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
First, the "nice fragrance" or, if you prefer, rumour (quite solid, from the source): @Hosteroid is or very soon will be upgrading their nodes to a newer XEON generation and, if I'm not mistaken, faster NVMes. Why first? Because it sets the stage and is linked to the matter at hand.
I'll be honest: I regarded @virtono for quite some years as a "no qualms" go-to place for a reasonably decent and reasonably cheap VPS at diverse locations, and I won't complain and was happy with them ... until the market changed quite a bit and they seemed to not adapt. And then I came across Hosteroid. Really decent pricing and decent VPSs (in the price range I was and often am looking for). And of course I like their promos, which is how I got my first few VPSs from them.
Let me be clear: that game isn't - or at least wasn't hint, hint - about high end. It was about from which provider I/one can get reasonably decent VPS in a *variety of locations at a very good price and with decent and relatively fast and good support. Keep that in mind when you read this review!
Now, to the hard cold data. The locations I tested are AT, Vienna, Slovakia, Bratislava, and UK (London I guess but don't know for sure nor care). My basis setting my expectations was my current, i.e. not yet upgraded, RO, Bucharest VPS. What I basically wanted to know was whether I can expect halfway similar value for money in those other locations as well.
A propos "value for money": We're talking about 2 vCores Xeon E5v4, 4 GB memory and a decently sized disk (50GB) preferably NVMe but SSD is acceptable as well, at €18/yr (IIRC)!!
Oh, and before I get to the benchmark data, a big THANK YOU to Hosteroid for providing access for a week or two to those VPS for free!
All data presented here are based on 100+ runs.
First location Vienna, and first, as usual sysinfo (only once because it's pretty much the same in all 3 locations), processor and memory.
Version 2.5.0a, (c) 2018+ jsg (->lowendtalk.com)
Machine: amd64, Arch.: amd64, Model: Intel(R) Xeon(R) CPU E5-2697A v4 @ 2.60GHz
OS, version: FreeBSD 14.2, Mem.: 3.989 GB
CPU - Cores: 2, Family/Model/Stepping: 6/79/1
Cache: 32K/32K L1d/L1i, 256K L2, 40M 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 hle avx2 smep bmi2 erms invpcid rtm rdseed
adx smap umip syscall nx pdpe1gb rdtscp lm lahf_lm lzcnt
AES? Yes
InNested Virt.? Yes
HW RNG? Yes
ProcMem SC [MB/s]: avg 141.7 - min 45.5 (32.1 %), max 266.6 (188.1 %)
ProcMem MA [MB/s]: avg 455.1 - min 352.4 (77.5 %), max 526.9 (115.8 %)
ProcMem MB [MB/s]: avg 482.6 - min 353.2 (73.2 %), max 553.1 (114.6 %)
ProcMem AES [MB/s]: avg 464.5 - min 320.9 (69.1 %), max 616.1 (132.6 %)
ProcMem RSA [kp/s]: avg 70.9 - min 45.5 (64.2 %), max 91.8 (129.6 %)
All the commonly desired flags are set and while the single-core performance isn't exactly great, the multi-core performance is really decent. The crypto performance is half-cooked but that's just the way it is with XEONs (even the most modern and best ones do not compare to Epyc, let alone Ryzen).
And keep in mind that we're talking about a €18/year box!
