A dependency datastructure represents a processing task. The dependency datastructure comprising a plurality of components, each component encapsulating a code unit. The dependency datastructure may include dependency arcs to inter-component dependencies. Dependencies that are not satisfied by components within the dependency datastructure may be represented as pseudo-components. An execution environment identifies components that can be executed (e.g., have satisfied dependencies), using the dependency datastructure and/or concurrency state metadata. The execution environment may identify and exploit concurrencies in the processing task, allowing for multiple components to be executed in parallel.