Holistic job manager on Kubernetes capable of (i) provides an abstraction for wrapping all resources of the job/application and treating them holistically, (ii) queuing job/application creation requests and applying different queuing policies, e.g., FIFO, Priority, (iii) dispatching the job to one of multiple clusters, where a queuing agent runs, using configurable dispatch policies, and (iv) auto-scaling pod sets balancing job demands and cluster availability.