96MB Low End VPS Review Part XXXVII – BuyVM KVM Series (Part 2 of 2)
In the previous post, we have reviewed the performance of the smallest siblings of BuyVM’s KVM offering. Although the disk I/O is pretty good at close to 100MB/s, the fact that only 1 CPU core is available caused the benchmarks to be pretty low. So, by paying more, other than getting larger RAM, do we also get better performance? In this post, we shall take a look at the performance of a 512MB KVM VPS.
General Information and Set Up
Since both 128MB and 512MB VPS use the same sign up process and the SolusVM control panel is pretty much the same, I will focus our discussion here on a few differences.
First difference is obviously the cost, the 521MB VPS cost 10 per month, and here is what you get for 10USD:
During the sign up process, the warning message that appeared for the 128MB VPS declaring that upgrading/downgrading is not possible for the VPS did not show up while I was signing up for the 521MB VPS, I assume that means for the 512MB VPS, the users are allowed to upgrade/downgrade at any time necessary.
The set up for this 512MB VPS is not instant as well and I waited for 33 minutes for the VPS to set up after received my payment confirmation email to get the new VPS information email (payment sent on 8:05PM, received the VPS information email on 8:38PM), which is actually pretty good.
Test on the VPS
As mentioned above, the VPS has 512MB of RAM and 50 GB of hard disk space and the VPS is located in San Jose, California, although I could have installed Windows, I chose to install Debian 6 32 bit instead for testing purposes.
When the Debian OS was first installed, 17MB of RAM was used, which is almost the same as the results in installing Debian on the 128MB VPS:
free -m
total used free shared buffers cached
Mem: 502 34 467 0 2 17
-/+ buffers/cache: 14 487
Swap: 979 0 979
Top output showing the processes running:
top - 22:43:57 up 9 min, 1 user, load average: 0.00, 0.00, 0.00
Tasks: 67 total, 1 running, 66 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: 514664k total, 35652k used, 479012k free, 2208k buffers
Swap: 1003512k total, 0k used, 1003512k free, 18304k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
1368 root 20 0 8260 2840 2320 S 0 0.6 0:00.23 sshd
1371 root 20 0 5540 2836 1452 S 0 0.6 0:00.46 bash
999 root 20 0 28340 1532 1024 S 0 0.3 0:00.16 rsyslogd
1391 root 20 0 2332 1128 900 R 0 0.2 0:00.09 top
1185 root 20 0 5496 976 584 S 0 0.2 0:00.00 sshd
1344 Debian-e 20 0 6512 928 620 S 0 0.2 0:00.04 exim4
390 root 16 -4 2256 788 416 S 0 0.2 0:00.12 udevd
859 statd 20 0 1936 768 640 S 0 0.1 0:00.01 rpc.statd
1122 root 20 0 3784 768 600 S 0 0.1 0:00.02 cron
1 root 20 0 2032 704 612 S 0 0.1 0:01.02 init
1095 root 20 0 1704 588 480 S 0 0.1 0:00.00 acpid
549 root 18 -2 2252 568 232 S 0 0.1 0:00.00 udevd
1363 root 20 0 1708 552 472 S 0 0.1 0:00.00 getty
1367 root 20 0 1708 552 472 S 0 0.1 0:00.01 getty
550 root 18 -2 2252 548 212 S 0 0.1 0:00.00 udevd
1364 root 20 0 1708 548 472 S 0 0.1 0:00.00 getty
1365 root 20 0 1708 548 472 S 0 0.1 0:00.00 getty
The corresponding htop output:
And less than 700MB of hard drive space was used:
df -h Filesystem Size Used Avail Use% Mounted on /dev/sda1 49G 689M 46G 2% / tmpfs 252M 0 252M 0% /lib/init/rw udev 247M 112K 247M 1% /dev tmpfs 252M 0 252M 0% /dev/shm
When the full LNMP stack was installed, more than 50MB of RAM was used, which was shown in the htop output below:
And the top output showing the entire LNMP stack running:
