Howdy, Stranger!

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


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
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.

[Updated] Hosteroid multi-location benchmarks and review plus a nice fragrance in the air

2»

Comments

  • jsgjsg Member, Resident Benchmarker

    @Mumbly said:

    @jsg said: That, guys and gals, IMO is the weak point.
    While connectivity within Europa is acceptable, global connectivity simply is crappy in my books.
    That said, within Europe connectivity certainly isn't great but good enough for most jobs one reasonably buys and uses such a cheap VPS for.

    I noticed this as well. I live quite close to Vienna, but it seems like everything is routed through Frankfurt.
    So in my case, the latency is worse, or even significantly worse than the latency from several not so close German VPS providers I use.
    I don't have any other complaints. It's stable and everything. It's just the damn routing going halfway across Europe only to come back almost to the starting point...
    For example, I have more than twice as low latency to their Bratislava location (which is geographically farther from me) than to Vienna.

    To be honest I didn't look at routes in detail (mtr, traceroute, ...) but just (automatically) gathered the data and then compiled them. But I, with good reason, show not just the average values but the minima and maxima as well plus how many failures there were with each target.
    But anyway, yes, I wouldn't be surprised at all if the Vienna location was, let's say a bit special, which almost certainly wouldn't be the provider's but the backend's/carrier's[/DC's?] fault.

    But maybe I should look at routing as well to have those data available if needed. But to generally publish them would look like overkill to me (huge amount of data). Let me think a bit about it ...

    But @Hosteroid is a really nice guy and I guess he might (again) provide test access for a few days. If so, I'm willing to collect routing info to all targets from Vienna.

  • jsgjsg Member, Resident Benchmarker
    edited July 2025

    upgraded RO, Buc VPS benchmark and review

    They just upgraded a (or multiple or all?) node in their home location in Bucharest. And of bloody course I benchmarked my Hosteroid VPS there right away. And it was worth it, I saw quite significant improvments!

    Based on about 100 runs.

    Version 2.5.0a, (c) 2018+ jsg (->lowendtalk.com)
    Machine: amd64, Arch.: amd64, Model: Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz
    OS, version: FreeBSD 14.2, Mem.: 3.989 GB
    CPU - Cores: 2, Family/Model/Stepping: 6/85/7
    Cache: 32K/32K L1d/L1i, 1024K L2, 27M L3
    Std. Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat
              pse36 cflsh mmx fxsr sse sse2 ss sse3 pclmulqdq vmx ssse3 fma cx16
              pdcm pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline aes xsave
              osxsave avx f16c rdrnd hypervisor
    Ext. Flags: fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid mpx pat pse36
              rdseed adx smap clflushopt clwb sha umip pku ospke syscall nx pdpe1gb
              rdtscp lm lahf_lm lzcnt
    
    AES? Yes
    InNested Virt.? Yes
    HW RNG? Yes
    
    ProcMem SC [MB/s]: avg 225.9 - min 80.1 (35.5 %), max 414.8 (183.6 %)
    ProcMem MA [MB/s]: avg 658.9 - min 567.6 (86.1 %), max 816.6 (123.9 %)
    ProcMem MB [MB/s]: avg 658.2 - min 586.7 (89.1 %), max 780.3 (118.6 %)
    ProcMem AES [MB/s]: avg 1128.8 - min 1067.2 (94.5 %), max 1254.7 (111.2 %)
    ProcMem RSA [kp/s]: avg 95.7 - min 80.1 (83.7 %), max 111.0 (115.9 %)
    

    Wow! Very significantly better now! Almost or about double of what we saw with the former E5v4 processor, both single and multicore. And crypto performance is in solid territory as well now. Very nice, I'm impressed and very happy with what I see. And I even didn't have to pay, @Hosteroid just upgraded the node I'm on. Kudos, very well done!

    Let's see whether the disk performance also is higher now.

    --- Disk 4 KB - Buffered ---
    Write seq. [MB/s]: avg 4.54 - min 3.90 (85.8%), max 4.77 (105.0%)
    Write rnd. [MB/s]: avg 4.15 - min 3.47 (83.7%), max 4.40 (106.1%)
    Read seq. [MB/s]:  avg 10.25 - min 8.73 (85.1%), max 11.03 (107.6%)
    Read rnd. [MB/s]:  avg 8.77 - min 7.45 (85.0%), max 9.70 (110.7%)
    --- Disk 4 KB - Sync/Direct ---
    Write seq. [MB/s]: avg 4.45 - min 3.55 (79.9%), max 4.92 (110.7%)
    Write rnd. [MB/s]: avg 4.16 - min 3.42 (82.2%), max 4.65 (111.8%)
    Read seq. [MB/s]:  avg 10.23 - min 8.36 (81.7%), max 10.87 (106.2%)
    Read rnd. [MB/s]:  avg 8.79 - min 7.83 (89.0%), max 9.38 (106.7%)
    
