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.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Netcup - New offer of root server gen. 11
This discussion has been closed.
Comments
Hi,
These "core"s, both in Netcup's offer and Yabs's counting, are actually threads [cores * 2], right? Or are they really cores?
2 cores == 4 Execution Threads
But main thing is these execution threads are NOT shared with other VPS users and that makes the real difference
They're so-called vCores, or threads
These 4 execution threads (aka 2 cores) are pretty powerful.
https://browser.geekbench.com/v6/cpu/5211752
Compare this performance with this dedicated server
Intel Xeon E3-1270 v3 … it has 4 cores (8 execution threads)
https://browser.geekbench.com/processors/intel-xeon-e3-1270-v3
That is the reason why this NetCup deal is extraordinary value
These are vCores. Host scheduler assigns them dynamically to any thread or core within CPU.
If host usage is low they will behave like 4 physical cores, if host usage is high they will perform worse than 4 threads (because of context switching).
You cannot have simpler answer. Its still a VPS, but its way less crowded and you wont get throttled.
With 4vCores you may get 300% multicore scaling, with 16vCores you may get 800-1000% scaling, cause with every new thread there is a risk of running into core that has another thread busy.
If you need 12vCores+ get dedi, if 4-8vCPU max then VDS/netcup root server gets you better value.
Excellent advice.
have glass?
Sorry I ment to say this is their VPS 1000 ARM G11 just for compression
is the voucher recurring? or just one time?
There is no vps image backup in Netcup. Only snapshot and raid 10 disks.
There were cases of data loss due to node failure in Contabo and Webtropia.
Has anyone experienced VPS data loss due to node failure? I searched the forums for Netcup but couldn't find it.
Are there really internal disaster recovery systems?
TBH, I never trust my backups with the same service providers. When shit happens, we should be prepared for the worst case.
This is why I use a external reliable hosting provider like iHostArt for my bulk backup requirements.
anyone has 8 core YABS?
from forum.netcup.de:
Edit: sorry for the formatting, was like this in the forum, tried to fix it a bit.
Edit: :sigh: got even worse, reverted it...
@Moopah
@kend
That's not what I want to ask
Of course, it's a good idea to save your own backups to an external location.
But how often does node failure occur in Netcup and has anyone experienced data loss due to this?
Vps is dead, there is no backup, you are on your own.. I'm looking for a story like this.
Of course, it's hard to know for sure, but I don't think that netcup has had a disaster (such as a node failure) that has led to data loss, at least not in recent times. Usually, the word spreads quickly in the case of a disaster. (That said, I may have missed something)
I've had one or more VPSes (vServers or Root-Servers) with netcup since 2017, and they've been remarkably reliable
Maybe they're implementing some kind of disaster recovery. I hope so
I can ask them about this but they won't confirm.
A few years ago my time4vps storage vps server went down due to node failure (disks and file system). The problem was them. They set up a new server and uploaded about 1 TB of my backup, although they did not promise it. This is a great thing. This is prestige and reliability. If we are the problem, we are the solution.
Yes same here !
One tip: Change hard drive driver to VIRTIO instead of SCSI to get double disk performance. (then reboot)
Because by default all installations will be on SCSI, which result a slower disk performance. If you check the forums then VIRTIO is the recommended for performance. And stable very well for me.
I've always kept to SCSI just because it was the default and also "recommended", but I was never sure why this was the case. It may just be a legacy setting that dates from the days when Virtio was less widely or less robustly supported by OSes
Can you post some yabs with virtio comparing with SCSI ?
Ugh.. I just moved from netcup (old root generation) to hetzner. Ran a script to compare the performance:
https://browser.geekbench.com/v6/cpu/compare/5216017?baseline=5310533
Seems netcup is beating my hetzner auction server?
Should I move from dedicated to root server? >_>
It depends on your needs, but in general I think it is worth it. Netcup root servers have dedicated resources and great uptime too.
Storage Space on Root Server is still a resource shared by many VPS users during read/write access
Something to keep in mind
Granted. Besides the increased single/multi core. I actually wonder if it's a good choice to move from a dedicated server to a KVM server. The DDR5 on netcup is a nice touch, so is NVMe. Not much fan of the daily bandwidth limitation though. But good thing there's no longer a monthly limitation.
Sorry I cant. I dont have any root server with SCSI enabled, only with VIRTIO, but expect double disk speeds with new linux distros almalinux-8-9 for example.
@Andru There is no change
YABS - VIRTIO
YABS - SCSI
So.. not so big difference..
Geekbench is not really indicator with VPS servers. Look at results of dedicated Epyc 9634 https://browser.geekbench.com/v6/cpu/4587873. You see 2048/17974 for DEDICATED EPYC server.
So how this dedicated divided into more than 20 parts can be more powerful than other dedicated that shows 1628/8150 (not that much difference with dedicated epyc)?