Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!


Cheap Dacentec offers (dedicated) - Page 24
New on LowEndTalk? Please Register and read our Community Rules.

All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.

Cheap Dacentec offers (dedicated)

12122242627

Comments

  • FredQc said: We should get comparable results, right

    I see a kernel difference. Could that impact the benchmark?

    2.6.32-042stab108.2 vs 3.2.0-4-amd64

  • FredQcFredQc Member

    @Silvenga said:

    Yes, thanks for pointing this fact. In regards of hardware they are both almost identical, however, the Dacentec system should get a better score because of SSDs.

    Dacentec

    [root@l5520 ~]# dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 2.85554 s, 376 MB/s
    

    Delimiter

    root@L5520:~# dd if=/dev/zero of=test_$$ bs=64k count=16k conv=fdatasync && rm -f test_$$
    16384+0 records in
    16384+0 records out
    1073741824 bytes (1.1 GB) copied, 12.7004 s, 84.5 MB/s
    
  • MarkTurnerMarkTurner Member
    edited June 2015

    @FredQC:

    On the disk copy, why is a single HDD performing better than an SSD?

    Dacentec (SSD):
    File Copy 1024 bufsize 2000 maxblocks 419759.4 KBps (30.0 s, 2 samples)
    File Copy 256 bufsize 500 maxblocks 112885.4 KBps (30.0 s, 2 samples)
    File Copy 4096 bufsize 8000 maxblocks 1282817.2 KBps (30.0 s, 2 samples) Delimiter (HDD):
    File Copy 1024 bufsize 2000 maxblocks 606701.3 KBps (30.0 s, 2 samples)
    File Copy 256 bufsize 500 maxblocks 170660.8 KBps (30.0 s, 2 samples)
    File Copy 4096 bufsize 8000 maxblocks 1743948.5 KBps (30.0 s, 2 samples)

    There is something definitely bizarre about that. You should be seeing the opposite on that. Are you sure its hitting an SSD and not a HDD?

    But irrespective the rest of the CPU tests should be the same. I would run the same kernel on both and retest. I wouldn't want to benchmark with an OpenVZ kernel.

    image

  • FredQcFredQc Member

    @MarkTurner said:
    FredQC:

    On the disk copy, why is a single HDD performing better than an SSD?

    That's weird like you said. But I've also did the test without the disks part. Here is the results:

    Dacentec:

    Benchmark Run: Thu Jun 11 2015 16:16:17 - 16:40:28
    16 CPUs in system; running 1 parallel copy of tests
    
    Dhrystone 2 using register variables       23148727.4 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                     2868.9 MWIPS (10.0 s, 7 samples)
    Execl Throughput                               3343.7 lps   (29.9 s, 2 samples)
    Pipe Throughput                             1221676.6 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                 160265.0 lps   (10.0 s, 7 samples)
    Process Creation                               8002.5 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                   5656.5 lpm   (60.0 s, 2 samples)
    Shell Scripts (16 concurrent)                  2052.5 lpm   (60.0 s, 2 samples)
    Shell Scripts (8 concurrent)                   3203.5 lpm   (60.0 s, 2 samples)
    System Call Overhead                        1576640.7 lps   (10.0 s, 7 samples)
    
    System Benchmarks Partial Index              BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0   23148727.4   1983.6
    Double-Precision Whetstone                       55.0       2868.9    521.6
    Execl Throughput                                 43.0       3343.7    777.6
    Pipe Throughput                               12440.0    1221676.6    982.1
    Pipe-based Context Switching                   4000.0     160265.0    400.7
    Process Creation                                126.0       8002.5    635.1
    Shell Scripts (1 concurrent)                     42.4       5656.5   1334.1
    Shell Scripts (16 concurrent)                     ---       2052.5      ---
    Shell Scripts (8 concurrent)                      6.0       3203.5   5339.1
    System Call Overhead                          15000.0    1576640.7   1051.1
                                                                       ========
    System Benchmarks Index Score (Partial Only)                         1046.5
    
    ------------------------------------------------------------------------
    Benchmark Run: Thu Jun 11 2015 16:40:28 - 17:04:19
    16 CPUs in system; running 16 parallel copies of tests
    
