Register or log in (lost password?):

Low End Talk

Hosting Websites on Bare Minimum VPS/Dedicated Servers

Low End Talk » Low End Talk

New eNS Cloud VPS

(10 posts)

  1. david3
    Member

    I've got a new eNS Cloud VPS from eNetSouth, and thought I'd post a few observations here rather than clutter the post on lowendbox:

    http://www.lowendbox.com/blog/ens-cloud-4-45-200mb-vmware-vps-exclusive-offer/

    Here are the results from Unix Bench 5.1.2:

    ========================================================================
   BYTE UNIX Benchmarks (Version 5.1.2)

   System: GNU/Linux
   OS: GNU/Linux -- 2.6.18-194.8.1.el5 -- #1 SMP Thu Jul 1 19:07:06 EDT 2010
   Machine: i686 (i386)
   Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
   CPU 0: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4533.5 bogomips)
          x86-64, MMX, Physical Address Ext
   CPU 1: Intel(R) Xeon(R) CPU E5520 @ 2.27GHz (4537.1 bogomips)
          x86-64, MMX, Physical Address Ext
   10:57:19 up  1:25,  2 users,  load average: 0.00, 0.02, 0.00; runlevel

------------------------------------------------------------------------
Benchmark Run: Sat Aug 07 2010 10:57:19 - 11:25:10
2 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       11047335.2 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     2146.7 MWIPS (9.8 s, 7 samples)
Execl Throughput                               4722.8 lps   (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        446862.3 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          125883.1 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       1123697.0 KBps  (30.0 s, 2 samples)
Pipe Throughput                              869599.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  55894.3 lps   (10.0 s, 7 samples)
Process Creation                              17545.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                   6324.0 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1199.8 lpm   (60.1 s, 2 samples)
System Call Overhead                         780754.1 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   11047335.2    946.6
Double-Precision Whetstone                       55.0       2146.7    390.3
Execl Throughput                                 43.0       4722.8   1098.3
File Copy 1024 bufsize 2000 maxblocks          3960.0     446862.3   1128.4
File Copy 256 bufsize 500 maxblocks            1655.0     125883.1    760.6
File Copy 4096 bufsize 8000 maxblocks          5800.0    1123697.0   1937.4
Pipe Throughput                               12440.0     869599.6    699.0
Pipe-based Context Switching                   4000.0      55894.3    139.7
Process Creation                                126.0      17545.8   1392.5
Shell Scripts (1 concurrent)                     42.4       6324.0   1491.5
Shell Scripts (8 concurrent)                      6.0       1199.8   1999.7
System Call Overhead                          15000.0     780754.1    520.5
                                                                   ========
System Benchmarks Index Score                                         850.1

------------------------------------------------------------------------
Benchmark Run: Sat Aug 07 2010 11:25:10 - 11:53:13
2 CPUs in system; running 2 parallel copies of tests

Dhrystone 2 using register variables       21162088.8 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     4184.4 MWIPS (10.4 s, 7 samples)
Execl Throughput                               9707.2 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks        212741.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks           59077.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks        627942.8 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1716555.3 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 538898.7 lps   (10.0 s, 7 samples)
Process Creation                              31942.8 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  10026.9 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   1582.4 lpm   (60.0 s, 2 samples)
System Call Overhead                        1486013.3 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   21162088.8   1813.4
Double-Precision Whetstone                       55.0       4184.4    760.8
Execl Throughput                                 43.0       9707.2   2257.5
File Copy 1024 bufsize 2000 maxblocks          3960.0     212741.4    537.2
File Copy 256 bufsize 500 maxblocks            1655.0      59077.9    357.0
File Copy 4096 bufsize 8000 maxblocks          5800.0     627942.8   1082.7
Pipe Throughput                               12440.0    1716555.3   1379.9
Pipe-based Context Switching                   4000.0     538898.7   1347.2
Process Creation                                126.0      31942.8   2535.1
Shell Scripts (1 concurrent)                     42.4      10026.9   2364.8
Shell Scripts (8 concurrent)                      6.0       1582.4   2637.3
System Call Overhead                          15000.0    1486013.3    990.7
                                                                   ========
System Benchmarks Index Score                                        1277.1

    Here's a disk test:

    hdparm -t /dev/sda1
/dev/sda1:
 Timing buffered disk reads:  100 MB in  0.77 seconds = 130.10 MB/sec

    So far it's looking good.

    Posted 1 year ago #

  2. drmike
    Member

    From reading the comments, the admin seems to know what he is doing. He's a bit rough but he gets points for knowing the technical stuff.

    Posted 1 year ago #

  3. david3
    Member

    The network connectivity is really excellent. I'm thinking about moving a few small websites from my Linode over, and if it continues to do well, maybe I'll move all of them.

    I'm testing a Xen PV Thrust VPS also, but this eNS Cloud VPS is looking better so far.

    Posted 1 year ago #

  4. LowEndAdmin
    Key Master

    With Xen PV I don't think it would be able to run some of the latest distros properly (Fedora 13, Ubuntu 10.04 for example) unless they are able to patch the latest upstream kernel like Linode has done.

    VMWare based VPS would have no such problem. I'll be interested once they have got their west coast VPS ready though.

    Posted 1 year ago #

  5. david3
    Member

    Yeah, ping time is higher for me to Chicago vs. the West Coast, but the bandwidth consistency has been so good so far that I don't mind.

    It's probably just a personal routing thing, but I'm able to get max download speed on my DSL line at all times from the eNS vps, whereas Linode in Fremont, CA and Thrust VPS in LA both get slower and develop sawtooth patterns at times (Linode still being better than Thrust, though).

    Posted 1 year ago #

  6. david3
    Member

    Well, I take it back. eNS had better bandwidth consistency, but tonight when I checked it, it was about the same as Linode and ThrustVPS. In any case, I'm pretty sure it's not their fault, it's just my personal isp and routing situation.

    Posted 1 year ago #

  7. david3
    Member

    For comparison, here are examples of what I'm talking about (these are all from my DSL line in Asia).

    This first bandwidth graph is from 3 days ago, and I continued to experience similar results up until tonight (during the evening):

    http://a.imageshack.us/img96/9507/enscloudspeedtest01.png

    Here's a bandwidth graph from tonight, where they're all about the same:

    http://a.imageshack.us/img64/7236/enscloudspeedtest02.png

    I think this is really a reflection on my isp, and not any of the 3 VPS services.

    Posted 1 year ago #

  8. david3
    Member

    Just to be complete, here's a comparison during the morning (all 3 are fast):

    http://a.imageshack.us/img713/6025/enscloudspeedtest03.png

    Posted 1 year ago #

  9. kprice
    Member

    It's most definitely your provider. A mere 1 Mbps home connection isn't going to allow you to truly test the speed of these providers. Use "mtr" in conjunction with your simple download tests to determine where the bottleneck may lie.

    Posted 1 year ago #

  10. david3
    Member

    Well, I was trying to see if certain services had better routing for me to Asia. Most of the time it's fine, but if things get congested, sometimes some routes seem to perform better than others.

    I don't think it's always consistent, though. One day one route may be better, and the next day a different route may be better. I don't blame the vps services.

    It's handy to have access to multiple servers in different data centers with different routing.

    Posted 1 year ago #

RSS feed for this topic

Reply

You must log in to post.