#342 Cost Optimizations - Queuing Workloads Issue created by migueltol22 Motivation: Why do you think this is important? Flyte workflows produce a set of compute tasks (both single and multi pod) that run on a set of physical machines. These physical machines are procured at different rates depending on availability at that moment (reserved, on-demand, spot, etc). At any given time a workload may be cheaper or more expensive to run on. Deferring execution at a later time (e.g when reserved instances are freed up) can provide substantial cost savings for any org adopting flyte. Goal: What should the final outcome look like, ideally? Reduce costs by running workflows at a later more cost efficient time for workflows in which the user has selected to opt in to the queuing. Describe alternatives you've considered A clear and concise description of any alternative solutions or features you've considered. Flyte component • [x ] Overall ☐ Flyte Setup and Installation scripts ☐ Flyte Documentation ☐ Flyte communication (slack/email etc) • [ x] FlytePropeller • [ x] FlyteIDL (Flyte specification language) • [ x] Flytekit (Python SDK) • [ x] FlyteAdmin (Control Plane service) • [ x] FlytePlugins ☐ DataCatalog ☐ FlyteStdlib (common libraries) ☐ FlyteConsole (UI) ☐ Other [Optional] Propose: Link/Inline https://docs.google.com/document/d/1nqAp5-DanklNvG6OeQ3uIacZYGvf0oXioz7CboP_CsU/edit?usp=sharing Additional context Add any other context or screenshots about the feature request here. Is this a blocker for you to adopt Flyte Please let us know if this makes it impossible to adopt Flyte flyteorg/flyte