.8 s, 7 samples)
Execl Throughput 1894.4 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 278777.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 76948.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 606654.3 KBps (30.0 s, 2 samples)
Pipe Throughput 602595.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 155244.0 lps (10.0 s, 7 samples)
Process Creation 5312.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3118.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 421.2 lpm (60.1 s, 2 samples)
System Call Overhead 524068.0 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 9732199.7 834.0
Double-Precision Whetstone 55.0 2229.3 405.3
Execl Throughput 43.0 1894.4 440.6
File Copy 1024 bufsize 2000 maxblocks 3960.0 278777.5 704.0
File Copy 256 bufsize 500 maxblocks 1655.0 76948.3 464.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 606654.3 1046.0
Pipe Throughput 12440.0 602595.4 484.4
Pipe-based Context Switching 4000.0 155244.0 388.1
Process Creation 126.0 5312.1 421.6
Shell Scripts (1 concurrent) 42.4 3118.2 735.4
Shell Scripts (8 concurrent) 6.0 421.2 702.0
System Call Overhead 15000.0 524068.0 349.4
========
System Benchmarks Index Score 547.7
Second time, the score was a little better, mostly due to disk caching:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
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: ******: GNU/Linux
OS: GNU/Linux -- 2.6.18-238.12.1.el5.pony6-1 -- #3 SMP Fri Jun 3 16:37:31 PDT 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 E5520 @ 2.27GHz (4522.9 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
23:49:16 up 12:44, 1 user, load average: 0.08, 1.50, 1.38; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Mon Aug 08 2011 23:49:16 - 00:17:25
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 10033436.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2227.4 MWIPS (10.0 s, 7 samples)
Execl Throughput 1917.9 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 267100.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 81690.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 616777.9 KBps (30.0 s, 2 samples)
Pipe Throughput 603938.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 151863.9 lps (10.0 s, 7 samples)
Process Creation 5203.3 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 3269.3 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 429.4 lpm (60.1 s, 2 samples)
System Call Overhead 525018.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 10033436.7 859.8
Double-Precision Whetstone 55.0 2227.4 405.0
Execl Throughput 43.0 1917.9 446.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 267100.3 674.5
File Copy 256 bufsize 500 maxblocks 1655.0 81690.2 493.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 616777.9 1063.4
Pipe Throughput 12440.0 603938.0 485.5
Pipe-based Context Switching 4000.0 151863.9 379.7
Process Creation 126.0 5203.3 413.0
Shell Scripts (1 concurrent) 42.4 3269.3 771.1
Shell Scripts (8 concurrent) 6.0 429.4 715.7
System Call Overhead 15000.0 525018.6 350.0
========
System Benchmarks Index Score 552.5
For GeekBench, the results are pretty good as well:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Ubuntu 10.04.3 LTS 2.6.18-238.12.1.el5.pony6-1 i686
Model: Linux PC (Intel Xeon E5520)
Motherboard: Unknown Motherboard
Processor: Intel Xeon E5520
Processor ID: GenuineIntel Family 6 Model 26 Stepping 5
Logical Processors: 1
Physical Processors: 1
Processor Frequency: 2.26 GHz
L1 Instruction Cache: 64.0 KB
L1 Data Cache: 64.0 KB
L2 Cache: 512 KB
L3 Cache: 0.00 B
Bus Frequency: 0.00 Hz
Memory: 6.80 GB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel Xeon E5520
Processor Cores: 1
Integer
Blowfish
single-threaded scalar 1454 |||||
multi-threaded scalar 1550 ||||||
Text Compress
single-threaded scalar 1588 ||||||
multi-threaded scalar 1575 ||||||
Text Decompress
single-threaded scalar 1613 ||||||
multi-threaded scalar 1677 ||||||
Image Compress
single-threaded scalar 1360 |||||
multi-threaded scalar 1255 |||||
Image Decompress
single-threaded scalar 1231 ||||
multi-threaded scalar 1217 ||||
Lua
single-threaded scalar 2423 |||||||||
multi-threaded scalar 2470 |||||||||
Floating Point
Mandelbrot
single-threaded scalar 1759 |||||||
multi-threaded scalar 1768 |||||||
Dot Product
single-threaded scalar 2638 ||||||||||
multi-threaded scalar 2921 |||||||||||
single-threaded vector 3488 |||||||||||||
multi-threaded vector 3770 |||||||||||||||
LU Decomposition
single-threaded scalar 2049 ||||||||
multi-threaded scalar 2135 ||||||||
Primality Test
single-threaded scalar 2833 |||||||||||
multi-threaded scalar 2178 ||||||||
Sharpen Image
single-threaded scalar 4653 ||||||||||||||||||
multi-threaded scalar 4533 ||||||||||||||||||
Blur Image
single-threaded scalar 4430 |||||||||||||||||
multi-threaded scalar 4044 ||||||||||||||||
Memory
Read Sequential
single-threaded scalar 2484 |||||||||
Write Sequential
single-threaded scalar 4428 |||||||||||||||||
Stdlib Allocate
single-threaded scalar 2848 |||||||||||
Stdlib Write
single-threaded scalar 2480 |||||||||
Stdlib Copy
single-threaded scalar 4796 |||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 3013 ||||||||||||
single-threaded vector 3521 ||||||||||||||
Stream Scale
single-threaded scalar 2756 |||||||||||
single-threaded vector 3569 ||||||||||||||
Stream Add
single-threaded scalar 2610 ||||||||||
single-threaded vector 3367 |||||||||||||
Stream Triad