top - 23:27:36 up 11:53, 1 user, load average: 0.11, 0.06, 0.02 Tasks: 77 total, 1 running, 76 sleeping, 0 stopped, 0 zombie Cpu(s): 0.2%us, 0.0%sy, 0.0%ni, 99.8%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st Mem: 514664k total, 505204k used, 9460k free, 24476k buffers Swap: 1003512k total, 28k used, 1003484k free, 435948k cached PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND 4559 www 20 0 14504 10m 724 S 0 2.1 0:00.06 nginx 15382 mysql 20 0 34808 5852 3176 S 0 1.1 0:00.02 mysqld 4519 root 20 0 24204 4564 1404 S 0 0.9 0:06.92 php-cgi 4527 www 20 0 24204 4168 1008 S 0 0.8 0:00.00 php-cgi 4528 www 20 0 24204 4168 1008 S 0 0.8 0:00.00 php-cgi 4529 www 20 0 24204 4168 1008 S 0 0.8 0:00.00 php-cgi 4530 www 20 0 24204 4168 1008 S 0 0.8 0:00.00 php-cgi 4531 www 20 0 24204 4168 1008 S 0 0.8 0:00.02 php-cgi 4921 root 20 0 5940 3296 1512 S 0 0.6 0:00.54 bash 4918 root 20 0 8396 2856 2324 S 0 0.6 0:00.49 sshd 999 root 20 0 28340 1420 912 S 0 0.3 0:00.35 rsyslogd 5833 root 20 0 2332 1132 900 R 0 0.2 0:00.05 top 1185 root 20 0 5496 888 496 S 0 0.2 0:00.00 sshd 1344 Debian-e 20 0 6512 812 504 S 0 0.2 0:00.09 exim4 1122 root 20 0 3784 768 600 S 0 0.1 0:00.10 cron 4557 root 20 0 4692 708 264 S 0 0.1 0:00.00 nginx 390 root 16 -4 2256 676 304 S 0 0.1 0:00.12 udevd
About 3GB of hard drive space were used:
df -h Filesystem Size Used Avail Use% Mounted on /dev/sda1 49G 3.0G 43G 7% / tmpfs 252M 0 252M 0% /lib/init/rw udev 247M 112K 247M 1% /dev tmpfs 252M 0 252M 0% /dev/shm
And the inodes were set to pretty reasonable value:
df -i Filesystem Inodes IUsed IFree IUse% Mounted on /dev/sda1 3219456 74482 3144974 3% / tmpfs 64333 5 64328 1% /lib/init/rw udev 63227 543 62684 1% /dev tmpfs 64333 1 64332 1% /dev/shm
There are two CPU cores available for this VPS, which is one more than its cheapest sibling that we have reviewed:
cat /proc/cpuinfo processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 13 model name : QEMU Virtual CPU version (cpu64-rhel6) stepping : 3 cpu MHz : 2499.862 cache size : 4096 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 4 wp : yes flags : fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pse3 6 clflush mmx fxsr sse sse2 syscall nx lm pni cx16 hypervisor lahf_lm bogomips : 4999.72 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 : 13 model name : QEMU Virtual CPU version (cpu64-rhel6) stepping : 3 cpu MHz : 2499.862 cache size : 4096 KB fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 4 wp : yes flags : fpu de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pse3 6 clflush mmx fxsr sse sse2 syscall nx lm pni cx16 hypervisor lahf_lm bogomips : 4999.72 clflush size : 64 cache_alignment : 64 address sizes : 40 bits physical, 48 bits virtual power management:
The output of meminfo:
cat /proc/meminfo MemTotal: 514664 kB MemFree: 8848 kB Buffers: 24500 kB Cached: 436636 kB SwapCached: 28 kB Active: 70908 kB Inactive: 413924 kB Active(anon): 2928 kB Inactive(anon): 20884 kB Active(file): 67980 kB Inactive(file): 393040 kB Unevictable: 0 kB Mlocked: 0 kB HighTotal: 0 kB HighFree: 0 kB LowTotal: 514664 kB LowFree: 8848 kB SwapTotal: 1003512 kB SwapFree: 1003484 kB Dirty: 0 kB Writeback: 0 kB AnonPages: 23700 kB Mapped: 7404 kB Shmem: 124 kB Slab: 16320 kB SReclaimable: 12868 kB SUnreclaim: 3452 kB KernelStack: 640 kB PageTables: 688 kB NFS_Unstable: 0 kB Bounce: 0 kB WritebackTmp: 0 kB CommitLimit: 1260844 kB Committed_AS: 105484 kB VmallocTotal: 503812 kB VmallocUsed: 5872 kB VmallocChunk: 490292 kB HardwareCorrupted: 0 kB HugePages_Total: 0 HugePages_Free: 0 HugePages_Rsvd: 0 HugePages_Surp: 0 Hugepagesize: 4096 kB DirectMap4k: 12276 kB DirectMap4M: 512000 kB
And that of time sync:
time sync real 0m0.037s user 0m0.000s sys 0m0.008s
Disk I/O is still pretty good:
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, 13.4374 s, 79.9 MB/s
Testing again showed even better 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, 11.8921 s, 90.3 MB/s
IOPing results are pretty stable as well:
