Howdy, Stranger!

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

Sign In Register

Quick Links


Shells Virtual Desktop
BMail.ag - Secure Email Service
Server.net
CPLicense.net
VPS Server
Buy VPN
Vultr
VMs for AI
HostDare
ReliableSite White-Label Dedicated Hosting for Resellers
InterServer VPS
BMail.ag - Secure Email Service
Best VPN
High-Performance Bare Metal Server Solutions
Karvl.com
Server Mania Cloud Hosting
DataWagon Hosting
AlphaVPS Hosting
Evoxt.com
Clouvider
VPS Hosting with NVMe
Residential IPs in the US & 4G Mobile Proxies in EU & US with Unlimited Bandwidth
ReliableSite White-Label Dedicated Hosting for Resellers
Rabisu - Hosting Solutions
Shells Virtual Desktop

Categories

In this Discussion

Home › Offers
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.

HostBrr | Two year anniversary deals | AMD EPYC Turin + Block Storage | Flash deals !

1…3435373940…59

Comments

  • AstroAstro Member

    Please @labze. Just a little bit.

    Thanked by 2cainyxues Flq
  • NaughtyWombatNaughtyWombat Member

    I have been watching this thread for about a week and haven't gotten anything yet.

    Thanked by 2_MS_ cainyxues
  • _MS__MS_ Member
    edited April 2025

    Kære Brrsen/@labze. Genfyld venligst forhandlerens hosting flash-udsalg. Eller bare PM mig en kode. Tak.

    Thanked by 1cainyxues
  • cainyxuescainyxues Member

    @MS said: Eller bare PM mig en kode. Tak.

    me, yes :+1:

    Thanked by 1_MS_
  • _MS__MS_ Member

    @cainyxues said:

    @MS said: Eller bare PM mig en kode. Tak.

    me, yes :+1:

    Tak. :+1:

    Thanked by 1cainyxues
  • AstroAstro Member

    What are we PMing!!!

    Thanked by 1cainyxues
  • cainyxuescainyxues Member

    @Astro said:
    What are we PMing!!!

    you pm'ing me a deal 🤝 :wink: :wink: [he was saying

    @_MS_

    to @labze to pm him a code in his dm]

    Thanked by 1_MS_
  • AstroAstro Member

    Gimme 50%. I don’t even want 60%

    Thanked by 1cainyxues
  • PineappleMPineappleM Member

    My hunch is the €7/yr deal will drop on the last day of the flash deals, whenever that will be.

    @labze will it be on a Turin CPU? If so, expect your website to crash when it drops.

  • hyperblasthyperblast Member

    @labze is there any news regarding windows server on the EPIC Anniversary version?

  • sh97sh97 Member, Host Rep

    Thanked by 11PineappleM barbaros vastness4594 NaughtyWombat cainyxues Blembim Hitori0221 NHNHNH000 Decicus _MS_ lukast__
  • cainyxuescainyxues Member

    Dealz Wen? Where?

    Thanked by 3ariq01 _MS_ Blembim
  • NHNHNH000NHNHNH000 Member
    edited April 2025

    why talking like you guys can get his flash dealz easily

    Thanked by 1cainyxues
  • maverickmaverick Member
    edited April 2025

    @darkimmortal said:
    Did some benchmarking on the block storage. It is cached so the YABS fio results are not representative

    fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vdb1):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 246.83 MB/s  (61.7k) | 2.17 GB/s    (33.9k)
Write      | 247.48 MB/s  (61.8k) | 2.18 GB/s    (34.1k)
Total      | 494.31 MB/s (123.5k) | 4.35 GB/s    (68.0k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 2.62 GB/s     (5.1k) | 3.76 GB/s     (3.6k)
Write      | 2.76 GB/s     (5.4k) | 4.01 GB/s     (3.9k)
Total      | 5.39 GB/s    (10.5k) | 7.78 GB/s     (7.5k)

    The real performance is visible in a longer test, e.g.