Now the disk
--- Disk 4 KB - Buffered ---
Write seq. [MB/s]: avg 1.45 - min 0.92 (63.2%), max 1.57 (107.9%)
Write rnd. [MB/s]: avg 1.66 - min 1.02 (61.6%), max 1.82 (109.9%)
Read seq. [MB/s]: avg 7.79 - min 2.90 (37.2%), max 8.89 (114.1%)
Read rnd. [MB/s]: avg 6.51 - min 2.59 (39.8%), max 7.66 (117.7%)
--- Disk 4 KB - Sync/Direct ---
Write seq. [MB/s]: avg 1.43 - min 0.89 (62.1%), max 1.54 (107.5%)
Write rnd. [MB/s]: avg 1.66 - min 1.02 (61.5%), max 1.81 (109.1%)
Read seq. [MB/s]: avg 7.76 - min 3.18 (41.0%), max 8.78 (113.2%)
Read rnd. [MB/s]: avg 6.44 - min 2.77 (43.0%), max 7.51 (116.6%)
--- Disk 64 KB - Buffered ---
Write seq. [MB/s]: avg 31.38 - min 19.31 (61.5%), max 35.61 (113.5%)
Write rnd. [MB/s]: avg 26.64 - min 15.99 (60.0%), max 29.53 (110.8%)
Read seq. [MB/s]: avg 982.84 - min 871.29 (88.7%), max 1265.82 (128.8%)
Read rnd. [MB/s]: avg 78.53 - min 36.64 (46.7%), max 88.62 (112.9%)
--- Disk 64 KB - Sync/Direct ---
Write seq. [MB/s]: avg 6.96 - min 4.27 (61.3%), max 7.69 (110.5%)
Write rnd. [MB/s]: avg 4.12 - min 2.56 (62.2%), max 4.70 (114.2%)
Read seq. [MB/s]: avg 964.11 - min 868.36 (90.1%), max 1112.98 (115.4%)
Read rnd. [MB/s]: avg 79.13 - min 35.07 (44.3%), max 88.71 (112.1%)
--- Disk 1 MB - Buffered ---
Write seq. [MB/s]: avg 44.74 - min 28.21 (63.1%), max 49.75 (111.2%)
Write rnd. [MB/s]: avg 80.87 - min 48.92 (60.5%), max 90.00 (111.3%)
Read seq. [MB/s]: avg 1763.46 - min 1507.47 (85.5%), max 2138.43 (121.3%)
Read rnd. [MB/s]: avg 404.08 - min 247.86 (61.3%), max 469.65 (116.2%)
--- Disk 1 MB - Sync/Direct ---
Write seq. [MB/s]: avg 17.68 - min 12.73 (72.0%), max 20.58 (116.4%)
Write rnd. [MB/s]: avg 15.60 - min 11.20 (71.8%), max 18.90 (121.1%)
Read seq. [MB/s]: avg 1780.87 - min 1402.50 (78.8%), max 2345.20 (131.7%)
Read rnd. [MB/s]: avg 402.16 - min 252.19 (62.7%), max 455.27 (113.2%)
--- Disk IOps (Sync/Direct) ---
Write seq. [MB/s]: avg 6.40 - min 3.92 (61.2%), max 6.76 (105.6%)
IOps : avg 1638.80 - min 1003.92 (61.3%), max 1729.73 (105.5%)
Not even 10 MB/s and not much over 1500 IOps, that's meeh. Certainly acceptable for a very cheap VPS, but I really welcome Hosteroid's rumoured plan to upgrade the disks.