    --- Disk 64 KB - Buffered ---
    Write seq. [MB/s]: avg 41.46 - min 34.63 (83.5%), max 45.11 (108.8%)
    Write rnd. [MB/s]: avg 40.52 - min 33.80 (83.4%), max 43.11 (106.4%)
    Read seq. [MB/s]:  avg 963.56 - min 823.74 (85.5%), max 1300.68 (135.0%)
    Read rnd. [MB/s]:  avg 117.44 - min 98.69 (84.0%), max 124.88 (106.3%)
    --- Disk 64 KB - Sync/Direct ---
    Write seq. [MB/s]: avg 8.67 - min 7.56 (87.2%), max 10.80 (124.6%)
    Write rnd. [MB/s]: avg 4.64 - min 4.17 (89.9%), max 5.98 (128.9%)
    Read seq. [MB/s]:  avg 955.39 - min 796.51 (83.4%), max 1305.34 (136.6%)
    Read rnd. [MB/s]:  avg 100.49 - min 89.52 (89.1%), max 116.02 (115.5%)
    
    --- Disk 1 MB - Buffered ---
    Write seq. [MB/s]: avg 90.66 - min 57.27 (63.2%), max 111.86 (123.4%)
    Write rnd. [MB/s]: avg 155.94 - min 98.52 (63.2%), max 192.91 (123.7%)
    Read seq. [MB/s]:  avg 1670.10 - min 1468.77 (87.9%), max 1930.36 (115.6%)
    Read rnd. [MB/s]:  avg 566.18 - min 513.29 (90.7%), max 620.41 (109.6%)
    --- Disk 1 MB - Sync/Direct ---
    Write seq. [MB/s]: avg 23.67 - min 21.63 (91.4%), max 27.13 (114.6%)
    Write rnd. [MB/s]: avg 18.12 - min 16.86 (93.0%), max 21.32 (117.6%)
    Read seq. [MB/s]:  avg 1684.26 - min 1465.79 (87.0%), max 2006.00 (119.1%)
    Read rnd. [MB/s]:  avg 613.57 - min 562.27 (91.6%), max 681.90 (111.1%)
    --- Disk IOps (Sync/Direct) ---
    Write seq. [MB/s]: avg 38.78 - min 35.75 (92.2%), max 46.93 (121.0%)
    IOps             : avg 9927.39 - min 9152.38 (92.2%), max 12014.06 (121.0%)
    

    Yep, it is, nice! An impressive improvement as well! Both, the 4k/4t results are about 5 x higher now!

    Re connectivity I'll only list targets that either changed significantly or are new/added, as indicated at the end of the first line of the listed targets.

    --- Europe ---
    
    NO OSL mirror.terrahost.no [F: 0] - was 165.8 MB/s
      DL [Mb/s]:      avg 217.4 - min 187.1 (86.1%), max 233.3 (107.3%)
      Ping [ms]:      avg 47.7 - min 47.0 (98.6%), max 51.4 (107.9%)
      Web ping [ms]:  avg 51.5 - min 47.0 (91.3%), max 87.1 (169.2%)
    
    IT MIL it1.mirror.vhosting-it.com [F: 28] - was 336.8 MB/s
      DL [Mb/s]:      avg 38.2 - min 0.0 (0.0%), max 232.0 (606.6%)
      Ping [ms]:      avg 27.7 - min 27.4 (99.1%), max 29.8 (107.7%)
      Web ping [ms]:  avg 30.4 - min 27.5 (90.4%), max 81.5 (268.1%)
    
    RO BUC almalinux.mirrors.orange.ro [F: 0] - was 1282.5 MB/s
      DL [Mb/s]:      avg 2112.4 - min 765.3 (36.2%), max 3486.9 (165.1%)
      Ping [ms]:      avg 1.2 - min 1.0 (82.4%), max 3.8 (313.3%)
      Web ping [ms]:  avg 5.4 - min 1.1 (20.4%), max 94.2 (1749.3%)
    
    RO BUC mirrors.hosterion.ro [F: 0] - was 2604.6 MB/s
      DL [Mb/s]:      avg 4493.1 - min 2870.7 (63.9%), max 6252.3 (139.2%)
      Ping [ms]:      avg 0.6 - min 0.4 (64.9%), max 1.6 (259.7%)
      Web ping [ms]:  avg 0.9 - min 0.5 (57.9%), max 5.5 (636.6%)
    