ioping -c 10 . 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=1 time=0.4 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=2 time=0.9 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=3 time=0.6 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=4 time=0.8 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=5 time=2.3 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=6 time=0.8 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=7 time=1.4 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=8 time=0.7 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=9 time=0.8 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=10 time=0.7 ms --- . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c) ioping statistics --- 10 requests completed in 9011.1 ms, 1079 iops, 4.2 mb/s min/avg/max/mdev = 0.4/0.9/2.3/0.5 ms
And the second test is is just as stable as the first, and the round trip time is slightly shorter as well:
ioping -c 10 . 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=1 time=0.3 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=2 time=0.5 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=3 time=0.6 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=4 time=0.9 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=5 time=0.7 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=6 time=0.7 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=7 time=1.1 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=8 time=0.9 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=9 time=0.6 ms 4096 bytes from . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c): request=10 time=0.8 ms --- . (ext3 /dev/disk/by-uuid/31f32b67-cf91-492d-87b3-6080f122678c) ioping statistics --- 10 requests completed in 9009.1 ms, 1411 iops, 5.5 mb/s min/avg/max/mdev = 0.3/0.7/1.1/0.2 ms
During the last review, the network speed turns out to be very dismal. Francisco commented and informed me that it was due to the wrong Network Interface Card (NIC) used, therefore, this time, I made sure I used the VIRTIO NIC card before running the test and let’s take a look at how the download test do first:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2011-11-21 23:28:09-- 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 8.22M/s in 12s 2011-11-21 23:28:21 (8.40 MB/s) - `/dev/null' saved [104857600/104857600]
As you can see, the test speed is pretty slow, however the second test, for some reason, actually showed really good results:
wget cachefly.cachefly.net/100mb.test -O /dev/null --2011-11-27 00:01:48-- 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 28.7M/s in 3.5s 2011-11-27 00:01:51 (28.7 MB/s) - `/dev/null' saved [104857600/104857600]
However, this did not sustain a lot longer and a few minute’s after, the results is back to square one again:
wget cachefly.cachefly.net/100mb.test --2011-11-27 00:05:01-- 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 9.57M/s in 11s 2011-11-27 00:05:12 (8.70 MB/s) - `100mb.test' saved [104857600/104857600]
Upload speed does not look too good as well, the first is with my test VPS from BuffaloVPS in Chicago, IL:
wget 209.141.48.68/100mb.test -O /dev/null --2011-11-26 11:05:08-- http://209.141.48.68/100mb.test Connecting to 209.141.48.68:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 1.73M/s in 2m 57s 2011-11-26 11:08:05 (579 KB/s) - `/dev/null' saved [104857600/104857600]
As I do not have any west coast VPS other than those from BuyVM, unfortunately I can not test for the upload speed along the west coast, so the only test left is with the QuickWeb VPS in London, UK:
wget 209.141.48.68/100mb.test -O /dev/null --2011-11-26 17:41:39-- http://209.141.48.68/100mb.test Connecting to 209.141.48.68:80... connected. HTTP request sent, awaiting response... 200 OK Length: 104857600 (100M) [application/octet-stream] Saving to: `/dev/null' 100%[======================================>] 104,857,600 1.33M/s in 95s 2011-11-26 17:43:14 (1.05 MB/s) - `/dev/null' saved [104857600/104857600]
As you can see, just like the 15 per year plan, network continues to be an issue for BuyVM for those who are looking for some good upload/download speed.
