96MB Low End VPS Community Member Review Part 3 – DireVPS 4GB RAM OpenVZ

Preface

The low end VPS market has received quite a bit of “shock” recently when DireVPS, which is an subsidiary of OpenITC/XenVZ, announced that they would be selling what they called “carefully oversold” VPS, offering 4GB of RAM at a price of less than 7USD per month. Soon after that, we saw a lot more offering from other VPS providers offering VPS at such an incredibly low price. Clearly DireVPS was the leader in this change and how have they been doing recently? The most valuable contributor of 96MB, jcaleb, has decided to sign up with them and take a look. Thanks jcaleb and without further to do, let’s take a look at his review:

Introduction

2GB OpenVZ has been more and more common nowadays, that many providers shoots up that offer from time to time. But last February (2013), OpenITC surprised the LET/LEB community by giving a new landmark to high RAM offer at LEB Prices: a whopping 4GB RAM, 100GB disk, and 2TB BW for just under $7 USD. It’s under their new brand DireVPS.

Despite the very generous amount of resources, it was received with skepticism/criticism and did not sold well as discussed in here. In my opinion, the fault is with how they promoted the new brand. The website is full of red flags, that will really scare away potential customers, instead of attract them.

Although the owners does not seem to be marketing genius, I still believe that DireVPS is worth giving a shot. As this is the same group of people that is running XenVZ – a brand with high reputation and trusted by many.

Disclaimer: I am doing this review for recreational purposes. I am not receiving any compensation from OpenITC/DireVPS in free service or financial terms. Although I am one of 14 people sean gave a free month of DireVPS service, for giving community feedback. I did not claim the free VPS, and did not contact OpenITC. I have paid for the VPS, under a different name, to be able to write this review.

General Information and Background

The offer is sweet and simple. The main attraction is their high amount of RAM, disk space, and bandwidth. They also have IPv6, which is very good news! The catch is that they don’t have support and they offer no refund.

direvps_deal

direvps_features

The turn-off’s for me, are in their FAQ:

direvps_prob1

their support policy.

direvps_prob3

and most importantly, their disclaimer in the home page:

direvps_prob2

I will not expound my opinion on above. It is what it is, and its upto the readers to interpret their meaning.

Signing up

They have the same sign up page for all the brands under OpenITC. Some of the sign-up options below are not relevant, as your base package already have large resources.

direvps_opt1

direvps_opt2

You can buy more IP’s and also control panel license.

direvps_opt3

The welcome email looks something like this. They don’t create your VPS instance right away. You have to provision it yourself.

direvps_welcome

They don’t use SolusVM, but their own control panel. In my opinion, it is also easy to use, unlike other custom control panel that I have tried.

direvps_panel

Provisioning your VPS is simple and straightforward.

direvps_prov1

direvps_prov2

direvps_prov3

direvps_prov4

The choice of OS templates are just basic. But they do have Ubuntu 12.04 which is very important to many people nowadays.

direvps_os_choice

Test on the VPS

For the tests carried out below, I have used Ubuntu 12.04 32 bit OS. I switch from 32/64 bit when doing Unix benchmark and geekbench, and report whichever is higher.

Checking memory when OS template was initially loaded reveals it’s a .32 kernel with 1GB vSwap memory. The template is trimmed down, as their XenVZ brand have low memory resources. But minimal template does not matter to us, we have so many.

    free -m
                 total       used       free     shared    buffers     cached
    Mem:          4096         14       4081          0          0          8
    -/+ buffers/cache:          5       4090
    Swap:         1024          0       1024

Top showing the processes running.

    top - 17:54:21 up 0 min,  1 user,  load average: 0.00, 0.00, 0.00
    Tasks:  16 total,   1 running,  15 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:   4194304k total,    14776k used,  4179528k free,        0k buffers
    Swap:  1048576k total,        0k used,  1048576k free,     8696k cached

      PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND                                                                                                    
        1 root      20   0  3348 1744 1280 S    0  0.0   0:00.09 init                                                                                                       
        2 root      20   0     0    0    0 S    0  0.0   0:00.00 kthreadd/3220                                                                                              
        3 root      20   0     0    0    0 S    0  0.0   0:00.00 khelper/3220                                                                                               
      118 root      20   0  2772  708  528 S    0  0.0   0:00.00 upstart-udev-br                                                                                            
      124 root      20   0  2788  996  724 S    0  0.0   0:00.00 udevd                                                                                                      
      156 root      20   0  2784  636  356 S    0  0.0   0:00.00 udevd                                                                                                      
      157 root      20   0  2784  636  356 S    0  0.0   0:00.00 udevd                                                                                                      
      174 root      20   0  2172  560  492 S    0  0.0   0:00.00 sh                                                                                                         
      187 root      20   0  2784  512  368 S    0  0.0   0:00.00 upstart-socket-                                                                                            
      219 root      20   0  6628 2356 1924 S    0  0.1   0:00.00 sshd                                                                                                       
      239 root      20   0  2564  828  668 S    0  0.0   0:00.00 cron                                                                                                       
      272 syslog    20   0  2348  736  600 S    0  0.0   0:00.00 syslogd                                                                                                    
      293 root      20   0  9592 3096 2464 S    0  0.1   0:00.08 sshd                                                                                                       
      305 root      20   0  3428 1772 1424 S    0  0.0   0:00.00 bash                                                                                                       
      316 root      20   0  2088  452  388 S    0  0.0   0:00.00 sleep                                                                                                      
      317 root      20   0  2668 1168  972 R    0  0.0   0:00.00 top                                                                                                        

And htop output:

direvps_htop

Hard drive space confirms we have 100GB to work with.

    df -h
    Filesystem      Size  Used Avail Use% Mounted on
    /dev/simfs      100G  415M  100G   1% /
    varrun          2.0G   40K  2.0G   1% /run
    varlock         2.0G     0  2.0G   0% /var/lock
    none            5.0M     0  5.0M   0% /run/lock
    none            2.0G     0  2.0G   0% /run/shm

And the iNodes are set to pretty high values:

    df -i
    Filesystem       Inodes IUsed    IFree IUse% Mounted on
    /dev/simfs     52428800 23395 52405405    1% /
    varrun           524288    47   524241    1% /run
    varlock          524288     1   524287    1% /var/lock
    none             524288     1   524287    1% /run/lock
    none             524288     1   524287    1% /run/shm

The VPS is fairly stable and little used, as shown in the output of uptime:

    uptime
    17:59:10 up 5 min,  1 user,  load average: 0.00, 0.00, 0.00

vmstat shows no IOwait as well, which is also a good sign:

    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 4179608      0   8980    0    0     0     2    0 100617  0  0 100  0

CPUInfo shows there are four CPU cores available but throttled only to 800 MHz per core. This could be either good news or bad news, depending on how you think of it.

cat /proc/cpuinfo
processor	: 0
vendor_id	: GenuineIntel
cpu family	: 6
model		: 44
model name	: Intel(R) Xeon(R) CPU           L5630  @ 2.13GHz
stepping	: 2
cpu MHz		: 800.024
cache size	: 12288 KB
physical id	: 1
siblings	: 8
core id		: 0
cpu cores	: 4
apicid		: 32
initial 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 pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 sse4_2 popcnt aes lahf_lm ida arat epb dts tpr_shadow vnmi flexpriority ept vpid
bogomips	: 4266.79
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management:

processor	: 1
vendor_id	: GenuineIntel
cpu family	: 6
model		: 44
model name	: Intel(R) Xeon(R) CPU           L5630  @ 2.13GHz
stepping	: 2
cpu MHz		: 800.024
cache size	: 12288 KB
physical id	: 1
siblings	: 8
core id		: 10
cpu cores	: 4
apicid		: 52
initial apicid	: 52
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 pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good xtopology nonstop_tsc aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 cx16 xtpr pdcm dca sse4_1 sse4_2 popcnt aes lahf_lm ida arat epb dts tpr_shadow vnmi flexpriority ept vpid
bogomips	: 4266.79
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management:

