It's possible to adjust the level of high availability within a virtual infrastructure by optimising all existing mechanisms in order to reach the best "cost to high availability" ratio for a given block.
Currently, virtual platforms make redundancy and high availability possible throughout each of the links in an infrastructure (storage, hypervisor, network). This high availability does come at a price. The logic of redundancy extends into the management of VMs themselves. Capabilities like HA are complemented by "fault tolerance", the replication of VMs. It's therefore possible to compensate for any hardware deficiency while ensuring continuous service automatically or with a single click.
This agility and level of high availability naturally come at a price, namely redundant attribution of physical resources and the allocation of duplicate storage space.
These capabilities cannot, of course, be replicated across an entire platform (except with absolutely unlimited budgets, but the 1990s are well behind us...).
On the other hand, they help adapt the level of high availability of each VM according to its level of criticality, financial constraints and resource management constraints--in a scalable manner throughout the platform's life.
Ingénieur Avant-Ventes chez NEOCLES, ma mission consiste en du conseil en avant-ventes et études d'opportunités de projets de transformation d'infrastructures et/ou poste de travail.