I have tried googling this topic and I usually can’t find a very good answer / data.
To put it simply, I’m curious how the performance of XCP-NG matches up against that of ESXi.
I’d be interested in all aspects of performance if anyone has any info or data, even a “I’ve used both and noticed no difference” is fine for me as I don’t have experience with XCP-NG yet, and if you have used both already, thats valuable experience to me.
For example, if you RDP to a VM on ESXi, and RDP to a VM on XCP-NG, same number of processors, RAM, backing storage, etc… would one be “snappier” than the other?
With XCP-NG, is it better to use NFS storage, or iSCSI, or does it not matter as long as you set it up correctly?
I will certainly be watching Tom’s videos on the initial config of XCP-NG and TrueNAS (I have done TrueNAS before, but refreshers never hurt) but any other “gotcha’s” you could think of like if theres a box you have to check for 10gbit to work well or things like that.
I know this post is a bit all over the place, but basically what I’m asking for is if there is anyone that has seen any performance differences between XCP-NG and ESXi when you have used both for your given workload.
Performance is always specific to the workload. As such it is not possible to answer your question in such broad terms. Plus the bottleneck is always one factor. Whether it is the hardware in general, or the hardware in the context of a certain scenario is another aspect.
In addition, performance is looked at in terms of throughput and latency. There are workloads where latency is more or less irrelevant (batch processing is a typical example), and others (e.g. web shops) where it is absolutely critical.
Sorry, but without a lot more details it will be difficult to answer your question as such. On the other hand, in the context of a home lab a delta of 10-20% is typically not that important. And usually 20% is hard to detect outside of artificial benchmarks or specific performance tests.
Thanks for the input. Perhaps my question wasn’t the most clear. Is there anyone out there that has gone from ESXi to XCP-NG, or vice-versa and can speak to the performance difference (if there is any) for their given workload.
For example, if anyone out there ran web servers, and saw no difference between the two hypervisors. or if anyone out there ran SQL servers, and found that ESXi was better performing. That type of opinion would be helpful to me.