Meminfo:

cat /proc/meminfo
MemTotal:        4194304 kB
MemFree:         4179612 kB
Cached:             9004 kB
Active:            10596 kB
Inactive:           1540 kB
Active(anon):       2100 kB
Inactive(anon):     1032 kB
Active(file):       8496 kB
Inactive(file):      508 kB
Unevictable:           0 kB
Mlocked:               0 kB
SwapTotal:       1048576 kB
SwapFree:        1048576 kB
Dirty:                 0 kB
Writeback:             0 kB
AnonPages:          3132 kB
Shmem:              2600 kB
Slab:               2548 kB
SReclaimable:       1368 kB
SUnreclaim:         1180 kB
tim

Disk I/O is fast

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, 4.23812 s, 253 MB/s

Testing again showed almost the same result:

16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 4.2675 s, 252 MB/s

The ioping results are very consistent as well.

    ioping  -c 10 .
    4096 bytes from . (simfs /vps/3220): request=1 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=2 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=3 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=4 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=5 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=6 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=7 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=8 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=9 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=10 time=0.1 ms

    --- . (simfs /vps/3220) ioping statistics ---
    10 requests completed in 9002.8 ms, 9132 iops, 35.7 mb/s
    min/avg/max/mdev = 0.1/0.1/0.1/0.0 ms

Testing again showed similar results:

    ioping  -c 10 .
    4096 bytes from . (simfs /vps/3220): request=1 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=2 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=3 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=4 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=5 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=6 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=7 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=8 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=9 time=0.1 ms
    4096 bytes from . (simfs /vps/3220): request=10 time=0.1 ms

    --- . (simfs /vps/3220) ioping statistics ---
    10 requests completed in 9002.8 ms, 8347 iops, 32.6 mb/s
    min/avg/max/mdev = 0.1/0.1/0.1/0.0 ms

Using the FreeVPS benchmark script, shows the network network speed. It seems capped at 100mbps and not impressive when outside UK.

    bash bench.sh
    CPU model :  Intel(R) Xeon(R) CPU           L5630  @ 2.13GHz
    Number of cores : 2
    CPU frequency :  800.024 MHz
    Total amount of ram : 4096 MB
    Total amount of swap : 1024 MB
    System uptime :   13 min,       
    Download speed from CacheFly: 8.18MB/s 
    Download speed from Coloat, Atlanta GA: 4.61MB/s 
    Download speed from Softlayer, Dallas, TX: 2.68MB/s 
    Download speed from Linode, Tokyo, JP: 2.20MB/s 
    Download speed from Linode, London, UK: 11.2MB/s 
    Download speed from Leaseweb, Haarlem, NL: 8.96MB/s 
    Download speed from Softlayer, Singapore: 1.59MB/s 
    Download speed from Softlayer, Seattle, WA: 2.93MB/s 
    Download speed from Softlayer, San Jose, CA: 2.07MB/s 
    Download speed from Softlayer, Washington, DC: 4.04MB/s 
    I/O speed :  257 MB/s

Testing again shows relatively same results:

    bash bench.sh
    CPU model :  Intel(R) Xeon(R) CPU           L5630  @ 2.13GHz
    Number of cores : 2
    CPU frequency :  800.024 MHz
    Total amount of ram : 4096 MB
    Total amount of swap : 1024 MB
    System uptime :   18 min,       
    Download speed from CacheFly: 3.43MB/s 
    Download speed from Coloat, Atlanta GA: 4.76MB/s 
    Download speed from Softlayer, Dallas, TX: 3.08MB/s 
    Download speed from Linode, Tokyo, JP: 1.98MB/s 
    Download speed from Linode, London, UK: 11.2MB/s 
    Download speed from Leaseweb, Haarlem, NL: 9.09MB/s 
    Download speed from Softlayer, Singapore: 1.28MB/s 
    Download speed from Softlayer, Seattle, WA: 3.46MB/s 
    Download speed from Softlayer, San Jose, CA: 2.40MB/s 
    Download speed from Softlayer, Washington, DC: 3.26MB/s 
    I/O speed :  255 MB/s

For benchmarks, the throttled CPU affected the result, if you would compare with other providers.