single-threaded scalar 2866 |||||||||||
single-threaded vector 2570 ||||||||||
Integer Score: 1617 ||||||
Floating Point Score: 3085 ||||||||||||
Memory Score: 3407 |||||||||||||
Stream Score: 3034 ||||||||||||
Overall Geekbench Score: 2630 ||||||||||
And second results is once again pretty consistent:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Ubuntu 10.04.3 LTS 2.6.18-238.12.1.el5.pony6-1 i686
Model: Linux PC (Intel Xeon E5520)
Motherboard: Unknown Motherboard
Processor: Intel Xeon E5520
Processor ID: GenuineIntel Family 6 Model 26 Stepping 5
Logical Processors: 1
Physical Processors: 1
Processor Frequency: 2.26 GHz
L1 Instruction Cache: 64.0 KB
L1 Data Cache: 64.0 KB
L2 Cache: 512 KB
L3 Cache: 0.00 B
Bus Frequency: 0.00 Hz
Memory: 6.80 GB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: Intel Xeon E5520
Processor Cores: 1
Integer
Blowfish
single-threaded scalar 1486 |||||
multi-threaded scalar 1560 ||||||
Text Compress
single-threaded scalar 1572 ||||||
multi-threaded scalar 1510 ||||||
Text Decompress
single-threaded scalar 1650 ||||||
multi-threaded scalar 1679 ||||||
Image Compress
single-threaded scalar 1158 ||||
multi-threaded scalar 1370 |||||
Image Decompress
single-threaded scalar 1249 ||||
multi-threaded scalar 1289 |||||
Lua
single-threaded scalar 2637 ||||||||||
multi-threaded scalar 2517 ||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1792 |||||||
multi-threaded scalar 1813 |||||||
Dot Product
single-threaded scalar 2825 |||||||||||
multi-threaded scalar 2964 |||||||||||
single-threaded vector 3430 |||||||||||||
multi-threaded vector 3900 |||||||||||||||
LU Decomposition
single-threaded scalar 2150 ||||||||
multi-threaded scalar 2165 ||||||||
Primality Test
single-threaded scalar 2803 |||||||||||
multi-threaded scalar 2266 |||||||||
Sharpen Image
single-threaded scalar 4606 ||||||||||||||||||
multi-threaded scalar 4785 |||||||||||||||||||
Blur Image
single-threaded scalar 4289 |||||||||||||||||
multi-threaded scalar 4305 |||||||||||||||||
Memory
Read Sequential
single-threaded scalar 2896 |||||||||||
Write Sequential
single-threaded scalar 5769 |||||||||||||||||||||||
Stdlib Allocate
single-threaded scalar 3087 ||||||||||||
Stdlib Write
single-threaded scalar 2397 |||||||||
Stdlib Copy
single-threaded scalar 4786 |||||||||||||||||||
Stream
Stream Copy
single-threaded scalar 2633 ||||||||||
single-threaded vector 3756 |||||||||||||||
Stream Scale
single-threaded scalar 2744 ||||||||||
single-threaded vector 3332 |||||||||||||
Stream Add
single-threaded scalar 2006 ||||||||
single-threaded vector 3516 ||||||||||||||
Stream Triad
single-threaded scalar 2767 |||||||||||
single-threaded vector 2411 |||||||||
Integer Score: 1639 ||||||
Floating Point Score: 3149 ||||||||||||
Memory Score: 3787 |||||||||||||||
Stream Score: 2895 |||||||||||
Overall Geekbench Score: 2722 ||||||||||
Conclusion
There has been quite a few improvements since the last time I have review BuyVM’s VPS and it is mostly for the better, there are new OS template selections added and they have moved to a better data centre. The VPS remains pretty solid in general, however it is still having the issue of inconsistent disk I/O, even after improvements were done, although the slowest disk I/O speed among the tests is still pretty fast for such a low end VPS. The network speed to Europe unfortunately still requires some improvements as well. Having said that, overall, for 15 per year, this is definitely a bargain and is good enough for a general purpose VPS for hosting some non-memory intensive website or VPN in general.
Thanks for the review
on the other, lighter note, RSS subscription sometime sucks, I got 7 approx notifications that the article was changed 😛
@Asim: Mmm…interesting, I was so hungry that I start to grab my first slice of pizza as soon as I finished typing up on LEB and post it on Twitter…and definitely did not have time to change it for 7 times…
I guess I’d better find out why before 96MB becomes 96SPAmb, LOL 🙂
Those bumps in your IO are really weird, you should log a ticket when you have time.
It’s quite possible there was a configuration error on our part with a raid cache, or just some really over the top abuse that our systems didn’t spot.
Either way, thanks for the heads up!
Francisco
@Francisco: Ticket Created #540859, thanks!
I have been trying to sign up for a vps with these guys literally for months and every time I try (every few days) it says out of stock.
Is there some trick to getting a VM with them? I even waited end of month/start of new month in case someone canceled 🙂
@Luma: I subscribed to their stock announcement list: http://eepurl.com/daaLk, also they have a live stock count list here: http://doesbuyvmhavestock.com/. That page has Json and XML feed, so you can use whatever tool you prefer to get the stock information.
Anyone noticed this on the CentOS 6 template?
“!!128MB is not enough ram to run this template!!”
If so, why does it work for me on my BuyVM 128, Hostigation (128mb OVZ), Secure Dragon (96mb Xen)?
Interesting, I guess they just want to be safe? I knew (from one of the posts on LEB) that CentOS installer takes a lot of memory because it was written in Python, but for OVZ they do not even need to run the installer, so I guess only Francisco could come and resolve the mystery for us…
Pingback: 96MB Low End VPS Review Part XXXVII – BuyVM KVM Series (Part 2 of 2) | | 96MB.com96MB.com