Systems, methods, and apparatuses for implementing a scheduler and workload manager with dynamic workload termination based on cost-benefit analysis

Non-transitory computer readable storage media having instructions stored thereupon that, when executed by a system having at least a processor and a memory therein, the instructions cause the system to perform operations including:

• allocating a cache within the memory of the system;
• identifying, via a workload discovery engine, pending workload tasks to be scheduled for execution from one or more workload queues and updating the cache;
• identifying, via a compute resource discovery engine, a plurality of computing resources available to execute the workload tasks and updating the cache;
• executing a scheduler via the processor of the system, wherein the scheduler performs at least the following operations:
• retrieving information from the cache specifying the one or more computing resources available to execute workload tasks and specifying the plurality of workload tasks to be scheduled for execution;
• analyzing each of the plurality of workload tasks to be scheduled to determine if a Return On Investment (ROI) for each respective workload task exceeds a threshold;
• canceling any of the plurality of workload tasks to be scheduled having an ROI below the threshold;
• and
• scheduling the plurality of workload tasks having an ROI which exceeds the threshold for execution via the plurality of computing resources available to execute the workload tasks.

A system to implement a scheduling service, wherein the system comprises:

• a processor and a memory to execute instructions at the system;
• a cache allocated within the memory of the system to store information on behalf of a compute resource discovery engine and a workload discovery engine and a scheduler;
• wherein the compute resource discovery engine is to identify a plurality of computing resources available to execute the workload tasks and update the cache specifying the identified computing resources;
• wherein the workload discovery engine is to identify pending workload tasks to be scheduled for execution from one or more workload queues and update the cache with the identified workload tasks;
• wherein the scheduler is to retrieve information from the cache specifying the one or more computing resources available to execute workload tasks and specifying the plurality of workload tasks to be scheduled for execution;
• wherein the scheduler is further to analyze each of the plurality of workload tasks to be scheduled to determine if a Return On Investment (ROI) for each respective workload task exceeds a threshold;
• wherein the scheduler is further to cancel any of the plurality of workload tasks to be scheduled having an ROI below the threshold;
• and
• wherein the scheduler is further to schedule the plurality of workload tasks having an ROI which exceeds the threshold for execution via the plurality of computing resources available to execute the workload tasks.