Finally the network
--- Europe ---
NO OSL mirror.terrahost.no [F: 0]
DL [Mb/s]: avg 247.8 - min 125.3 (50.6%), max 270.8 (109.3%)
Ping [ms]: avg 44.5 - min 42.5 (95.5%), max 93.6 (210.3%)
Web ping [ms]: avg 45.8 - min 42.5 (92.9%), max 93.6 (204.5%)
UK LON lon.speedtest.clouvider.net [F: 0]
DL [Mb/s]: avg 393.7 - min 200.9 (51.0%), max 431.2 (109.5%)
Ping [ms]: avg 27.7 - min 26.0 (94.0%), max 29.9 (108.0%)
Web ping [ms]: avg 27.8 - min 26.0 (93.6%), max 32.4 (116.6%)
NL AMS nl.mirrors.clouvider.net [F: 69]
DL [Mb/s]: avg 159.6 - min 0.0 (0.0%), max 443.1 (277.7%)
Ping [ms]: avg 27.8 - min 25.3 (91.0%), max 134.7 (484.3%)
Web ping [ms]: avg 27.9 - min 25.3 (90.8%), max 134.7 (483.6%)
DE FRA mirror.plusline.net [F: 0]
DL [Mb/s]: avg 390.6 - min 160.3 (41.0%), max 424.5 (108.7%)
Ping [ms]: avg 27.5 - min 0.0 (0.0%), max 82.3 (299.6%)
Web ping [ms]: avg 28.1 - min 0.0 (0.0%), max 82.3 (292.7%)
FR PAR mirror.in2p3.fr [F: 68]
DL [Mb/s]: avg 141.2 - min 0.0 (0.0%), max 391.6 (277.4%)
Ping [ms]: avg 29.3 - min 28.0 (95.5%), max 79.8 (272.2%)
Web ping [ms]: avg 42.4 - min 28.0 (66.0%), max 245.2 (578.3%)
CH GEN pkg.adfinis-on-exoscale.ch [F: 0]
DL [Mb/s]: avg 435.4 - min 192.1 (44.1%), max 468.9 (107.7%)
Ping [ms]: avg 24.9 - min 23.6 (94.8%), max 76.4 (306.9%)
Web ping [ms]: avg 27.7 - min 23.7 (85.4%), max 76.4 (275.4%)
IT MIL it1.mirror.vhosting-it.com [F: 69]
DL [Mb/s]: avg 104.7 - min 0.0 (0.0%), max 291.1 (278.1%)
Ping [ms]: avg 40.2 - min 38.5 (95.7%), max 90.4 (224.7%)
Web ping [ms]: avg 41.1 - min 38.5 (93.7%), max 90.4 (219.9%)
ES MAD mirror.es.stackscale.com [F: 0]
DL [Mb/s]: avg 223.7 - min 113.8 (50.9%), max 238.1 (106.5%)
Ping [ms]: avg 47.7 - min 46.6 (97.6%), max 83.2 (174.3%)
Web ping [ms]: avg 56.8 - min 47.7 (83.9%), max 131.5 (231.4%)
RO almalinux.mirrors.orange.ro [F: 70]
DL [Mb/s]: avg 37.3 - min 0.0 (0.0%), max 176.4 (472.8%)
Ping [ms]: avg 40.5 - min 38.6 (95.3%), max 80.3 (198.3%)
Web ping [ms]: avg 61.3 - min 38.6 (63.0%), max 328.4 (535.8%)
RU MOS speedtest.hostkey.ru [F: 0]
DL [Mb/s]: avg 218.2 - min 58.1 (26.6%), max 366.0 (167.7%)
Ping [ms]: avg 33.5 - min 32.8 (97.9%), max 37.5 (111.9%)
Web ping [ms]: avg 46.2 - min 32.9 (71.2%), max 79.2 (171.3%)
--- Asia / Oceania ---
RU SIB mirror.truenetwork.ru [F: 110]
DL [Mb/s]: avg 0.0 - min 0.0 (0.0%), max 0.0 (0.0%)
Ping [ms]: avg 111.0 - min 108.0 (97.3%), max 145.8 (131.4%)
Web ping [ms]: avg 111.0 - min 108.0 (97.3%), max 145.8 (131.4%)
IR TEH mirror.mobinhost.com [F: 0]
DL [Mb/s]: avg 112.7 - min 48.5 (43.1%), max 129.5 (114.9%)
Ping [ms]: avg 100.8 - min 0.0 (0.0%), max 191.8 (190.3%)
Web ping [ms]: avg 108.0 - min 0.0 (0.0%), max 195.5 (181.1%)
IN MUM mirrors.piconets.webwerks.in [F: 0]
DL [Mb/s]: avg 73.2 - min 59.5 (81.3%), max 77.7 (106.1%)
