We here provide an extension to the Linux Completely Fair Scheduler to support scheduling of heterogeneous architectures, i.e., architectures including GPUs, FPGAs, or other specialized co-processors.
Using an associated programming model we are able to leverage the available hardware by being able to run several applications concurrently on the available hardware. In addition our programming model supports time-shared use of the architectures by performing cooperative multitasking.
Please refer to the documentation for a detailed description of our implementation and instructions of how to install and use our scheduler.