    Dhrystone 2 using register variables      204843052.2 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                    37502.7 MWIPS (9.9 s, 7 samples)
    Execl Throughput                              35117.2 lps   (29.9 s, 2 samples)
    Pipe Throughput                            10405386.4 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                1731745.1 lps   (10.0 s, 7 samples)
    Process Creation                              70940.3 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                  43878.0 lpm   (60.0 s, 2 samples)
    Shell Scripts (16 concurrent)                  2993.3 lpm   (60.1 s, 2 samples)
    Shell Scripts (8 concurrent)                   5980.8 lpm   (60.1 s, 2 samples)
    System Call Overhead                        6041094.5 lps   (10.0 s, 7 samples)
    
    System Benchmarks Partial Index              BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0  204843052.2  17553.0
    Double-Precision Whetstone                       55.0      37502.7   6818.7
    Execl Throughput                                 43.0      35117.2   8166.8
    Pipe Throughput                               12440.0   10405386.4   8364.5
    Pipe-based Context Switching                   4000.0    1731745.1   4329.4
    Process Creation                                126.0      70940.3   5630.2
    Shell Scripts (1 concurrent)                     42.4      43878.0  10348.6
    Shell Scripts (16 concurrent)                     ---       2993.3      ---
    Shell Scripts (8 concurrent)                      6.0       5980.8   9968.0
    System Call Overhead                          15000.0    6041094.5   4027.4
                                                                       ========
    System Benchmarks Index Score (Partial Only)                         7581.9
    

    Delimiter:

    ------------------------------------------------------------------------
    Benchmark Run: Tue Jun 09 2015 18:40:32 - 19:04:52
    16 CPUs in system; running 1 parallel copy of tests
    
    Dhrystone 2 using register variables       26882773.0 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                     3177.9 MWIPS (10.0 s, 7 samples)
    Execl Throughput                               2600.7 lps   (29.7 s, 2 samples)
    Pipe Throughput                             1941070.8 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                 163969.0 lps   (10.0 s, 7 samples)
    Process Creation                               4184.4 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                   7744.1 lpm   (60.0 s, 2 samples)
    Shell Scripts (16 concurrent)                  2900.6 lpm   (60.0 s, 2 samples)
    Shell Scripts (8 concurrent)                   4502.2 lpm   (60.0 s, 2 samples)
    System Call Overhead                        3452120.4 lps   (10.0 s, 7 samples)
    
    System Benchmarks Partial Index              BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0   26882773.0   2303.6
    Double-Precision Whetstone                       55.0       3177.9    577.8
    Execl Throughput                                 43.0       2600.7    604.8
    Pipe Throughput                               12440.0    1941070.8   1560.3
    Pipe-based Context Switching                   4000.0     163969.0    409.9
    Process Creation                                126.0       4184.4    332.1
    Shell Scripts (1 concurrent)                     42.4       7744.1   1826.4
    Shell Scripts (16 concurrent)                     ---       2900.6      ---
    Shell Scripts (8 concurrent)                      6.0       4502.2   7503.7
    System Call Overhead                          15000.0    3452120.4   2301.4
                                                                       ========
    System Benchmarks Index Score (Partial Only)                         1205.9
    
    ------------------------------------------------------------------------
    Benchmark Run: Tue Jun 09 2015 19:04:52 - 19:29:18
    16 CPUs in system; running 16 parallel copies of tests
    
    Dhrystone 2 using register variables      207694210.7 lps   (10.0 s, 7 samples)
    Double-Precision Whetstone                    41037.6 MWIPS (10.0 s, 7 samples)
    Execl Throughput                              34237.7 lps   (29.8 s, 2 samples)
    Pipe Throughput                            17203969.6 lps   (10.0 s, 7 samples)
    Pipe-based Context Switching                2762998.0 lps   (10.0 s, 7 samples)
    Process Creation                              89458.9 lps   (30.0 s, 2 samples)
    Shell Scripts (1 concurrent)                  70540.3 lpm   (60.0 s, 2 samples)
    Shell Scripts (16 concurrent)                  4745.8 lpm   (60.1 s, 2 samples)
    Shell Scripts (8 concurrent)                   9460.7 lpm   (60.0 s, 2 samples)
    System Call Overhead                        6862767.8 lps   (10.0 s, 7 samples)
    
    System Benchmarks Partial Index              BASELINE       RESULT    INDEX
    Dhrystone 2 using register variables         116700.0  207694210.7  17797.3
    Double-Precision Whetstone                       55.0      41037.6   7461.4
    Execl Throughput                                 43.0      34237.7   7962.3
    Pipe Throughput                               12440.0   17203969.6  13829.6
    Pipe-based Context Switching                   4000.0    2762998.0   6907.5
    Process Creation                                126.0      89458.9   7099.9
    Shell Scripts (1 concurrent)                     42.4      70540.3  16636.9
    Shell Scripts (16 concurrent)                     ---       4745.8      ---
    Shell Scripts (8 concurrent)                      6.0       9460.7  15767.8
    System Call Overhead                          15000.0    6862767.8   4575.2
                                                                       ========
    System Benchmarks Index Score (Partial Only)                         9834.7
    

    2252 pts of difference :/

  • @FredQC - if you look at my comparison table its clear that its not the disks dragging it down. There is an actual performance issue in compute.