Ping [ms]: avg 148.2 - min 0.0 (0.0%), max 191.2 (129.0%)
Web ping [ms]: avg 153.0 - min 0.0 (0.0%), max 195.9 (128.1%)
SG SGP mirror.sg.gs [F: 68]
DL [Mb/s]: avg 25.4 - min 0.0 (0.0%), max 68.2 (268.4%)
Ping [ms]: avg 167.5 - min 165.3 (98.7%), max 213.4 (127.4%)
Web ping [ms]: avg 168.3 - min 165.3 (98.2%), max 213.4 (126.8%)
CN HKG mirrors.xtom.hk [F: 110]
DL [Mb/s]: avg 0.0 - min 0.0 (0.0%), max 0.0 (0.0%)
Ping [ms]: avg 263.2 - min 207.8 (78.9%), max 326.7 (124.1%)
Web ping [ms]: avg 263.2 - min 207.8 (78.9%), max 326.7 (124.1%)
CN NAJ mirror.nyist.edu.cn [F: 70]
DL [Mb/s]: avg 13.0 - min 0.0 (0.0%), max 43.9 (338.0%)
Ping [ms]: avg 279.1 - min 245.1 (87.8%), max 349.8 (125.3%)
Web ping [ms]: avg 290.9 - min 245.1 (84.3%), max 450.4 (154.8%)
JP TOK ftp.udx.icscoe.jp [F: 0]
DL [Mb/s]: avg 41.6 - min 37.1 (89.2%), max 43.7 (105.2%)
Ping [ms]: avg 267.0 - min 264.5 (99.1%), max 334.1 (125.2%)
Web ping [ms]: avg 268.1 - min 264.5 (98.6%), max 353.3 (131.8%)
AU SYD mirror.internet.asn.au [F: 68]
DL [Mb/s]: avg 15.7 - min 0.0 (0.0%), max 41.9 (267.3%)
Ping [ms]: avg 273.1 - min 270.6 (99.1%), max 326.3 (119.5%)
Web ping [ms]: avg 273.6 - min 270.6 (98.9%), max 326.3 (119.3%)
--- Africa ---
ZA JOB mirror.datakeepers.co.za [F: 69]
DL [Mb/s]: avg 20.3 - min 0.0 (0.0%), max 61.8 (304.0%)
Ping [ms]: avg 184.5 - min 182.8 (99.1%), max 216.2 (117.2%)
Web ping [ms]: avg 192.5 - min 182.8 (94.9%), max 418.3 (217.2%)
KE NAI mirror.liquidtelecom.com [F: 110]
DL [Mb/s]: avg 0.0 - min 0.0 (0.0%), max 0.0 (0.0%)
Ping [ms]: avg 174.2 - min 171.2 (98.3%), max 218.8 (125.6%)
Web ping [ms]: avg 174.2 - min 171.2 (98.3%), max 218.8 (125.6%)
--- America ---
US NYC nyc.mirrors.clouvider.net [F: 67]
DL [Mb/s]: avg 45.5 - min 0.0 (0.0%), max 119.1 (261.4%)
Ping [ms]: avg 94.9 - min 93.8 (98.9%), max 99.9 (105.3%)
Web ping [ms]: avg 95.4 - min 93.8 (98.4%), max 108.7 (114.0%)
US CHI linux-mirrors.fnal.gov [F: 66]
DL [Mb/s]: avg 36.2 - min 0.0 (0.0%), max 99.5 (274.8%)
Ping [ms]: avg 121.2 - min 119.6 (98.7%), max 154.3 (127.3%)
Web ping [ms]: avg 125.0 - min 119.6 (95.6%), max 160.1 (128.0%)
US LAX mirror.alma.lax1.serverforge.org [F: 69]
DL [Mb/s]: avg 26.8 - min 0.0 (0.0%), max 74.8 (279.5%)
Ping [ms]: avg 154.3 - min 150.9 (97.8%), max 185.2 (120.0%)
Web ping [ms]: avg 154.9 - min 150.9 (97.4%), max 185.2 (119.6%)
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.
Comments
Now, Slovakia, Bratislava, a location I like a lot albeit mostly for historical and personal reasons.
Again, first processor (also E5-2697A v4 @ 2.60GHz) and memory
Now, sorry, but this really is meeh. All criteria worse than the Vienna VPS. Nuff said.
Let's look at the disk
That's more like what I want to see! Solidly more than double the Vienna VPS, nice.