    --- Asia / Oceania ---
    
    CN HKG mirrors.xtom.hk [F: 0] - was 35.4 MB/s
      DL [Mb/s]:      avg 50.0 - min 48.5 (97.0%), max 51.1 (102.2%)
      Ping [ms]:      avg 216.5 - min 216.3 (99.9%), max 217.1 (100.3%)
      Web ping [ms]:  avg 223.6 - min 216.8 (97.0%), max 249.7 (111.7%)
    
    CN BEJ mirrors.bfsu.edu.cn [F: 1] - was 45.6 MB/s
      DL [Mb/s]:      avg 60.9 - min 0.0 (0.0%), max 74.7 (122.6%)
      Ping [ms]:      avg 154.6 - min 153.8 (99.5%), max 156.3 (101.1%)
      Web ping [ms]:  avg 170.2 - min 153.9 (90.4%), max 243.8 (143.3%)
    
    AU SYD mirror.internet.asn.au [F: 0] - was 24.5 MB/s
      DL [Mb/s]:      avg 38.3 - min 23.6 (61.5%), max 39.9 (104.2%)
      Ping [ms]:      avg 280.9 - min 94.0 (33.5%), max 294.6 (104.9%)
      Web ping [ms]:  avg 288.3 - min 286.1 (99.2%), max 304.3 (105.5%)
    
    AU SYD gsl-syd.mm.fcix.net [F: 0] - was 5 MB/s
      DL [Mb/s]:      avg 38.0 - min 36.5 (96.1%), max 39.4 (103.6%)
      Ping [ms]:      avg 287.1 - min 49.5 (17.2%), max 313.9 (109.3%)
      Web ping [ms]:  avg 294.3 - min 285.0 (96.9%), max 313.9 (106.7%)
    
    AU SYD mirrors.xtom.au [F: 0] - was 25.9 MB/s
      DL [Mb/s]:      avg 37.8 - min 36.8 (97.2%), max 38.8 (102.7%)
      Ping [ms]:      avg 285.2 - min 285.1 (99.9%), max 285.4 (100.1%)
      Web ping [ms]:  avg 293.2 - min 285.4 (97.3%), max 346.6 (118.2%)
    
    --- America ---
    
    CA MTL speedtest.mtl2.ca.leaseweb.net [F: 0] - new target
      DL [Mb/s]:      avg 90.4 - min 48.0 (53.1%), max 97.4 (107.8%)
      Ping [ms]:      avg 116.7 - min 115.5 (99.0%), max 122.6 (105.1%)
      Web ping [ms]:  avg 162.2 - min 115.5 (71.2%), max 479.5 (295.7%)
    
    US NYC nyc.mirrors.clouvider.net [F: 0] - was 68.1 MB/s
      DL [Mb/s]:      avg 98.0 - min 95.4 (97.4%), max 101.5 (103.6%)
      Ping [ms]:      avg 108.6 - min 108.4 (99.8%), max 108.8 (100.2%)
      Web ping [ms]:  avg 114.1 - min 108.9 (95.4%), max 130.4 (114.3%)
    
    US NYC mirrors-nyj.hawkhost.com [F: 0] - was 52.1 MB/s
      DL [Mb/s]:      avg 98.5 - min 96.0 (97.5%), max 100.4 (102.0%)
      Ping [ms]:      avg 109.8 - min 109.8 (100.0%), max 110.1 (100.2%)
      Web ping [ms]:  avg 110.9 - min 109.8 (99.0%), max 114.3 (103.1%)
    
    US ASH ash.speedtest.clouvider.net [F: 0] - new target
      DL [Mb/s]:      avg 87.1 - min 67.6 (77.7%), max 91.4 (105.0%)
      Ping [ms]:      avg 122.1 - min 121.8 (99.8%), max 122.7 (100.5%)
      Web ping [ms]:  avg 122.3 - min 121.9 (99.7%), max 122.7 (100.3%)
    
    US WDC speedtest.wdc2.us.leaseweb.net [F: 0] - new target
      DL [Mb/s]:      avg 80.9 - min 38.7 (47.8%), max 95.9 (118.5%)
      Ping [ms]:      avg 116.2 - min 115.8 (99.7%), max 116.6 (100.3%)
      Web ping [ms]:  avg 116.6 - min 115.9 (99.4%), max 116.8 (100.2%)
    
    US PIB mirror.pit.teraswitch.com [F: 0] - new target
      DL [Mb/s]:      avg 87.8 - min 86.4 (98.4%), max 90.7 (103.2%)
      Ping [ms]:      avg 125.6 - min 123.7 (98.5%), max 137.8 (109.7%)
      Web ping [ms]:  avg 126.6 - min 123.7 (97.7%), max 137.8 (108.8%)
    