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   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: 57: GNU/Linux
   OS: GNU/Linux -- 2.6.32-042stab072.10 -- #1 SMP Wed Jan 16 18:54:05 MSK 2013
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU L5630 @ 2.13GHz (4266.8 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   CPU 1: Intel(R) Xeon(R) CPU L5630 @ 2.13GHz (4266.8 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   18:11:22 up  1:08,  2 users,  load average: 0.08, 0.02, 0.01; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Sun Mar 03 2013 18:11:22 - 18:39:53
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       12249470.7 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     1989.5 MWIPS (9.9 s, 7 samples)
Execl Throughput                               1580.5 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        359012.9 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           97759.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        997133.5 KBps  (30.0 s, 2 samples)
Pipe Throughput                              703305.1 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 108396.1 lps   (10.0 s, 7 samples)
Process Creation                               5620.4 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3401.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    448.7 lpm   (60.1 s, 2 samples)
System Call Overhead                         796506.7 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   12249470.7   1049.7
Double-Precision Whetstone                       55.0       1989.5    361.7
Execl Throughput                                 43.0       1580.5    367.6
File Copy 1024 bufsize 2000 maxblocks          3960.0     359012.9    906.6
File Copy 256 bufsize 500 maxblocks            1655.0      97759.7    590.7
File Copy 4096 bufsize 8000 maxblocks          5800.0     997133.5   1719.2
Pipe Throughput                               12440.0     703305.1    565.4
Pipe-based Context Switching                   4000.0     108396.1    271.0
Process Creation                                126.0       5620.4    446.1
Shell Scripts (1 concurrent)                     42.4       3401.9    802.3
Shell Scripts (8 concurrent)                      6.0        448.7    747.9
System Call Overhead                          15000.0     796506.7    531.0
                                                                   ========
System Benchmarks Index Score                                         612.7

------------------------------------------------------------------------
Benchmark Run: Sun Mar 03 2013 18:39:53 - 19:10:50
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       12068016.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3979.0 MWIPS (9.9 s, 7 samples)
Execl Throughput                               1565.6 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        361951.5 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           97285.5 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1004438.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                              704878.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 105103.2 lps   (10.0 s, 7 samples)
Process Creation                               5461.1 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3378.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    444.3 lpm   (60.1 s, 2 samples)
System Call Overhead                         786294.8 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   12068016.8   1034.1
Double-Precision Whetstone                       55.0       3979.0    723.5
Execl Throughput                                 43.0       1565.6    364.1
File Copy 1024 bufsize 2000 maxblocks          3960.0     361951.5    914.0
File Copy 256 bufsize 500 maxblocks            1655.0      97285.5    587.8
File Copy 4096 bufsize 8000 maxblocks          5800.0    1004438.9   1731.8
Pipe Throughput                               12440.0     704878.3    566.6
Pipe-based Context Switching                   4000.0     105103.2    262.8
Process Creation                                126.0       5461.1    433.4
Shell Scripts (1 concurrent)                     42.4       3378.9    796.9
Shell Scripts (8 concurrent)                      6.0        444.3    740.5
System Call Overhead                          15000.0     786294.8    524.2
                                                                   ========
System Benchmarks Index Score                                         643.7

Testing again showed similar results:

   #    #  #    #  #  #    #          #####   ######  #    #   ####   #    #
   #    #  ##   #  #   #  #           #    #  #       ##   #  #    #  #    #
   #    #  # #  #  #    ##            #####   #####   # #  #  #       ######
   #    #  #  # #  #    ##            #    #  #       #  # #  #       #    #
   #    #  #   ##  #   #  #           #    #  #       #   ##  #    #  #    #
    ####   #    #  #  #    #          #####   ######  #    #   ####   #    #

   Version 5.1.3                      Based on the Byte Magazine Unix Benchmark

   Multi-CPU version                  Version 5 revisions by Ian Smith,
                                      Sunnyvale, CA, USA
   January 13, 2011                   johantheghost at yahoo period com


1 x Dhrystone 2 using register variables  1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone  1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput  1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks  1 2 3

1 x File Copy 256 bufsize 500 maxblocks  1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks  1 2 3

1 x Pipe Throughput  1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching  1 2 3 4 5 6 7 8 9 10