Is connectivity also better?
Nuh, it isn't, actually it's worse. While Vienna at least was decent within Europe, this one here is meeh even in Europe.
Let's just walk away and look at the UK VPS ...
Again, first sysinfo (differences only), processor and memory of the UK VPS.
Yep, I like that! Clearly the best processor and memory performance of the three (and also much better than my RO, Buc. VPS).
If compute power is your priority, this one clearly is the one for you!
Does the disk performance also look good?
Hmm, let's say "kinda". It's better than the Vienna one but worse than the Bratislava VPS.
But IMO well acceptable with a dirt-cheap VPS.
Maybe connectivity is a positive surprise?
Nuh, not really. Less mediocre than the Bratislava box but less performant than the Vienna one, albeit closer in performance to the better one.
Maybe important to some sidenote: The Vienna promo VPS normally comes with only about half the resources but I explicitely requested the one with the same resources the other two come with in order to keep the comparison fair.
TL;DR my verdict: it might surprise you but I'm OK with what I see (modulo the Vienna VPS which is significantly less attractive due to the price).
Why? Remember what I said in the intro. For me this was about a specific kind of provider, namely one with multiple locations and also good prices. And of course one can only expect so much from a 2vCores, 4 GB mem. 50 GB disk VPS for less than €20/year!
IMO Hosteroid's main competitors aren't providers who from time to time offer a particularly tasty and cheap VPS, nope, it's e.g. Greencloud that is, providers who also offer multiple and diverse locations and at good prices. Because many of us can't just rely on the occasional super-cheap VPS at a location that happens to meet our needs and/or preferences, nope, we also need a decent source that is, a provider who consistently sells relatively (or even very) cheap and a bit all over the world or at least within one's continent.
But still, if Hosteroid asked me for advice I'd suggest mainly two points, (a) to have disks with >= 10 MB/s and >= 5000 IOps (4k/4t) everywhere, and (b) to significantly improve their connectivity. And by that I don't mean impressive high-end connectivity like e.g. Hybula, but rather well acceptable connectivity or expressed less politely, connectivity with a solid distance to crappy.
That all said, I'm happy with both my Hosteroid VPS, one of which is even poorer than the three I benchmarked and reviewed here. Simple reason: bang per buck is fine; I do not expect miracles for less than €20/year, plus their support is relatively quick, good, and friendly.
If asked which of the 3 tested VPS I personally would buy the official answer is None, I'd wait for the improved nodes (and connectivity hopefully), and the inofficial answer is the one in Bratislava, albeit for personal and not necessarily rational reasons *g
When will the upgraded nodes come? I don't know but it seems the answer is "soon". And I'll likely get access to one early.
Finally: kudos to Hosteroid for showing the - well based - chuzpah to invite me to benchmark and review their VPSs!
Thank you so much for the detailed review and all the benchmark effort you’ve put in! 🙏 We really appreciate the kind words and the time you took to test multiple locations.
Yes, rumors are correct we are proceeding slowly with nodes upgrades to Intel Xeon Gold 6248.
when EPYC
coming here to show my precious romanian idler
I think it's like over 2 years already that I've been using @Hosteroid, and they've yet to disappoint me, honestly. The only locations I can't vouch are US, AT and NL, but all the rest have been quite rock solid.
@jsg you should give a try to a VPS in LT, IMO amongst the best location they have network-wise
@Hosteroid still hoping for APAC expansion
I actually might ... * hint, hint
As for Epyc: Nuh. I think Xeon is really plenty good enough for such a cheap 2 vCore, 4 GB and generous disk space VPS! I'd rather advise to "round some corners" - just as Hosteroid according to rumours plans ...
Oh, and I'm also a happy Hosteroid customer and have no complaints
honestly, the biggest red flag in a guy is when his vps has Xeon E5v4
I get your point, to some degree, but, let me word it like this: good luck finding 2 EPYC vCores, 4 GB memory and a 50 GB disk for less than €20/yr ...
So in conclusion, the nice fragrance in the air is a load of sh**?