As usual, the benchmarks is the last item for this section. I was hoping the performance of this VPS will be better than its 128MB sibling since there are two CPU cores assigned, however the results does not seems to indicate so:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
2 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
2 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
2 x Execl Throughput 1 2 3
2 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
2 x File Copy 256 bufsize 500 maxblocks 1 2 3
2 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
2 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
2 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
2 x Process Creation 1 2 3
2 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
2 x Shell Scripts (1 concurrent) 1 2 3
2 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: *****: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-686 -- #1 SMP Mon Oct 3 04:15:24 UTC 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
12:46:53 up 3 days, 13:41, 2 users, load average: 0.08, 0.02, 0.01; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 12:46:53 - 13:15:00
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 10924826.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2059.7 MWIPS (9.4 s, 7 samples)
Execl Throughput 268.6 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 423747.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 143747.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 658701.5 KBps (30.0 s, 2 samples)
Pipe Throughput 1150889.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 69944.5 lps (10.0 s, 7 samples)
Process Creation 747.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1022.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 224.4 lpm (60.2 s, 2 samples)
System Call Overhead 1773486.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 10924826.9 936.1
Double-Precision Whetstone 55.0 2059.7 374.5
Execl Throughput 43.0 268.6 62.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 423747.3 1070.1
File Copy 256 bufsize 500 maxblocks 1655.0 143747.0 868.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 658701.5 1135.7
Pipe Throughput 12440.0 1150889.6 925.2
Pipe-based Context Switching 4000.0 69944.5 174.9
Process Creation 126.0 747.0 59.3
Shell Scripts (1 concurrent) 42.4 1022.0 241.0
Shell Scripts (8 concurrent) 6.0 224.4 373.9
System Call Overhead 15000.0 1773486.2 1182.3
========
System Benchmarks Index Score 411.6
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 13:15:00 - 13:43:08
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 19963598.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3885.0 MWIPS (9.6 s, 7 samples)
Execl Throughput 830.1 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 291376.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 84376.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 629088.4 KBps (30.0 s, 2 samples)
Pipe Throughput 2216933.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 128200.2 lps (10.0 s, 7 samples)
Process Creation 1721.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 2052.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 296.7 lpm (60.3 s, 2 samples)
System Call Overhead 1955301.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 19963598.8 1710.7
Double-Precision Whetstone 55.0 3885.0 706.4
Execl Throughput 43.0 830.1 193.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 291376.3 735.8
File Copy 256 bufsize 500 maxblocks 1655.0 84376.0 509.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 629088.4 1084.6
Pipe Throughput 12440.0 2216933.4 1782.1
Pipe-based Context Switching 4000.0 128200.2 320.5
Process Creation 126.0 1721.8 136.6
Shell Scripts (1 concurrent) 42.4 2052.0 484.0
Shell Scripts (8 concurrent) 6.0 296.7 494.4
System Call Overhead 15000.0 1955301.5 1303.5
========
System Benchmarks Index Score 603.0
Yes, I know it is close to 100 points up from the benchmark score obtained by the 128MB VPS, but I was expect a bit more.
