[UPDATE/FINAL] rootnerds multi-week VPS benchmark/test

Front up: It seems I can't change the title of my original thread https://www.lowendtalk.com/discussion/147995/rootnerds-multi-week-vps-benchmark-test to reflect that this is an update and final version. So if the mods like it's fine with me to move this content here to the original thread and to edit the title to this one (adding "[UPDATE/FINAL]" to it).

Otherwise, dear readers, pls have a look at the linked above original thread for context.

Here are the benchmark numbers since my last intermediate report (all values in MB/s or in Mb/s for network tests):

Proc/mem single-core avg: 239.9 (min: 233.6, max 243.7)
Proc/mem multi-core avg: 853.9 (min: 500.0, max 887.6)
Disk Test seq. write avg: 965.0 (min: 865.2, max 1005.0)
Disk Test rnd. write avg: 1469.8 (min: 1286.0, max 1624.0)
Disk Test seq. read avg: 3158.7 (min: 2500.0, max 3762.0)
Disk Test rnd. read avg: 2873.2 (min: 2432.0, max 3418.0)
Net Test AU, MEL avg: 57.5 (min: 40.4, max 61.6)
Net Test IN, CHE avg: 62.0 (min: 39.1, max 76.4)
Net Test UK, LON avg: 404.3 (min: 25.0, max 681.0)
Net Test DE, FRA avg: 656.8 (min: 267.7, max 930.2)
Net Test IT, MIL avg: 531.4 (min: 50.0, max 797.4)
Net Test FR, PAR avg: 575.3 (min: 50.0, max 792.6)
Net Test RO, BUC avg: 196.9 (min: 50.0, max 205.8)
Net Test GR, UNK avg: 265.5 (min: 10.0, max 344.1)
Net Test US, DAL avg: 118.6 (min: 10.0, max 127.9)
Net Test US, SJC avg: 103.2 (min: 10.0, max 108.3)
Net Test US, WDC avg: 131.1 (min: 10.0, max 183.5)
Net Test BR, SAO avg: 49.0 (min: 10.0, max 58.4)
Net Test JP, TOK avg: 37.0 (min: 10.0, max 48.3)
Net Test NO, OSL avg: 147.7 (min: 20.0, max 277.1)

And here for your convenience is a processed report of ALL test sets during the whole testing period:

Proc/mem single-core avg: 227.8 (min: 172.3, max 243.7)
Proc/mem multi-core avg: 732.6 (min: 97.0, max 904.8)
Disk Test seq. write avg: 2291.7 (min: 849.4, max 13096.0)
Disk Test rnd. write avg: 2270.7 (min: 1286.0, max 9185.0)
Disk Test seq. read avg: 4184.6 (min: 2500.0, max 12886.0)
Disk Test rnd. read avg: 3511.0 (min: 2311.0, max 9162.0)
Net Test AU, MEL avg: 54.8 (min: 0.0, max 62.7)
Net Test IN, CHE avg: 59.0 (min: 0.4, max 78.1)
Net Test UK, LON avg: 461.6 (min: 25.0, max 742.0)
Net Test DE, FRA avg: 696.0 (min: 267.7, max 930.2)
Net Test IT, MIL avg: 514.5 (min: 50.0, max 797.4)
Net Test FR, PAR avg: 607.2 (min: 50.0, max 794.0)
Net Test RO, BUC avg: 186.9 (min: 39.8, max 206.7)
Net Test GR, UNK avg: 277.7 (min: 10.0, max 349.1)
Net Test US, DAL avg: 118.5 (min: 10.0, max 131.0)
Net Test US, SJC avg: 102.7 (min: 10.0, max 108.3)
Net Test US, WDC avg: 144.4 (min: 7.6, max 184.1)
Net Test BR, SAO avg: 48.0 (min: 10.0, max 58.4)
Net Test JP, TOK avg: 36.6 (min: 10.0, max 48.3)
Net Test NO, OSL avg: 183.9 (min: 20.0, max 292.3)

Notes, thoughts, discussion

There are some network test numbers that have a nicely "round" min. value like "10.0". This is due to the logic in my evaluation script which assumes a reasonable minimum and only replaces it with a real measurement value if there is one that is actually below the assumed minimum.
In other words: Those "round" min. values are a positive sign indicating that the values never reached or dropped below those reasonable assumptions.

(Keep in mind that the benchmark was run MANY TIMES during some 3 or so weeks and the numbers you see here are all the numbers from all those test sets processed).

Looking at the differences over the whole testing period one notices some changes, sometimes positive sometimes negative, but those are easily with the frame of what is to be expected on a VPS, particularly an OpenVZ one. Possible (positive) exception: It seems that the Asia and Australia connectivity has got better, probably due to some enhanced routes/routing.

Verdict: clearly positive.

A really nice VPS for an OpenVZ VPS with quite good connectivity. While support reaction is quick their support seems to be (partly?) India based and a bit on the canned response side. That said, I'm not the average or even unexperienced OpenVZ customer and maybe @rootnerds has good reasons for that kind of first level support. Once (the very few and only during the start up phase) support issues had been escalated rootnerds 2nd level support was quick, professional, and helpful.

Clear recommendation, especially for OpenVZ fans.

Disclaimer: I'm not and have never been a rootnerds customer and I did not ask for or receive anything (other than the needed couple of weeks test access) from rootnerds and they also did not ask any favours or anything else whatsoever.