Howdy, Stranger!

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


rootnerds multi-week VPS benchmark/test
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.

rootnerds multi-week VPS benchmark/test

jsgjsg Member, Resident Benchmarker
edited July 2018 in Reviews

I'm pleased to present a @rootnerds VPS multi-week benchmark/test. This is an intermediate report but I see no reason so far to expect wildly different final results.

Prolog

As some might remember I did a (small, short) rootnerds VPS benchmark a while ago and the result was, uhm, not great. The rootnerds people investigated and my test seems to have taken place during a phase of major abuse. Once things were sorted out, rootnerds asked me to do another benchmark/test and this time over a couple of weeks.

I want to explicitly and clearly state that the rootnerds people have stayed cool and professional at all times. Plus they have in no way whatsoever interfered with my testing nor did they demand any favours. Kudos @rootnerds!

According to rootnerds the tested VPS is a "rootnerds myXXL" plan. The provided disk is 400 GB.

One (well, the only one) not exactly satifying experience I made is not to do with the VPS but with rootnerds support. While their reaction time is really good I had quite a bit of support ping pong during the initial phase with multiple support guys with indian names. Gladly though one of their managers (I suppose) saved the situation and had the problem sorted out.

Other than those first two or so days I experienced no significant problem with the VPS. It was available and responsive over the whole period so far.

The numbers

I present the numbers in two parts. Part one is the normal vpsbench sysinfo which doesn't change.

    Version 1.02b, (c) 2018+ jsg/moerm (->lowendtalk.com)
    Machine: amd64, Arch.: x86_64, Model: Intel(R) Xeon(R) CPU E5-2620 v4 @ 2.10GHz
    OS, version: Linux 2.6.32, Mem.: 4.0 GB
    CPU - Cores: 4, Family/Model/Stepping: 6/79/1
    Cache: 32K/32K L1d/L1i, 256K L2, 20M L3
    Std. Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat
              pse36 cflsh ds acpi mmx fxsr sse sse2 ss htt tm pbe sse3 pclmulqdq
              dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm
              pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline aes xsave
              osxsave avx f16c rdrnd
    Ext. Flags: fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm pqm
              fpcsds pqe rdseed adx smap intel_pt syscall nx pdpe1gb rdtscp lm
              lahf_lm lzcnt

Part two is the result of processing dozens of result sets over the whole period so far and shows average, min. and max. for each test. All numbers are in MB/s or for the network tests in Mb/s unless explicitely noted otherwise. The benchmarks were done multiple times every day and at different times over the full 24 hours.

Proc/mem single-core avg: 220.8 (min: 172.3, max 243.7)
Proc/mem multi-core avg: 661.8 (min: 97.0, max 902.9)
Disk Test seq. write avg: 3049.3 (min: 849.4, max 13.1 GB/s)
Disk Test rnd. write avg: 2731.6 (min: 1293.0, max 9.19 GB/s)
Disk Test seq. read avg: 4775.1 (min: 2500.0, max 12.9 GB/s)
Disk Test rnd. read avg: 3869.2 (min: 2311.0, max 9.2 GB/s)
Net Test AU, MEL avg: 53.3 (min: 0.0, max 62.7)
Net Test IN, CHE avg: 56.9 (min: 0.4, max 78.1)
Net Test UK, LON avg: 503.7 (min: 25.0, max 742.0)
Net Test DE, FRA avg: 711.2 (min: 374.2, max 929.6)
Net Test IT, MIL avg: 503.6 (min: 50.0, max 785.8)
Net Test FR, PAR avg: 619.5 (min: 50.0, max 794.0)
Net Test RO, BUC avg: 181.0 (min: 39.8, max 206.7)
Net Test GR, UNK avg: 284.3 (min: 10.0, max 349.1)
Net Test US, DAL avg: 118.1 (min: 10.0, max 131.0)
Net Test US, SJC avg: 102.3 (min: 10.0, max 107.1)
Net Test US, WDC avg: 151.6 (min: 7.6, max 184.1)
Net Test BR, SAO avg: 47.5 (min: 10.0, max 58.3)
Net Test JP, TOK avg: 36.1 (min: 10.0, max 42.1)
Net Test NO, OSL avg: 207.6 (min: 20.0, max 292.3)

Observations, Discussion

While the single core performance/memory results were within a reasonably tight frame the multi-core results had major variations which however are probably to be expected with an OpenVZ VPS. On average the results are very good.

The disk tests also had some variations but always stayed within a well acceptable frame. Generally speaking the provided vDisks perform really well.

The average network results are good to very good. 30 or even 50 Mb/s to destinations like Chennai, Melbourne and Tokyo are quite respectable. Cross-Atlantic connectivity is also quite good and in the range from 100 to 150 Mb/s. The Frankfurt destination which also happens to be where the VPS is hosted demonstrates that the port speed is quite close to 1 Gb/s and inner European connectivity is still quite good; all major points on the European internet are about half a Gb/s which is quite good.

As the min and max values suggest the average connectivity isn't too far from max which is positive. Unfortunately though, the min results indicate that while not often the connectivity can occasionally be really poor.

Summary

I personally don't like OpenVZ VPSs but it seems that rootnerds provides really nice VPSs. For an OpenVZ VPS the rootnerds VPS performs very well in all areas provided the nodes are well maintained (e.g. kept free from abusers) which seems to be the case.

I'm impressed both by the VPS and by the transparent and professional attitude of the rootnerds people. I wouldn't hesitate to recommend a good look at rootnerds for anyone interested in a OpenVZ VPS.

Thanked by 1RootNerds
Sign In or Register to comment.