Testing again showed slightly worse result:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
2 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
2 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
2 x Execl Throughput 1 2 3
2 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
2 x File Copy 256 bufsize 500 maxblocks 1 2 3
2 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
2 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
2 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
2 x Process Creation 1 2 3
2 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
2 x Shell Scripts (1 concurrent) 1 2 3
2 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: *****: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-686 -- #1 SMP Mon Oct 3 04:15:24 UTC 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
14:14:48 up 3 days, 15:09, 3 users, load average: 0.00, 0.00, 0.44; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 14:14:48 - 14:42:59
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 10861387.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2033.5 MWIPS (10.9 s, 7 samples)
Execl Throughput 256.6 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 417818.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 146276.7 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 737302.7 KBps (30.0 s, 2 samples)
Pipe Throughput 1124513.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 71972.2 lps (10.0 s, 7 samples)
Process Creation 775.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 921.2 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 216.0 lpm (60.1 s, 2 samples)
System Call Overhead 1778788.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 10861387.5 930.7
Double-Precision Whetstone 55.0 2033.5 369.7
Execl Throughput 43.0 256.6 59.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 417818.3 1055.1
File Copy 256 bufsize 500 maxblocks 1655.0 146276.7 883.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 737302.7 1271.2
Pipe Throughput 12440.0 1124513.0 903.9
Pipe-based Context Switching 4000.0 71972.2 179.9
Process Creation 126.0 775.6 61.6
Shell Scripts (1 concurrent) 42.4 921.2 217.3
Shell Scripts (8 concurrent) 6.0 216.0 359.9
System Call Overhead 15000.0 1778788.6 1185.9
========
System Benchmarks Index Score 410.1
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 14:42:59 - 15:11:13
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 20213567.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4068.6 MWIPS (9.9 s, 7 samples)
Execl Throughput 864.0 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 283615.3 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 82226.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 652770.2 KBps (30.0 s, 2 samples)
Pipe Throughput 2181158.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 132381.7 lps (10.0 s, 7 samples)
Process Creation 1564.4 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1979.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 284.3 lpm (60.2 s, 2 samples)
System Call Overhead 1974525.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 20213567.7 1732.1
Double-Precision Whetstone 55.0 4068.6 739.7
Execl Throughput 43.0 864.0 200.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 283615.3 716.2
File Copy 256 bufsize 500 maxblocks 1655.0 82226.0 496.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 652770.2 1125.5
Pipe Throughput 12440.0 2181158.4 1753.3
Pipe-based Context Switching 4000.0 132381.7 331.0
Process Creation 126.0 1564.4 124.2
Shell Scripts (1 concurrent) 42.4 1979.0 466.7
Shell Scripts (8 concurrent) 6.0 284.3 473.8
System Call Overhead 15000.0 1974525.7 1316.4
========
System Benchmarks Index Score 599.7
And I decided to give them yet another try, but this time, the score is even worse:
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
Version 5.1.3 Based on the Byte Magazine Unix Benchmark
Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com
1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
1 x Execl Throughput 1 2 3
1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
1 x File Copy 256 bufsize 500 maxblocks 1 2 3
1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
1 x Process Creation 1 2 3
1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
1 x Shell Scripts (1 concurrent) 1 2 3
1 x Shell Scripts (8 concurrent) 1 2 3
2 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
2 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
2 x Execl Throughput 1 2 3
2 x File Copy 1024 bufsize 2000 maxblocks 1 2 3
2 x File Copy 256 bufsize 500 maxblocks 1 2 3
2 x File Copy 4096 bufsize 8000 maxblocks 1 2 3
2 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10
2 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
2 x Process Creation 1 2 3
2 x System Call Overhead 1 2 3 4 5 6 7 8 9 10
2 x Shell Scripts (1 concurrent) 1 2 3
2 x Shell Scripts (8 concurrent) 1 2 3
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: *****: GNU/Linux