    Normalise the OS and kernel version (loose OpenVZ) and then recompare the results.

    Which L5520 do you have from Delimiter, is it the BL2x220C blade?

  • toirdaprrtoirdaprr Member
    edited June 2015

    edit

  • FredQcFredQc Member

    @MarkTurner said:
    FredQC - if you look at my comparison table its clear that its not the disks dragging it down. There is an actual performance issue in compute.

    Normalise the OS and kernel version (loose OpenVZ) and then recompare the results.

    Which L5520 do you have from Delimiter, is it the BL2x220C blade?

    Yes, that's the BL2x220C! I've installed the OpenVZ Kernel on the Delimiter box and will run the tests again...

  • dacentec.com is down

  • Their DNS is also down. My server can't resolve anything but I can SSH to it and use other setup ports :/

    Thanked by 1stallion
  • hotsnowhotsnow Veteran

    it seems also have some route issue, can't reach some network from dacentec

  • SadySady Member

    Do you guys check your email inbox? :p.

    Maintenance Window
    
    Start: Thursday, June 11, 2015, 03:00 PM EDT
    End:   Thursday, June 11, 2015, 03:15 PM EDT
    
    Services Affected: internet/Network
    
    Data Center Location(s): 801 Main Street NW, Lenoir, NC
    
    Expected Impact: During the maintenance window customers may experience a brief period of instability as routes fail over to alternate providers and re-converge throughout the internet.
    
    Details: Due to recent DDoS attacks we are adding a line card to one of our core routers to increase capacity. During the maintenance window customers may experience a brief period of instability as routes fail over to alternate providers and re-converge throughout the internet.
    
    Dacentec’s support team will be available for assistance 24/7. For questions, concerns, or to report a problem, please contact our Operations Center:

    .
    Might be possible that maintenance took a bit longer?

  • @hotsnow said:
    it seems also have some route issue, can't reach some network from dacentec.

    I'm not to sure what is causing this issue but it s actually pretty bad. Looks liek their Servers are down under DDoS or some serious crap going down. Could probably switch the DNS and that would probably fix the issue if they are running their own DNS systems. BUt hey speculation for now. Will probably get something official when it comes back up.

  • dacentecdacentec Member, Host Rep

    Charter lost a 100 gig bundle to Atlanta, it is still down. There was some routing instability failing over to Charlotte, but it seems all the issues have been resolved. Contact [email protected] if you are still having any issues.

    Thanked by 2hotsnow FrankZ
  • Are 5420s out of stock. When will they be available again?

  • dacentecdacentec Member, Host Rep

    They are back today.

  • hawchawc Moderator, LIR
    edited June 2015

    Anyone else on AS5089 (Virgin Media Limited) noticed different routing between the two? Noticing a lot more use of Level3 which is cool! However, VM peer directly with Level3, so on the return path, there is no need to do Dacentec -> Level3 -> Cable and Wireless -> Virgin... On the way to Dacentec it goes Virgin -> Level3 -> Dacentec

    Strange

    Anyway, nice to see the network changing

    Thanks
    Harry

  • wychwych Member

    hawc said: Anyone else on AS5089 (Virgin Media Limited) noticed different routing between the two?

    I noticed some the other night, nothing major though.

  • Would have thought VM sent traffic over Tinet.

  • howardsl2howardsl2 Member
    edited June 2015

    Ordered a server and Dacentec staff have been very helpful with quick ticket responses (< 30 min) and (up to 24 hours) free IP KVM access upon request. The server was provisioned fast within a few hours. They even let me swap it for another model with better specs and refunded the original one. Would recommend to friends.

    A few helpful hints for all:

    1. Avoid the use of "poweroff" or "shutdown -h now" unless the BIOS setting of "After Power Failure" is set to "Power On". Otherwise some data center staff will have to press the physical power button on your server. (To enter the BIOS, first request IP KVM access via ticket, then reboot your server and hit the "DEL" key.)

    2. In order for Dacentec's auto provisioning system to work, the NVIDIA boot agent (PXE) must be set before the hard disk in your server's "boot priority" BIOS settings.

