Instruction sets for generating schedules for task execution in computing systems

A method comprising:

• receiving application data indicative of a plurality of runnables corresponding to a computing application, the application data indicating:
• a dependency between two or more of the runnables of the plurality of runnables;
• respective compute engine types for execution of one or more runnables of the plurality of runnables;
• and
• respective timing constraints of one or more runnables of the plurality of runnables;
• generating, based at least on the application data and prior to execution of the computing application, a static execution schedule for execution of the plurality of runnables using a plurality of compute engines, the static execution schedule being deterministic and including one or more commands corresponding to one or more timing fences, the one or more timing fences dictating a timing and order of execution between at least a first runnable and a second runnable of the plurality of runnables;
• and
• sending the execution schedule to one or more machines, the one or more machines to using the execution schedule to execute multiple different iterations of the computing application deterministically as part of performing one or more planning, control, or navigation operations.

A system comprising:

• one or more central processing units (CPUs);
• and
• one or more graphics processing units (GPUs), wherein the one or more CPUs and the one or more GPUs are to execute, together, a computing application according to a static and deterministic execution schedule such that the one or more CPUs and the one or more GPUs execute the computing application in a same manner across multiple different execution iterations of the computing application, wherein:
• the execution schedule is based at least on receiving application data indicative of:
• a dependency between two or more of runnables of a plurality of runnables corresponding to the computing application;
• respective compute engine types for execution of one or more runnables of the plurality of runnables;
• and
• respective timing constraints of one or more runnables of the plurality of runnables;
• and
• the execution schedule dictates a same timing and order of execution of the plurality of runnables across the one or more CPUs and the one or more GPUs for individual execution iterations of the multiple different execution iterations of the computing application.