OS: GNU/Linux -- 2.6.32-5-686 -- #1 SMP Mon Oct 3 04:15:24 UTC 2011
Machine: i686 (unknown)
Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
CPU 0: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: QEMU Virtual CPU version (cpu64-rhel6) (4999.7 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
15:18:58 up 3 days, 16:13, 3 users, load average: 0.00, 1.38, 2.17; runlevel 2
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 15:18:58 - 15:46:59
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 11080325.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2180.4 MWIPS (10.2 s, 7 samples)
Execl Throughput 268.7 lps (29.8 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 437718.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 143590.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 653738.3 KBps (30.0 s, 2 samples)
Pipe Throughput 1113214.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 70348.3 lps (10.0 s, 7 samples)
Process Creation 736.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 913.0 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 201.4 lpm (60.3 s, 2 samples)
System Call Overhead 1694785.2 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 11080325.5 949.5
Double-Precision Whetstone 55.0 2180.4 396.4
Execl Throughput 43.0 268.7 62.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 437718.2 1105.3
File Copy 256 bufsize 500 maxblocks 1655.0 143590.4 867.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 653738.3 1127.1
Pipe Throughput 12440.0 1113214.1 894.9
Pipe-based Context Switching 4000.0 70348.3 175.9
Process Creation 126.0 736.7 58.5
Shell Scripts (1 concurrent) 42.4 913.0 215.3
Shell Scripts (8 concurrent) 6.0 201.4 335.7
System Call Overhead 15000.0 1694785.2 1129.9
========
System Benchmarks Index Score 404.4
------------------------------------------------------------------------
Benchmark Run: Sun Nov 13 2011 15:46:59 - 16:15:11
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 18055615.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3907.2 MWIPS (9.0 s, 7 samples)
Execl Throughput 767.4 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 278454.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 84660.8 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 614216.4 KBps (30.0 s, 2 samples)
Pipe Throughput 2114469.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 131640.0 lps (10.0 s, 7 samples)
Process Creation 1610.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 1998.5 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 291.5 lpm (60.2 s, 2 samples)
System Call Overhead 2019926.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 18055615.2 1547.2
Double-Precision Whetstone 55.0 3907.2 710.4
Execl Throughput 43.0 767.4 178.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 278454.9 703.2
File Copy 256 bufsize 500 maxblocks 1655.0 84660.8 511.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 614216.4 1059.0
Pipe Throughput 12440.0 2114469.5 1699.7
Pipe-based Context Switching 4000.0 131640.0 329.1
Process Creation 126.0 1610.0 127.8
Shell Scripts (1 concurrent) 42.4 1998.5 471.3
Shell Scripts (8 concurrent) 6.0 291.5 485.8
System Call Overhead 15000.0 2019926.7 1346.6
========
System Benchmarks Index Score 586.2
The GeekBench testing score is not really that impressive as well, at least for a 512MB RAM VPS:
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 i686
Model: Linux PC (QEMU Virtual CPU version (cpu64-rhel6))
Motherboard: Unknown Motherboard
Processor: QEMU Virtual CPU version (cpu64-rhel6)
Processor ID: GenuineIntel Family 6 Model 13 Stepping 3
Logical Processors: 2
Physical Processors: 1
Processor Frequency: 2.50 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: 503 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: QEMU Virtual CPU version (cpu64-rhel6)
Processor Cores: 2
Integer
Blowfish
single-threaded scalar 1759 |||||||
multi-threaded scalar 2992 |||||||||||
Text Compress
single-threaded scalar 1928 |||||||
multi-threaded scalar 2314 |||||||||
Text Decompress
single-threaded scalar 1722 ||||||
multi-threaded scalar 3274 |||||||||||||
Image Compress
single-threaded scalar 1664 ||||||
multi-threaded scalar 2346 |||||||||
Image Decompress
single-threaded scalar 1356 |||||
multi-threaded scalar 1518 ||||||
Lua
single-threaded scalar 3264 |||||||||||||
multi-threaded scalar 5052 ||||||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1842 |||||||
multi-threaded scalar 3339 |||||||||||||
Dot Product
single-threaded scalar 3346 |||||||||||||
multi-threaded scalar 6241 ||||||||||||||||||||||||
single-threaded vector 2483 |||||||||
multi-threaded vector 4461 |||||||||||||||||
LU Decomposition
single-threaded scalar 1979 |||||||
multi-threaded scalar 1951 |||||||
Primality Test
single-threaded scalar 2706 ||||||||||
multi-threaded scalar 3792 |||||||||||||||
Sharpen Image
single-threaded scalar 5261 |||||||||||||||||||||
multi-threaded scalar 8383 |||||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 4222 ||||||||||||||||