My staff right now:
Disclaimer: take it with a grain of salt
then again
@Hosteroid
Happy to see you also aimed for the Scale Gen 2, you will most definitely like it.
Cheers and good speed !!!
@host_c Time for upgrades has came
Yep! While even 2nd gen Xeons still are, well, Xeons (read: not really high-end) I've seen significantly better results with Gen2 vs. v4.
That said, I for one still consider even v4 a decent processor for a very cheap VPS.
But sure, @Hosterion, bring on them upgrades! After all, there's no such thing as "too fast" on LET
imma put it in your ass tf does this comment have to do with Xeon E5v4 VPSes being shit
I had E5 tooo don't judge me
i am judging you so hard rn
China?
honestly i wish he was chinese but sadly no
why u wish me dimsumman? i love dimsum
... and so I did
@all
First, again, keep in mind what we're talking about: a 2 vCore, 4 GB RAM, + decently sized NVMe for less than €20 per year!
I emphasize that because I remember well what we considered a steal for$20/yr not that long ago: single vCore, 1 GB RAM and 20 GB or even just 15 GB SSD ...
Now to the LT VPS. First, as usual, sysinfo, processor and memory, based on 100+ runs.
I'll put it bluntly: single core performance is meeh. But multi-core is kind of OK, still a bit worse than Bratislava but acceptable, on can work with that.
Maybe we're luckier with the disk, let's see
YEP! That's more like it. Looks almost as if @Hosteroid had listened to my advice above and quickly adapted the LT VPS *g
Not yet fully there but close, quite close to the 20 MB/s and 5k IOps I talked about. Nice!
Now, let's look at the "amongst the best location they have network-wise" of zgato ...
Yes, zgato's take is correct, that indeed is the best connectivity I've seen so far with Hosteroid.
Let me explain: "good connectivity" for me is not just high performance, low latency, etc. Nope, the first thing I look at when I compile result sets is "how many failures?" that is, whether all (carefully selected and tested) targets are reached. That sets my basic impression. After all what's the worth of high performance with some targets when others aren't reached at all?
This LT VPS did reach all targets, even in Iran, China, and Ozzyland. That's something I only rarely see, even with way more expensive VPS!
Let's split it into continents:
Europe - Pretty much all, at least major, targets are about 250 Mb/s or more. Very nice, especially at the super-low price!
Funny side note: Moscow (officially "the enemy") achieved by far the best result! Oh well, fibers and photons don't care about politics *g
Asia / Oceania - A bit weird IMO but still quite decent overall. Hmmm, probably I should have put that the other way around: quite decent but with a few weird kinks.
Nothing much to explain about "quite decent", I mean most targets show decent results or even very decent ones (e.g. China). So I'll explain "weird kinks": Well, SGP is a bit slower than China - weird. Even weirder, after really good results in - notoriously difficult - China, suddenly a very significant drop in Japan. Also, far, far away Ozzyland showing (slightly) better results than Japan? Weird.
That said, those results all are decent, nothing to complain about, just wondering a bit.
Africa - well, both targets show good results, nice!
America - Well, NYC (just barely) passes my personal "I want to see at least 100 Mb/s" barrier, Chicago is quite close, and LAX is, oh well, LAX, i.e. not great but (a bit) over 70 Mb/s pass as "not bad".
There you have it. That LT VPS indeed is a good - and very cheap! - way to get decent and really good global connectivity. Well judged, zgato!
Summary/verdict: processor and mem. meeh, disk almost high-end (in that price range), connectivity decent to really good. Certainly a really good deal.
Want to get a peek at @Hosteroid's next gen? Then you are right here ...
This time we look at a USA, New Jersey VPS.
First sysinfo (full because it's worth it), processor and memory
Guys and gals, did you notice? We're almost in Epyc territory! Which of course also includes finally decent crypto hardware support. (Slightly) over 1 GB/s AES and almost 100 RSA keypairs, very nice!
