For XO, I suggest you start a separate thread.
Regarding the host issue, without more details or information, it's difficult to say anything at the moment, especially since I haven't been able to reproduce it myself.
@dom0 As already previously mentioned XCP-ng Center / XenCenter are not officially supported and a third-party product.
It is generally advised to use Xen Orchestra for all administration / management tasks.
If it is a requirement for you to use a thick client (such as XCP-ng Center) you might want to try XenAdminQt: https://github.com/benapetr/XenAdminQt
It is also not officially supported but a very new project that gets updated frequently. Maybe that one works better for you.
@Pilow
While that could be true only 1 vm was shuffled to another SR. But all VMs now being backup with CBT enabled.
Well... i just check again at backups and after the last scheduled backup at 1am all vms fell back to full backup again.
Difficile de parler de « réalité » avec les benchmarks. Vérifiez aussi l'iodepth (qui dépend du type matériel que vous avez, sur du flash/NVMe vous pouvez monter à 128 ou 256), la latence entre en compte aussi bien sûr.
Le bottleneck principal est le monothreading de tapdisk, si vous testez sur plusieurs VMs différentes la somme va monter de manière assez régulière.