    # fio --size=30000m --filename=fiotest.tmp --ioengine=libaio --direct=1 --name=hi --bs=1M --rw=write --iodepth=32
hi: (g=0): rw=write, bs=(R) 1024KiB-1024KiB, (W) 1024KiB-1024KiB, (T) 1024KiB-1024KiB, ioengine=libaio, iodepth=32
fio-3.33
Starting 1 process
hi: Laying out IO file (1 file / 30000MiB)
Jobs: 1 (f=1): [W(1)][100.0%][w=64.0MiB/s][w=64 IOPS][eta 00m:00s] 
hi: (groupid=0, jobs=1): err= 0: pid=4286: Fri Apr 18 23:09:02 2025
  write: IOPS=119, BW=120MiB/s (126MB/s)(29.3GiB/250557msec); 0 zone resets
    slat (usec): min=7, max=172493k, avg=5897.74, stdev=996062.52
    clat (usec): min=285, max=172508k, avg=261351.84, stdev=5592367.60
     lat (usec): min=320, max=172508k, avg=267249.58, stdev=5680118.12
    clat percentiles (usec):
     |  1.00th=[     873],  5.00th=[    1369], 10.00th=[    1991],
     | 20.00th=[    4047], 30.00th=[    7046], 40.00th=[   10421],
     | 50.00th=[   12911], 60.00th=[   16188], 70.00th=[   20055],
     | 80.00th=[   25822], 90.00th=[   36963], 95.00th=[   89654],
     | 99.00th=[ 1719665], 99.50th=[ 2634023], 99.90th=[17112761],
     | 99.95th=[17112761], 99.99th=[17112761]
   bw (  KiB/s): min= 2052, max=3410597, per=100.00%, avg=705266.30, stdev=979846.93, samples=87
   iops        : min=    2, max= 3330, avg=688.69, stdev=956.86, samples=87
  lat (usec)   : 500=0.12%, 750=0.41%, 1000=1.21%
  lat (msec)   : 2=8.31%, 4=9.71%, 10=18.92%, 20=30.99%, 50=23.16%
  lat (msec)   : 100=2.39%, 250=0.83%, 500=0.89%, 750=0.72%, 1000=0.42%
  lat (msec)   : 2000=1.06%, >=2000=0.85%
  cpu          : usr=0.36%, sys=0.14%, ctx=14696, majf=0, minf=13
  IO depths    : 1=0.1%, 2=0.1%, 4=0.1%, 8=0.1%, 16=0.1%, 32=99.9%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.1%, 64=0.0%, >=64=0.0%
     issued rwts: total=0,30000,0,0 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: bw=120MiB/s (126MB/s), 120MiB/s-120MiB/s (126MB/s-126MB/s), io=29.3GiB (31.5GB), run=250557-250557msec

Disk stats (read/write):
  vdb: ios=0/30935, merge=0/4, ticks=0/2509060, in_queue=2720648, util=99.45%

    tl;dr write speed around 100MB/s, latency odd (note the 99.9%ile latency of 17 seconds!)

    With fsync'd writes the performance is worse, around 1 IOPS!

    # fio --size=3000m --filename=fiotest2.tmp --ioengine=libaio --direct=1 --name=hi --bs=1M --rw=randwrite --iodepth=32 --fsync=1
hi: (g=0): rw=randwrite, bs=(R) 1024KiB-1024KiB, (W) 1024KiB-1024KiB, (T) 1024KiB-1024KiB, ioengine=libaio, iodepth=32
fio-3.33
Starting 1 process
^Cbs: 1 (f=1): [w(1)][2.2%][w=1022KiB/s][w=0 IOPS][eta 39m:08s]    
fio: terminating on signal 2