1 x Process Creation  1 2 3

1 x System Call Overhead  1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent)  1 2 3

1 x Shell Scripts (8 concurrent)  1 2 3

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: 57: GNU/Linux
   OS: GNU/Linux -- 2.6.32-042stab072.10 -- #1 SMP Wed Jan 16 18:54:05 MSK 2013
   Machine: x86_64 (x86_64)
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
   CPU 0: Intel(R) Xeon(R) CPU L5630 @ 2.13GHz (4266.8 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   CPU 1: Intel(R) Xeon(R) CPU L5630 @ 2.13GHz (4266.8 bogomips)
          Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
   19:19:14 up  2:16,  2 users,  load average: 0.00, 1.11, 1.99; runlevel 2

------------------------------------------------------------------------
Benchmark Run: Sun Mar 03 2013 19:19:14 - 19:47:42
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       12217326.3 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2000.5 MWIPS (9.9 s, 7 samples)
Execl Throughput                               1590.0 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        354646.2 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           96583.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        996923.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                              717088.7 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 108656.7 lps   (10.0 s, 7 samples)
Process Creation                               5584.5 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3398.8 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    446.5 lpm   (60.1 s, 2 samples)
System Call Overhead                         794744.2 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   12217326.3   1046.9
Double-Precision Whetstone                       55.0       2000.5    363.7
Execl Throughput                                 43.0       1590.0    369.8
File Copy 1024 bufsize 2000 maxblocks          3960.0     354646.2    895.6
File Copy 256 bufsize 500 maxblocks            1655.0      96583.7    583.6
File Copy 4096 bufsize 8000 maxblocks          5800.0     996923.1   1718.8
Pipe Throughput                               12440.0     717088.7    576.4
Pipe-based Context Switching                   4000.0     108656.7    271.6
Process Creation                                126.0       5584.5    443.2
Shell Scripts (1 concurrent)                     42.4       3398.8    801.6
Shell Scripts (8 concurrent)                      6.0        446.5    744.2
System Call Overhead                          15000.0     794744.2    529.8
                                                                   ========
System Benchmarks Index Score                                         612.3

------------------------------------------------------------------------
Benchmark Run: Sun Mar 03 2013 19:47:42 - 20:18:42
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       12066140.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     3971.6 MWIPS (10.0 s, 7 samples)
Execl Throughput                               1573.7 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        357243.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           95900.8 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        997728.6 KBps  (30.0 s, 2 samples)
Pipe Throughput                              702558.0 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 104997.3 lps   (10.0 s, 7 samples)
Process Creation                               5482.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   3391.1 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                    446.7 lpm   (60.1 s, 2 samples)
System Call Overhead                         792904.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   12066140.8   1033.9
Double-Precision Whetstone                       55.0       3971.6    722.1
Execl Throughput                                 43.0       1573.7    366.0
File Copy 1024 bufsize 2000 maxblocks          3960.0     357243.3    902.1
File Copy 256 bufsize 500 maxblocks            1655.0      95900.8    579.5
File Copy 4096 bufsize 8000 maxblocks          5800.0     997728.6   1720.2
Pipe Throughput                               12440.0     702558.0    564.8
Pipe-based Context Switching                   4000.0     104997.3    262.5
Process Creation                                126.0       5482.8    435.1
Shell Scripts (1 concurrent)                     42.4       3391.1    799.8
Shell Scripts (8 concurrent)                      6.0        446.7    744.5
System Call Overhead                          15000.0     792904.3    528.6
                                                                   ========
System Benchmarks Index Score                                         642.9

Naturally, Geekbench scores are also affected.

direvps_geek1

Testing again showed relatively same results:

direvps_geek2

Benefits of High RAM

Even if you have throttled CPU, depending on your application, the high RAM and fast IO can more than make up for it

A simple example I did is just restoring a production MySQL database. From default MySQL config, it takes 5 minutes.

root@57:~# ls -lh
total 115M
-rw-r--r-- 1 root root 115M Mar  4 18:31 big.sql

root@57:~# pv big.sql | mysql -u root -pXXXX big
 115MB 0:05:09 [ 380kB/s] [======================================================================>] 100%

Changing a bit the memory configuration, makes the same operation almost twice faster.

innodb_buffer_pool_size = 1024M

root@57:~# pv big.sql | mysql -u root -plonganisa6 big
 115MB 0:03:12 [ 610kB/s] [======================================================================>] 100%

This is just a simple example, as I am only making a point. I believe experienced system administrators can utilize this high amount of RAM to run servers a magnitude of 10 or 100 faster, than a VPS with 256 or 512mb RAM. As there is room for lots and lots of caching.

