Optimized deployment of analytic models in an edge topology

A computer program product, the computer program product comprising a computer readable storage medium having program code embodied therewith, the program code executable by at least one processor to perform operations for:

• receiving a description of a multi-tiered edge topology with a plurality of nodes, a multimedia stream analytics composition that is executed on the plurality of nodes, and performance objectives, wherein the multimedia stream analytics composition includes tasks that use analytic models, wherein the analytic models comprise neural networks;
• optimizing at least one of the analytic models to form a list of optimized analytic models by:
• for each unique pair of an optimization technique of a plurality of optimization techniques and a precision of a parameter of that analytic model of a plurality of precisions,
• generating an optimized analytic model based on the optimization technique and the precision;
• and
• adding the optimized analytic model to the list;
• benchmark testing the optimized analytic models on the list using test data and the plurality of nodes in the multi-tiered edge topology;
• clustering the optimized analytic models on the list based on benchmark results o form clusters of optimized analytic models;
• selecting a representative optimized analytic model from each of the clusters of optimized analytic models;
• determining a configuration recommendation that indicates deployment of the tasks and of each selected representative optimized analytic model on the plurality of nodes to meet the performance objectives;
• generating one or more workflows from the configuration recommendation to execute the tasks of the multimedia stream analytics composition on nodes of different tiers of the multi-tiered edge topology;
• and
• executing the one or more workflows on the plurality of nodes to generate output for the multimedia stream analytics composition.

A computer system, comprising:

• one or more processors, one or more computer-readable memories and one or more computer-readable, tangible storage devices;
• and
• program instructions, stored on at least one of the one or more computer-readable, tangible storage devices for execution by at least one of the one or more processors via at least one of the one or more memories, to perform operations comprising:
• receiving a description of a multi-tiered edge topology with a plurality of nodes, a multimedia stream analytics composition that is executed on the plurality of nodes, and performance objectives, wherein the multimedia stream analytics composition includes tasks that use analytic models, wherein the analytic models comprise neural networks;
• optimizing at least one of the analytic models to form a list of optimized analytic models by:
• for each unique pair of an optimization technique of a plurality of optimization techniques and a precision of a parameter of that analytic model of a plurality of precisions,
• generating an optimized analytic model based on the optimization technique and the precision;
• and
• adding the optimized analytic model to the list;
• benchmark testing the optimized analytic models on the list using test data and the plurality of nodes in the multi-tiered edge topology;
• clustering the optimized analytic models on the list based on benchmark results to form clusters of optimized analytic models;
• selecting a representative optimized analytic model from each of the clusters of optimized analytic models;
• determining a configuration recommendation that indicates deployment of the tasks and of each selected representative optimized analytic model on the plurality of nodes to meet the performance objectives;
• generating one or more workflows from the configuration recommendation to execute the tasks of the multimedia stream analytics composition on nodes of different tiers of the multi-tiered edge topology;
• and
• executing the one or more workflows on the plurality of nodes to generate output for the multimedia stream analytics composition.