hi: (groupid=0, jobs=1): err= 0: pid=4368: Fri Apr 18 23:15:10 2025
  write: IOPS=1, BW=1906KiB/s (1952kB/s)(96.0MiB/51567msec); 0 zone resets
    slat (usec): min=17, max=196, avg=64.39, stdev=35.17
    clat (msec): min=965, max=24679, avg=16349.89, stdev=6830.34
     lat (msec): min=965, max=24679, avg=16349.96, stdev=6830.35
    clat percentiles (msec):
     |  1.00th=[  969],  5.00th=[ 3037], 10.00th=[ 5604], 20.00th=[ 9463],
     | 30.00th=[11610], 40.00th=[17113], 50.00th=[17113], 60.00th=[17113],
     | 70.00th=[17113], 80.00th=[17113], 90.00th=[17113], 95.00th=[17113],
     | 99.00th=[17113], 99.50th=[17113], 99.90th=[17113], 99.95th=[17113],
     | 99.99th=[17113]
   bw (  KiB/s): min= 2048, max= 4096, per=100.00%, avg=2511.70, stdev=865.31, samples=53
   iops        : min=    2, max=    4, avg= 2.45, stdev= 0.85, samples=53
  lat (msec)   : 1000=1.04%, 2000=1.04%, >=2000=97.92%
  fsync/fdatasync/sync_file_range:
    sync (msec): min=3903, max=5613.2k, avg=103173.50, stdev=688707.50
    sync percentiles (msec):
     |  1.00th=[ 3910],  5.00th=[ 5604], 10.00th=[ 8221], 20.00th=[10939],
     | 30.00th=[16174], 40.00th=[17113], 50.00th=[17113], 60.00th=[17113],
     | 70.00th=[17113], 80.00th=[17113], 90.00th=[17113], 95.00th=[17113],
     | 99.00th=[17113], 99.50th=[17113], 99.90th=[17113], 99.95th=[17113],
     | 99.99th=[17113]
  cpu          : usr=0.01%, sys=0.02%, ctx=183, majf=0, minf=9
  IO depths    : 1=1.0%, 2=3.1%, 4=4.2%, 8=8.3%, 16=16.7%, 32=135.4%, >=64=0.0%
     submit    : 0=0.0%, 4=100.0%, 8=0.0%, 16=0.0%, 32=0.0%, 64=0.0%, >=64=0.0%
     complete  : 0=0.0%, 4=98.5%, 8=0.0%, 16=0.0%, 32=1.5%, 64=0.0%, >=64=0.0%
     issued rwts: total=0,96,0,66 short=0,0,0,0 dropped=0,0,0,0
     latency   : target=0, window=0, percentile=100.00%, depth=32

Run status group 0 (all jobs):
  WRITE: bw=1906KiB/s (1952kB/s), 1906KiB/s-1906KiB/s (1952kB/s-1952kB/s), io=96.0MiB (101MB), run=51567-51567msec

Disk stats (read/write):
  vdb: ios=0/164, merge=0/0, ticks=0/50978, in_queue=102072, util=97.95%

    The VPS performance is absolutely top class, but I'm afraid I can't recommend the block storage - you will probably be disappointed coming from a single dedicated HDD or storage VPS offerings at other providers

    Other observations: read performance is hard to measure fairly with the caching setup, seems to be somewhere in the region of equivalent to a single dedicated HDD (200MB/s, 300 IOPS). Thin provisioning is being used judging by extremely high read speed on uninitialised/discarded areas.

    Did only fsync test, exactly as you did it, to be able to compare numbers.

    HOST-C storage VPS runs at full 127 MiB/s despite fsync (the storage is limited to that speed). Knowing @host_c who swears by HW RAID controllers :wink: he probably also has a good chunk of onboard RAM cache which is battery protected, and in that case fsync is basically a noop, so the good speed is the result of proper enterprise technology involved.

    On another storage VPS, but with rather complex setup (iSCSI through private networking) I got around 35 MiB/s. I consider it good enough (for the purpose).

    Finally, on a physical 4 x 8 TB HDD SW RAID0 array, the result is 42 MiB/s.

    Considering all that, ~ 2 MiB/s you got is quite dissapointing, I agree.

    Thanked by 6wadhah NHNHNH000 PineappleM darkimmortal lukast__ host_c
  • PineappleMPineappleM Member

    Does anyone who have the Flash-1 server be able to benchmark the performance of its HDD? Is it similar?

  • labzelabze Member, Patron Provider

    @hyperblast said:
    @labze is there any news regarding windows server on the EPIC Anniversary version?

    Latest benchmark gave me 2684 on Windows 2022, so about to be where it should be expected.

    Seems like the power profile on the host system on Turin CPUs is much more sensitive and had to be set more aggressive.

    At least this is from my initial findings.

    Thanked by 3cainyxues PineappleM hyperblast
  • PineappleMPineappleM Member

    @labze said:

    @hyperblast said:
    @labze is there any news regarding windows server on the EPIC Anniversary version?

    Latest benchmark gave me 2684 on Windows 2022, so about to be where it should be expected.

    Seems like the power profile on the host system on Turin CPUs is much more sensitive and had to be set more aggressive.

    At least this is from my initial findings.

    Do you think that would boost everyone’s GB6 scores on the node?

  • labzelabze Member, Patron Provider

    @PineappleM said:

    @labze said:

    @hyperblast said:
    @labze is there any news regarding windows server on the EPIC Anniversary version?

    Latest benchmark gave me 2684 on Windows 2022, so about to be where it should be expected.