The Big Question

I have to admit, the reason why I was intrigued to review DireVPS, is because there was a comment in LET doubting will DireVPS really allow you to use your allocated resources?. So I spent a whole month with DireVPS, to give my experience.

First, I pulled some files and back-ups from my other VPS to fill up 92% of my allocated disk space. I waited a few days, if I will receive any complaint from my host.

angel@57:~/test1$ df -h
Filesystem      Size  Used Avail Use% Mounted on
/dev/simfs      100G   92G  8.5G  92% /
varrun          2.0G   56K  2.0G   1% /run
varlock         2.0G     0  2.0G   0% /var/lock
none            5.0M     0  5.0M   0% /run/lock
none            2.0G     0  2.0G   0% /run/shm

Having no problem using disk space, next I run multiple copies of Java programs, each consuming around 500mb each, to fill up around 3.5GB of RAM. Since no one is consuming the services of the Java program, the load is minimal. Only memory is used. I let this sit for about 1 week.

root@57:~# free -m
             total       used       free     shared    buffers     cached
Mem:          4096       3517        578          0          0         75
-/+ buffers/cache:       3441        654
Swap:         1024         76        947

Since I was not terminated yet, I tried to test a little bit more. I wrote a Java program to consume around 500mb again and also do processing. Lots of math computation, read hundred gigs of data, and write a few gig of data. Repeated every few minutes. Running multiple copies to fill up 3.5GB Ram and about 2.0 load average. I let this run again for about a week.

direvps_load1

After a week, I modified my programs to use a little more load. Again, I let it sit for about a week, using resource more or less what is shown below.

direvps_load2

I stopped monitoring my VPS after 3 weeks of usage, and just let what I was running in there. I just monitored my email if I will get notice of termination, which did not happen.

Customer Service and Support

Although DireVPS has no support, they do update their customers through email with important events and information.

Personally, I find it really difficult to swallow their support policy, above all the things I read in their website. Sure, I am responsible for my VPS as it is unmanaged. But there are gray areas where you really need to communicate with your provider. It is intriguing to know if they will really ignore any ticket you submit to them, and be completely unreasonable.

After 3 weeks of using their service, I finally found a reason to contact them without needing help with setting up my VPS. I have an issue with their control panel always showing 0GB bandwidth used, when I am sure to have consumed alot already.

After 2 days of submitting a ticket in their contact form, I have not received a reply. I was disappointed, but decided to submit again a ticket just to be fair. To my delight, I received a reply after about 7hrs.

direvps_support

Conclusion

DireVPS, hands down offers great value for money. Lots of RAM, disk space, and bandwidth all for under $7 USD. And by testing them for a month, it was shown that they will allow you to use them. Although I admit that I have not tested the bandwidth, as I am out of ideas on how to consume 2TB in under a month.

Although CPU is throttled and network is just okay, Disk IO is fast and overall performance of the VPS is satisfactory. Discussing with 96mb, we believe this is because they have not sold well, and the node I was on is not overloaded. I hope performance will not degrade much, when they become popular.

Although their website is more likely to discourage you than to convince you to sign-up, they have shown that they are very reasonable. Personally, I recommend their service, as long as you are okay with no support. Perhaps not for production, but for personal and development purposes. (E.g. as Continuous Integration server (Jenkins) that requires lots of memory and disk space.) There is no need to worry that they will just terminate your VPS with no just reason.

3 thoughts on “96MB Low End VPS Community Member Review Part 3 – DireVPS 4GB RAM OpenVZ

  1. For me, that vps is quiet reasonable, although I don’t do java or C heavily, but if you limiting for bandwith in your ISP and not dealing with torrent, this service was good.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.