Howdy, Stranger!

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


Time4vps 50% off for VPS, storage hosting and VPN. Today only 100 coupons!
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.

Time4vps 50% off for VPS, storage hosting and VPN. Today only 100 coupons!

EvpaEvpa Member
edited September 2018 in General

Today is Programmer's Day and european provider time4vps.eu gives you 50% OFF VPS, storage hosting and VPN services! The discount is valid only for today!

Use promo code: HELLOWORLD

Use this promo code and get 50% OFF Time4VPS Standard VPS, Storage hosting and VPN service. The discount is not recurring and can be applied once. Only 100 coupons are provided and can be used only at 13th of September (GMT +3).

My refferal link for kind and generous sysadmins who pay their invoices in full and on time :) https://www.time4vps.eu/?affid=3263

Link for others: https://www.time4vps.eu/

Thanked by 1default

Comments

  • @Chuck Seen. :smile:

  • Quick money with non recurring discounts.

  • mfsmfs Banned, Member

    non recurring

    Thanked by 2corbpie leonari
  • not for KVM :(

  • HukinHukin Member
    edited September 2018

    Tripleflix said: not for KVM

    They still keep it for a sweet.

  • DedicadudeDedicadude Member
    edited September 2018

    I used time4vps a few years back. The I/O write speed was terrible not sure about now. Still a good offer for a 1 month (or 1 year) project... Non recurring? no problem, no contract...

    1GB RAM (non kvm) for 38.94 EUR per 1st year is it worth it guys? I'm tempted to get one...

    I don't mean to hijack this thread by I saw another offer on LEB... https://www.sparkvps.com/crm/cart.php?a=confproduct&i=0 $39/yr for 6GB RAM + 100 GB SSD

  • @Dedicadude said:
    I used time4vps a few years back. The I/O write speed was terrible not sure about now.

    They set disk I/O limit to 100MB/s.

  • i'm running a home server and have 6x Samsung 850 Evo 250GB SSDs (xfs perc h700 raid controller)...

    output of a write test:

    [root@localhost] sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 1.95942 s, 548 MB/s

    output of a read test:

    [root@localhost]# dd if=tempfile of=/dev/null bs=1M count=1024
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 0.171463 s, 6.3 GB/s

    but 100MB/s? come on now...

  • Ah, nice, grabbed one 1T! Thanks for the information!

  • @Dedicadude said:
    i'm running a home server and have 6x Samsung 850 Evo 250GB SSDs (xfs perc h700 raid controller)...

    output of a write test:

    [root@localhost] sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 1.95942 s, 548 MB/s

    output of a read test:

    [root@localhost]# dd if=tempfile of=/dev/null bs=1M count=1024
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 0.171463 s, 6.3 GB/s

    but 100MB/s? come on now...

    Noisy neighbours are not welcome? :)

  • AnthonySmithAnthonySmith Member, Patron Provider

    Dedicadude said: 6x Samsung 850 Evo 250GB SSDs

    storage server, ramp that up to 20TB and lets see how you feel about spinning drive speeds for remote storage :)

  • Evpa said: Noisy neighbours are not welcome?

    Let us join and ruin it. :smiley:

  • @Dedicadude said:
    i'm running a home server and have 6x Samsung 850 Evo 250GB SSDs (xfs perc h700 raid controller)...

    output of a write test:

    [root@localhost] sync; dd if=/dev/zero of=tempfile bs=1M count=1024; sync
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 1.95942 s, 548 MB/s

    output of a read test:

    [root@localhost]# dd if=tempfile of=/dev/null bs=1M count=1024
    1024+0 records in
    1024+0 records out
    1073741824 bytes (1.1 GB) copied, 0.171463 s, 6.3 GB/s

    but 100MB/s? come on now...

    6.3GB/s does not compute. You have 6 drives in raid0 at roughly half a gig/sec each. Do the math. This might be one of those cases where people complain about using dd like this incorrectly because the results are wildly wrong.

    Copy 50GB image and check speed.

  • DedicadudeDedicadude Member
    edited September 2018

    @TimboJones said:
    6.3GB/s does not compute. You have 6 drives in raid0 at roughly half a gig/sec each. Do the math. This might be one of those cases where people complain about using dd like this incorrectly because the results are wildly wrong.

    Copy 50GB image and check speed.

    You're right. That 6.3GB/s is probably being done from cache. lol. dd isn't an appropriate way to measure I/O performance. I tried other benchmark applications like iozone, bonnie++, etc. and tweak them to my desired read/write patterns.

    I'm playing with nr_requests and read_ahead_kb. I've also found that with iozone instead of dd I get quite different numbers and the differences between the servers seem to vanish.

    Edit: Thanks for the heads up!

  • jlayjlay Member
    edited September 2018

    @Dedicadude said:

    @TimboJones said:
    6.3GB/s does not compute. You have 6 drives in raid0 at roughly half a gig/sec each. Do the math. This might be one of those cases where people complain about using dd like this incorrectly because the results are wildly wrong.

    Copy 50GB image and check speed.

    You're right. That 6.3GB/s is probably being done from cache. lol. dd isn't an appropriate way to measure I/O performance. I tried other benchmark applications like iozone, bonnie++, etc. and tweak them to my desired read/write patterns.

    I'm playing with nr_requests and read_ahead_kb. I've also found that with iozone instead of dd I get quite different numbers and the differences between the servers seem to vanish.

    Edit: Thanks for the heads up!

    It's okay to use dd if you're only testing sequential speeds, the other tools you mention are better for overall profiling (eg: random IOPs, latency measurements at queue lengths, etc). The important thing with dd is to write enough that'll saturate the write cache so you don't get inflated numbers like this. Different block sizes and such (like the 'conv' option) have a lot of impact as well. It's still a perfectly decent test tool, it's just the variables are a little wilder.

  • Got one to try their VPN and its quite good so far

Sign In or Register to comment.