multi-threaded scalar 7705 ||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 1837 |||||||
Write Sequential
single-threaded scalar 2857 |||||||||||
Stdlib Allocate
single-threaded scalar 2017 ||||||||
Stdlib Write
single-threaded scalar 1103 ||||
Stdlib Copy
single-threaded scalar 677 ||
Stream
Stream Copy
single-threaded scalar 1624 ||||||
single-threaded vector 1700 ||||||
Stream Scale
single-threaded scalar 1558 ||||||
single-threaded vector 1723 ||||||
Stream Add
single-threaded scalar 1385 |||||
single-threaded vector 1698 ||||||
Stream Triad
single-threaded scalar 1548 ||||||
single-threaded vector 1304 |||||
Integer Score: 2432 |||||||||
Floating Point Score: 4122 ||||||||||||||||
Memory Score: 1698 ||||||
Stream Score: 1567 ||||||
Overall Geekbench Score: 2790 |||||||||||
And testing again showed similar results:
System Information
Platform: Linux x86 (32-bit)
Compiler: GCC 4.1.2 20070925 (Red Hat 4.1.2-33)
Operating System: Linux 2.6.32-5-686 i686
Model: Linux PC (QEMU Virtual CPU version (cpu64-rhel6))
Motherboard: Unknown Motherboard
Processor: QEMU Virtual CPU version (cpu64-rhel6)
Processor ID: GenuineIntel Family 6 Model 13 Stepping 3
Logical Processors: 2
Physical Processors: 1
Processor Frequency: 2.50 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: 503 MB
Memory Type: N/A
SIMD: 1
BIOS: N/A
Processor Model: QEMU Virtual CPU version (cpu64-rhel6)
Processor Cores: 2
Integer
Blowfish
single-threaded scalar 1757 |||||||
multi-threaded scalar 2933 |||||||||||
Text Compress
single-threaded scalar 1836 |||||||
multi-threaded scalar 3622 ||||||||||||||
Text Decompress
single-threaded scalar 1722 ||||||
multi-threaded scalar 3096 ||||||||||||
Image Compress
single-threaded scalar 1699 ||||||
multi-threaded scalar 2668 ||||||||||
Image Decompress
single-threaded scalar 1332 |||||
multi-threaded scalar 1796 |||||||
Lua
single-threaded scalar 3243 ||||||||||||
multi-threaded scalar 6052 ||||||||||||||||||||||||
Floating Point
Mandelbrot
single-threaded scalar 1837 |||||||
multi-threaded scalar 3133 ||||||||||||
Dot Product
single-threaded scalar 3325 |||||||||||||
multi-threaded scalar 6246 ||||||||||||||||||||||||
single-threaded vector 2450 |||||||||
multi-threaded vector 4039 ||||||||||||||||
LU Decomposition
single-threaded scalar 2163 ||||||||
multi-threaded scalar 2193 ||||||||
Primality Test
single-threaded scalar 2743 ||||||||||
multi-threaded scalar 4058 ||||||||||||||||
Sharpen Image
single-threaded scalar 5347 |||||||||||||||||||||
multi-threaded scalar 7786 |||||||||||||||||||||||||||||||
Blur Image
single-threaded scalar 4224 ||||||||||||||||
multi-threaded scalar 7647 ||||||||||||||||||||||||||||||
Memory
Read Sequential
single-threaded scalar 1919 |||||||
Write Sequential
single-threaded scalar 2721 ||||||||||
Stdlib Allocate
single-threaded scalar 1988 |||||||
Stdlib Write
single-threaded scalar 1057 ||||
Stdlib Copy
single-threaded scalar 560 ||
Stream
Stream Copy
single-threaded scalar 1711 ||||||
single-threaded vector 1892 |||||||
Stream Scale
single-threaded scalar 1489 |||||
single-threaded vector 1724 ||||||
Stream Add
single-threaded scalar 1453 |||||
single-threaded vector 1629 ||||||
Stream Triad
single-threaded scalar 1516 ||||||
single-threaded vector 1232 ||||
Integer Score: 2646 ||||||||||
Floating Point Score: 4085 ||||||||||||||||
Memory Score: 1649 ||||||
Stream Score: 1580 ||||||
Overall Geekbench Score: 2843 |||||||||||
Customer Service and Support
In our previous review for BuyVM’s OpenVZ product line, although the response on IRC channel were quick, responding tickets sometimes take several days. However, this time I am happy to say that BuyVM seems to be able to respond to tickets much faster, one of my tickets sent to them on a Monday at 8:57pm was responded by 9:47pm with the actual problem resolved (instead of “we are looking into it”). Another ticket that I have raised on 9:55pm was answered by 6:48AM, which is not bad as well, particularly if we bear in mind (and BuyVM have indicated this almost everywhere, in their sign up page, welcome email and on their website), that the services they provide is unmanaged, which is one of the reasons why they could afford to offer the service at a low price.
Conclusion
We have reviewed both the cheapest KVM offering from BuyVM as well as the not-so-cheap (but still relatively cheap compare to other KVM providers!) 512MB KVM plan. Both offers pretty good disk I/O but the networks may not be too optimal if you need heavy uploading and downloading. The CPU resources seems to be pretty low as well, which unfortunately resulted in some not-so-great benchmarks. However, if you are just looking for a VPS to “play around with”, 25 per year is a good price if you need full virtualization, or if you are looking for a Windows VPS at a cheap price. Otherwise, I would still recommend their OpenVZ products since it has been around a lot longer and have performed much better in the past 96MB.com reviews.