    US MIA speedtest.mia11.us.leaseweb.net [F: 0] - new target
      DL [Mb/s]:      avg 78.4 - min 77.5 (98.9%), max 79.3 (101.2%)
      Ping [ms]:      avg 142.9 - min 142.8 (100.0%), max 143.1 (100.2%)
      Web ping [ms]:  avg 142.9 - min 142.8 (99.9%), max 143.1 (100.1%)
    
    US CHI ord.mirror.rackspace.com [F: 0]
      DL [Mb/s]:      avg 90.5 - min 89.8 (99.3%), max 91.3 (100.9%)
      Ping [ms]:      avg 125.7 - min 125.5 (99.9%), max 125.8 (100.1%)
      Web ping [ms]:  avg 129.2 - min 126.1 (97.6%), max 148.3 (114.8%)
    
    US CHI speedtest.chi11.us.leaseweb.net [F: 0] - was 39.0 MB/s
      DL [Mb/s]:      avg 90.0 - min 86.8 (96.5%), max 91.1 (101.2%)
      Ping [ms]:      avg 122.6 - min 122.5 (99.9%), max 122.8 (100.1%)
      Web ping [ms]:  avg 123.3 - min 122.6 (99.4%), max 125.9 (102.1%)
    
    US ATL atl.speedtest.clouvider.net [F: 0] - new target
      DL [Mb/s]:      avg 80.1 - min 64.7 (80.8%), max 83.5 (104.3%)
      Ping [ms]:      avg 134.0 - min 133.9 (99.9%), max 134.3 (100.2%)
      Web ping [ms]:  avg 135.7 - min 133.9 (98.7%), max 147.4 (108.6%)
    
    US PHO phx.speedtest.clouvider.net [F: 0] - new target
      DL [Mb/s]:      avg 64.5 - min 62.7 (97.1%), max 65.3 (101.2%)
      Ping [ms]:      avg 169.4 - min 169.2 (99.9%), max 169.5 (100.1%)
      Web ping [ms]:  avg 169.9 - min 169.4 (99.7%), max 172.7 (101.6%)
    
    US PHO speedtest.phx1.us.leaseweb.net [F: 0] - new target
      DL [Mb/s]:      avg 63.8 - min 62.9 (98.6%), max 64.2 (100.7%)
      Ping [ms]:      avg 175.4 - min 175.1 (99.8%), max 175.6 (100.1%)
      Web ping [ms]:  avg 175.5 - min 175.1 (99.8%), max 175.7 (100.1%)
    
    US PTL mirrors.cat.pdx.edu [F: 0] - new target
      DL [Mb/s]:      avg 62.3 - min 36.1 (58.0%), max 68.7 (110.2%)
      Ping [ms]:      avg 174.4 - min 174.3 (99.9%), max 174.6 (100.1%)
      Web ping [ms]:  avg 176.0 - min 174.3 (99.0%), max 180.9 (102.8%)
    
    US LAX mirror.alma.lax1.serverforge.org [F: 0] - was 39.9 MB/s
      DL [Mb/s]:      avg 62.8 - min 44.7 (71.2%), max 66.7 (106.1%)
      Ping [ms]:      avg 172.0 - min 171.7 (99.8%), max 172.6 (100.3%)
      Web ping [ms]:  avg 176.0 - min 171.8 (97.6%), max 183.6 (104.3%)
    
    US SEA speedtest.sea11.us.leaseweb.net [F: 0] - new target
      DL [Mb/s]:      avg 65.1 - min 38.0 (58.3%), max 69.9 (107.3%)
      Ping [ms]:      avg 161.8 - min 161.7 (99.9%), max 161.9 (100.0%)
      Web ping [ms]:  avg 161.9 - min 161.8 (99.9%), max 162.1 (100.1%)
    

    As usual split into continents/regions.

    Europe - with one single exception, IT, Milano, all targets show very similar performance or even significantly better performance, as indicated at the end of the first line of the listed targets.

    Asia/Oceania - Iran had quite decent about 100 Mb/s but now, probably due to political situation is very low. But quite a few targets show significantly better results now, again as indicated in the first line of each target.

    America - again, some targets show significantly improved performance, plus there are quite a few new targets.

    It seems that M247 (which seems to be Hosterions backend) have improved their routing. While cogent (sadly) appears often as the second leg, in particular to America, to Asia and Ozzyland I see GSL as second leg; a few times also zayo, L3, and twelve99 also appear.

    Whatever, overall there are quite some improvements wrt connectivity, but the main heroes of the upgrade are of course the new - and very significantly better! - processor and the much, much higher disk performance.

    TL;DR Very well done, @Hosteroid! You've managed to turn a decent and cheap but somewhat boring VPS into a really attractive one I enjoy! Now, of bloody course we wait for your other locations to follow.

    Thanked by 2Hosteroid JohnnySac
Sign In or Register to comment.