Currently, pulumi deployment executors are homogeneous machines with no configuration options available. This is understandable for compute and memory since those can be somewhat costly. It would be great to at least have the option to allocate more ephemeral storage to a runner for large docker image build jobs, etc. In an ideal world, we could also specify things like cpu, memory, accelerators etc, but for now, I'd be more than happy with just storage quota configuration.
I could see this being gated behind different pricing tiers, and configurable both at the organization and stack level, but a simple org-level config would be a great place to start.
Currently, pulumi deployment executors are homogeneous machines with no configuration options available. This is understandable for compute and memory since those can be somewhat costly. It would be great to at least have the option to allocate more ephemeral storage to a runner for large docker image build jobs, etc. In an ideal world, we could also specify things like cpu, memory, accelerators etc, but for now, I'd be more than happy with just storage quota configuration.
I could see this being gated behind different pricing tiers, and configurable both at the organization and stack level, but a simple org-level config would be a great place to start.
Affected feature
Pulumi Deployments