P.S.: Sorry for the late posting on this one, my three-year old laptop seems to get more and more grumpy and decided to shut itself off once a while without any warning. And I just found, as I type, the option to enable auto-recovery, arghh…
P.S. again: The next few weeks will be the end-of-term exam period for me, as such, I might not have the time to publish reviews “frequently” as I would be busy studying. However the good news is that I will be done before Christmas, so we will definitely have a VPS review as my Christmas present to you!
I’m a happy BuyVM user with both an OpenVZ and KVM node.
I haven’t noticed any bandwidth issues.
From my OpenVZ node:
wget cachefly.cachefly.net/100mb.test -O /dev/null
--2011-12-04 10:17:42-- http://cachefly.cachefly.net/100mb.test
Resolving cachefly.cachefly.net... 205.234.175.175
Connecting to cachefly.cachefly.net|205.234.175.175|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `/dev/null'
100%[======================================>] 104,857,600 40.7M/s in 2.5s
2011-12-04 10:17:44 (40.7 MB/s) - `/dev/null' saved [104857600/104857600]
Download from my YardVPS 128MB OpenVZ node (also West Coast)
root@us2:~# wget http://xxx/100mb.test -O /dev/null
--2011-12-04 10:24:29-- http://us.busisoft.net/100mb.test
Resolving us.busisoft.net... 216.24.197.14
Connecting to us.busisoft.net|216.24.197.14|:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/x-tar]
Saving to: `/dev/null'
100%[======================================>] 104,857,600 11.0M/s in 9.2s
2011-12-04 10:24:38 (10.9 MB/s) - `/dev/null' saved [104857600/104857600]
Unfortunately I can’t test upload, as YardVPS seems to have terrible download performance, at least at the moment.
@Peter: Thanks for the comments, do you mind running a cachefly download test on the KVM VPS? I just want to see if there is something with the KVM VPS that was causing the issue..thanks a lot!
Sure no problem. The same 512MB KVM plan as you reviewed, however running Windows 2003 32-bit and the emulated Intel PRO/1000 Adaptor.
I used a Win32 build of wget from http://users.ugent.be/~bpuype/wget/
C:\Documents and Settings\Administrator\Desktop>wget cachefly.cachefly.net/100mb
.test -O nul
–2011-12-05 19:59:11– 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: `nul’
100%[======================================>] 104,857,600 12.2M/s in 9.1s
2011-12-05 19:59:20 (11.0 MB/s) – `nul’ saved [104857600/104857600]
Changing to the Virtio NIC (it took a full shutdown before the changes took place, a reboot initiated from the client didn’t affect it):
For completeness of the OpenVZ at a similar time:
root@us2:~# wget cachefly.cachefly.net/100mb.test -O /dev/null
–2011-12-05 13:23:47– 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 38.5M/s in 2.6s
2011-12-05 13:23:49 (38.5 MB/s) – `/dev/null’ saved [104857600/104857600]
So it’s looking like a KVM issue…
Ok ran a test between the two VMs over the internal LAN link, this is the result from the KVM machine.
I think I’ll need to raise a ticket, I think something may be misconfigured.
@Peter: Thanks so much for the reply, do let us know if BuyVM/Frantech get back to you about this, since I think it is perhaps the biggest issue that is preventing people from enjoy the cheap service from BuyVM’s KVM line!
I’ve certainly been very happy with BuyVM, they’re fantastic value.
I did log a ticket as an FYI to them, it has been closed, but no improvement as yet. Hopefully it’s being looked into. I’m not overly worried though, 20Mbyte/sec from Cachefly is still over 100Mbit!
Thanks for the review. I might grab one soon
@Bob: I cancelled my 512MB box after the testing, so there might be your chance
I happen to have some routing issues with BuyVM because of he.net. Can you experience this also using traceroutes?
Pingback: 96MB Low End VPS Review Part XXXVIII – servaRICA | | 96MB.com96MB.com
Pingback: BuyVM – $25/Year KVM VPS in Buffalo, NY – Low End Box