Multithreaded executor

The primary objective is to manage the processing of numerous tasks concurrently through multithreading. This involves controlling the submission of tasks to prevent potential OutOfMemoryExceptions and other errors. The TaskManager serves as a logging mechanism to monitor the current state of queues and other relevant information.

There are two queues:

1. Queue to which we submit WorkUnit
2. Queue which is used by ThreadPoolExecutor to proceed submitted WorkUnits.

Both queues are blocking queues with configurable size. Queue for WorkUnit is limited by Config.workUnitsDequeSize Queue for tasks is limited by Config.tasksDequeSize

Usual use ('all code' example can be found AbstractTaskManagerTest):

        var taskManager = new AbstractTaskManager<...,...>{...};  // create a taskManager
        
        var workUnit = new WorkUnit(){...};  // create WorkUnit
                
         while(stopSubmit){
             taskManager.submit(workUnit)  // submit WorkUnit to taskManager to execute it
         }
        
        // Notify taskManager that we'll not have more tasks and wait for having all submitted tasks to be proceeded
        taskManager.wait(CustomWorkOfUnit.LAST_VALUE);

        // Log final statistics
        taskManager.logStatistics();

We can submit workUnits until size of its dequeue is less or equal toConfig.workUnitsDequeSize. The submit process will be blocked until at least one workOfUnit will go to tasksDeque (so a place for one more workUnit is freed).

TaskManager will take workUnit from workUnit's queue and will submit the task to execute this workUnit using his own thread pool executor with LimitedQueue as reservoir.