    Seems like the power profile on the host system on Turin CPUs is much more sensitive and had to be set more aggressive.

    At least this is from my initial findings.

    Do you think that would boost everyone’s GB6 scores on the node?

    No, Windows seems to be primarily affected by this. Linux systems to a much lesser degree. I did see 50-100 higher in GB5/6 but I'd say it is within the margin of error.

    Thanked by 1PineappleM
  • hyperblasthyperblast Member

    @labze said:

    @hyperblast said:
    @labze is there any news regarding windows server on the EPIC Anniversary version?

    Latest benchmark gave me 2684 on Windows 2022, so about to be where it should be expected.

    Seems like the power profile on the host system on Turin CPUs is much more sensitive and had to be set more aggressive.

    At least this is from my initial findings.

    does this mean that you have tuned something else on the host system? i would then try again with an epic-epyc in conjunction with windows server 2022.

  • hyperblasthyperblast Member

    gb6 2500+ with windows server on a potent turin node are ok!

    Thanked by 1COLBYLICIOUS
  • emperoremperor Member
    edited April 2025

    @labze is today the day for 7€/y VPS deal ?

    Thanked by 1sh97
  • labzelabze Member, Patron Provider

    @emperor said:
    @labze is today the day for 7€/y VPS deal ?

    Nah, it'll be the last deal.

    Thanked by 6emperor anakara ikibsys PineappleM Flq lukast__
  • emperoremperor Member

    @labze said:

    @mandala said:

    @labze said:
    All deals are intended to be made available twice, spaced out, so in maybe 8-16 hours for next batch.

    • unpaid orders will be canceled and made available for new orders.

    Since it's 2nd anniversary this flashing madness is lasting 2 days weeks right?

    The 2nd anniversary flashes will continue for 2 years. I ain't stopping until the CPUs reach the same Geekbench scores as some managed to do earlier.

    That would be a lot of time for 7€/y deal :D :D

    Thanked by 1PineappleM
  • NHNHNH000NHNHNH000 Member

    @labze no more restock for flash-1? :/

    Thanked by 1zrj766
  • fmgpfmgp Member

    @labze no more restock for flash-1? :/

  • PineappleMPineappleM Member

    @labze said:

    @emperor said:
    @labze is today the day for 7€/y VPS deal ?

    Nah, it'll be the last deal.

    Aha, I called it! But then it begs the question: how many days of deals are you rolling out? It’s now half a week.

  • FlqFlq Member

    @labze said:

    @emperor said:
    @labze is today the day for 7€/y VPS deal ?

    Nah, it'll be the last deal.

    So more deals ready to come...!?
    Can't wait!

  • codelockcodelock Member

    @labze drop some promo code please

    Thanked by 2sh97 cainyxues
  • itachikonohaitachikonoha Member 
    # ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
#              Yet-Another-Bench-Script              #
#                     v2025-04-20                    #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #

Wed Apr 23 21:43:54 IST 2025

Basic System Information:
---------------------------------
Uptime     : 0 days, 0 hours, 3 minutes
Processor  : AMD EPYC 9655 96-Core Processor
CPU cores  : 2 @ 2600.002 MHz
AES-NI     : v Enabled
VM-x/AMD-V : v Enabled
RAM        : 5.7 GiB
Swap       : 0.0 KiB
Disk       : 58.0 GiB
Distro     : Ubuntu 24.04.2 LTS
Kernel     : 6.8.0-57-generic
VM Type    : KVM
IPv4/IPv6  : v Online / ❌ Offline

IPv4 Network Information:
---------------------------------
ISP        : IP Interactive
ASN        : AS214902 Philip Fjaera trading as PFWeb Solutions
Host       : rootbash.com - IT services
Location   : Berlin, State of Berlin (BE)
Country    : Germany

Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value
                |
Single Core     | 2768
Multi Core      | 5042
Full Test       | https://browser.geekbench.com/v6/cpu/11652615

YABS completed in 5 min 16 sec

    @labze has made the competition even more challenging.

    Now I've 3 host which I can rely on for high computation tasks.

    1. Berohost (GB 6: 2900-3000)
    2. Hostbrr (GB 6: 2700-2800)
    3. Computebox (GB 6: 2400-2500)
    Thanked by 2cainyxues PineappleM
  • ahdahd Member
    edited April 2025

    @labze Looking glass or test IP available for Turin instance? Thanks

Sign In or Register to comment.