2.1.1 Auto-Scaling
Auto-scaling is the act of dynamically adding local resources to a service in response to an increase in demand. This requires designing a vApp with the following characteristics:
The vApp should have one to multiple tiers. The
Tier 1 virtual machines
likely
require a connection to the load-balanced
network. The vApp must have at least one virtual machine
in each tier.
The vApp should be designed so that each tier can grow without having to reconfigure the vApp virtual machine’s
operating system and applications. The alternative is to customize
the scaling workflow to reconfigure these.
The different tiers can
be part of the same vApp (distributed) or different vApps. Different vApps allow for the centralization of
a tier. For example,
the database can stay on the private cloud.
The Tier 1 virtual machine
s in the vApp must use a static p
ool so that
IP addresses can be gathered to add or remove members to the load balancer pools.