Multi-device processing activity allocation

A system, comprising:

• a processor configured to:
• identify, for a computer-executable process, a plurality of computing activities of the computer-executable process based on a detection of a computer-executable instruction used to perform the computer-executable process, wherein
• the computer-executable instruction is executed by one of a plurality of communicatively coupled computing devices;
• estimate, in real-time, for each computing activity of the plurality of computing activities, computing resources required to execute a respective computing activity of the plurality of computing activities on a computing device of the plurality of communicatively coupled computing devices;
• determine, in real-time, a current condition and configuration of each of the plurality of communicatively coupled computing devices;
• receive, from a user, an allocation criterion and a selection of a machine learning model from a plurality of machine learning models, wherein
• the allocation criterion is associated with an effect of a plurality of effects,
• each machine learning model of the plurality of machine learning models corresponds to a respective effect of the plurality of effects, and
• the selected machine learning model corresponds to the allocation criterion;
• predict, for each of the plurality of communicatively coupled computing devices and using the selected machine learning model, the effect induced by execution of one or more of the plurality of computing activities on the computing device, based on the current condition and configuration of the computing device;
• and
• allocate, in real-time, based on the allocation criterion and the predicted effect for each of the plurality of communicatively coupled computing devices, each of the plurality of computing activities among the plurality of communicatively coupled computing devices, wherein
• the machine learning model is trained based on data collected from prior real-time processing of example computing activities.

A computer program product, the computer program product comprising:

• one or more computer-readable storage media and program instructions collectively stored on the one or more computer-readable storage media, the program instructions executable by a processor to cause the processor to initiate operations including:
• identifying, for a computer-executable process, a plurality of computing activities of the computer-executable process based on a detection of a computer-executable instruction used to perform the computer-executable process, wherein
• the computer-executable instruction is executed by one of a plurality of communicatively coupled computing devices;
• estimating, in real-time, for each computing activity of the plurality of computing activities, computing resources required for executing a respective computing activity of the plurality of computing activities on a computing device of the plurality of communicatively coupled computing devices;
• determining, in real-time a current condition and configuration of each of the plurality of communicatively coupled computing devices;
• receiving, from a user, an allocation criterion and a selection of a machine learning model from a plurality of machine learning models, wherein
• the allocation criterion is associated with an effect of a plurality of effects,
• each machine learning model of the plurality of machine learning models corresponds to a respective effect of the plurality of effects, and
• the selected machine learning model corresponds to the allocation criterion;
• predicting, for each of the plurality of communicatively coupled computing devices and using the selected machine learning model, the effect induced by executing one or more of the plurality of computing activities on the computing device, based on the current condition and configuration of the computing device;
• and
• allocating, in real-time, based on the allocation criterion and the predicted effect for each of the plurality of communicatively coupled computing devices, each of the plurality of computing activities among the plurality of communicatively coupled computing devices, wherein
• the machine learning model is trained based on data collected from prior real-time processing of example computing activities.