    3. Per Dacentec's earlier comment, you can enter the non-destructive "Rescue Mode" from the auto provision system, by selecting one of the "Sysrcd" options. After reboot, connect to the VNC server on port 5901, or to the SSH server with your chosen password.

    4. If the "Virtual Media" in IP KVM does not work or the BIOS "boot priority" does not show the virtual disk, try toggling USB and PS/2 modes under "Keyboard/Mouse Settings", or try the "Reset" under "Maintenance".

  • Could someone paste a (necessary redacted) traceroute from a delimiter atlanta IP to mirror.dacentec.com ?

  • @vimalware said:
    Could someone paste a (necessary redacted) traceroute from a delimiter atlanta IP to mirror.dacentec.com ?

    Sending it in PM, I suck something chronic at Markdown.

  • @Umcookies said:
    Sending it in PM, I suck something chronic at Markdown.

    Just use < pre > and < /pre > tags (no spaces), put your info between them.

  • SadySady Member

    They've some awesome new offers available at the moment. Have just grabbed L5420 16GB 2x1TB RTO $30/month :)

  • Sady said: They've some awesome new offers available at the moment. Have just grabbed L5420 16GB 2x1TB RTO $30/month :)

    Thanks for the heads up. I picked up the Dell 2xL5420 16GB 2X2TB Rent Only, as I'm not sure whether colocation costs would make it worthwhile to own one.

  • SadySady Member

    @Droidzone said:
    Thanks for the heads up. I picked up the Dell 2xL5420 16GB 2X2TB Rent Only, as I'm not sure whether colocation costs would make it worthwhile to own one.

    I've that one as well, truely not sure what I'll do with these :3

  • @vimalware said:
    Could someone paste a (necessary redacted) traceroute from a delimiter atlanta IP to mirror.dacentec.com ?

    root@atl:~# ping mirror.dacentec.com
    PING mirror.dacentec.com (199.191.56.90) 56(84) bytes of data.
    64 bytes from mirror.dacentec.com (199.191.56.90): icmp_req=1 ttl=57 time=9.75 ms
    64 bytes from mirror.dacentec.com (199.191.56.90): icmp_req=2 ttl=57 time=9.58 ms
    64 bytes from mirror.dacentec.com (199.191.56.90): icmp_req=3 ttl=57 time=9.76 ms
    64 bytes from mirror.dacentec.com (199.191.56.90): icmp_req=4 ttl=57 time=9.55 ms
    64 bytes from mirror.dacentec.com (199.191.56.90): icmp_req=5 ttl=57 time=9.73 ms
    ^C
    --- mirror.dacentec.com ping statistics ---
    5 packets transmitted, 5 received, 0% packet loss, time 4006ms
    rtt min/avg/max/mdev = 9.552/9.678/9.764/0.108 ms
    
    root@atl:~# mtr mirror.dacentec.com --report
    HOST: atl                         Loss%   Snt   Last   Avg  Best  Wrst StDev
      1.|-- 107.189.xxx.xxx             0.0%    10    0.4   2.2   0.3  17.8   5.5
      2.|-- te1-1-0d0.cir1.atlanta6-g  0.0%    10    0.4   1.1   0.3   5.1   1.5
      3.|-- 216.156.9.6.ptr.us.xo.net  0.0%    10   30.1   3.9   0.4  30.1   9.3
      4.|-- ???                       100.0    10    0.0   0.0   0.0   0.0   0.0
      5.|-- CENTRILOGIC.ear1.Atlanta2  0.0%    10    0.9   0.9   0.9   0.9   0.0
      6.|-- dct-cr01--v51.dacentec.co  0.0%    10   10.1  11.9   9.6  16.2   2.4
      7.|-- mirror.dacentec.com        0.0%    10    9.7   9.6   9.5   9.8   0.1
  • tomletomle Member, LIR

    $20 are still available!

  • Do the DELL or HP models come with native KVM/iLo card support? @dacentec

    Or is it on-demand Lantronix for every model/manufacturer?

  • dacentecdacentec Member, Host Rep

    Some Dell L5520, HP DL 180, E3 and others have full onboard iKVM. We only support the lantronix loaner and our provisioning system. But we know the some customers are using the onboard management.

  • sinsin Member

    Dacentec's support is awesome, Chris H. helped me out a lot yesterday morning and was extremely fast with this replies/help too. I just renewed my server a few days ago so I have had it for a little over a month and it's just an amazing deal.

    Thanked by 1dacentec
Sign In or Register to comment.