I'd like to show mercy and keep it at simply saying "meeh", but I'm afraid I have to look sternly at Hosteroid and ask him "Really? Man, how could you pair such a nice processor with such a mediocre disk? Put some decent disk into that box and you have a winner!"
On to the network
Yay! Not bad at all! Let's break it down to continents
Europe - Me impressed, almost all target above 100 Mb/s, some even significantly higher, and only 2 somewhat worse. Plus almost no failures, really nice. I wish I'd see almost 150 Mb/s to the east coast from Europe that is, the other way around.
Asia/Oceania - decent as well. Funny side note: They seem to go to major parts of Asia via Europe, but then turn around and go to Japan and Ozzyland via the west cost. Probably makes sense from an american perspective. And quite decent results as well via the west coast it seems to me at least (with my Europe-centric perspective and very little experience with the routing over there). I mean, almost 60 Mb/s +- 5 Mb/s to Japan and Ozzyland is something I hardly could dream of from Europe. In short: nice!
America - WUT? 155 Mb/s from LAX? Almost unbelievable (for a European). Seriously though, OK, no Gb/s numbers, not even close but all in all quite decent! I've seen far worse for 3 times the price.
Summary/verdict: Modulo the mediocre disk a really nice surprise! I'm certainly not complaining, at all, about a VPS - for less than $20 per year, mind you - giving me 100 Mb/s or higher both within Murrica and to Europe.
In other words: This one is for you, Murricans, as well for Europeans who want good connectivity across the ocean as well as decent connectivity within Europe.
Finally again a THANK YOU to Hosteroid who provided those boxes for free for testing, as well as giving me - and now all of us - a glimpse at their next VPS generation!
Now I'll patiently (Haha, as if ..) for seeing their european VPS boxes upgraded to the processor level of this New Jersey box and the disk level of the LT box ...
Well done, Hosteroid!
Thanks a lot @jsg for your full review of our locations!
We will soon start rolling out nodes with better CPUs, and we will take care of those NVMe disks
Just got a .lt domain, would love to have a LT location, any anys
Thanks for sharing the very detailed review @jsg
If that is the case, you are lucky we have Vytautas package 2 more available in stock
how are you testing the network speed? too many fails on truenetwork doesn't seem right, especially when ping worked fine. what's webping is not clear.
Download speed on low-end servers from far away servers depends on tcp buffer sizes which are dynamic and depend on RAM size.
for example a 512mb ram vps with default and increased tcp buffers: https://imgur.com/a/FuV4RwS
results are vastly different
timing is also important, testing in the evening and in the day can also provide vastly different results.
consider adding this to your script: https://imgur.com/a/network-testing-script-ysZpjQf
You obviously don't know anything about how I do it. Although I often publicly explained it.
Don't you say, how surprising! Btw, do you even know what you're talking about?
I always do dozens of runs during day time and night time.
Certainly not. Among others for the reason that I'm not using a script (but a compiled program).
and this is obviously the reason i asked the question on how you do it, because i don't know, and you didn't answer it at all. you post too many comments and there is no search through them. i have no idea why you responsed this way, you clearly get mad answering the same question multiple times, so add all information to the first post.
thanks for another answer that doesn't make any sense, again... with your results it's impossible to compare vps providers networks because there will always be bias towards servers with more ram, and you refused to add a test which will always be the same on all servers.
there is little point in running the tests during the night, and my comment was about evening when the root node will be loaded the most, and it's the best time to see how many resources are left for your server.
you didn't answer again. (and who even cares if it's a script or a program...)
wow, your answer was completely pointless and toxic, i was trying to help you and you just showed your dirty ego without accomplishing anything, enjoy more failed tests which you don't even care about, misinforming yourself and others.
You arrived at the same conclusion as the rest of us. This guy is an incompetent poser. It's really a poor reflection on @jbiloh to give this tool the "Resident Benchmarker" handle when his test tool is buggy AF and the test results useless to the average user.
@Hosteroid is it possible for Schnitzel to be in specs like others specials with same price? ? I really like that location, it have super ping to my home.. Cheers
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.