Library rt.restructuring.analysis.arrival.workload_bound

From rt.util Require Import sum.
From rt.restructuring.behavior Require Export all.
From rt.restructuring.model Require Import task schedule.priority_based.priorities.
From rt.restructuring.model.aggregate Require Import task_arrivals.
From rt.restructuring.model.arrival Require Import arrival_curves.
From rt.restructuring.analysis Require Import
workload ideal_schedule basic_facts.arrivals.

From mathcomp Require Import ssreflect ssrbool eqtype ssrnat seq path fintype bigop.

Task Workload Bounded by Arrival Curves

Section TaskWorkloadBoundedByArrivalCurves.

Consider any type of tasks ...

Context {Task : TaskType}.
Context `{TaskCost Task}.

... and any type of jobs associated with these tasks.

  Context {Job : JobType}.
  Context `{JobTask Job Task}.
  Context `{JobArrival Job}.
  Context `{JobCost Job}.

Consider any arrival sequence with consistent, non-duplicate arrivals...

  Variable arr_seq : arrival_sequence Job.
  Hypothesis H_arrival_times_are_consistent : consistent_arrival_times arr_seq.
  Hypothesis H_arr_seq_is_a_set : arrival_sequence_uniq arr_seq.

... and any ideal uniprocessor schedule of this arrival sequence.

Variable sched : schedule (ideal.processor_state Job).
Hypothesis H_jobs_come_from_arrival_sequence : jobs_come_from_arrival_sequence sched arr_seq.

Consider an FP policy that indicates a higher-or-equal priority relation.

Variable higher_eq_priority : FP_policy Task.
Let jlfp_higher_eq_priority := FP_to_JLFP Job Task.

For simplicity, let's define some local names.

Let arrivals_between := arrivals_between arr_seq.

We define the notion of request bound function.

Section RequestBoundFunction.

Let MaxArrivals denote any function that takes a task and an interval length and returns the associated number of job arrivals of the task.

Context `{MaxArrivals Task}.

In this section, we define a bound for the workload of a single task under uniprocessor FP scheduling.

Section SingleTask.

Consider any task tsk that is to be scheduled in an interval of length delta.

Variable tsk : Task.
Variable delta : duration.

We define the following workload bound for the task.

Definition task_request_bound_function := task_cost tsk × max_arrivals tsk delta.

End SingleTask.

In this section, we define a bound for the workload of multiple tasks.

Section AllTasks.

Consider a task set ts...

Variable ts : list Task.

...and let tsk be any task in task set.

Variable tsk : Task.

Let delta be the length of the interval of interest.

Variable delta : duration.

Recall the definition of higher-or-equal-priority task and the per-task workload bound for FP scheduling.

Let is_hep_task tsk_other := higher_eq_priority tsk_other tsk.
Let is_other_hep_task tsk_other := higher_eq_priority tsk_other tsk && (tsk_other != tsk ).

Using the sum of individual workload bounds, we define the following bound for the total workload of tasks in any interval of length delta.

Definition total_request_bound_function :=
\sum_(tsk <- ts ) task_request_bound_function tsk delta.

Similarly, we define the following bound for the total workload of tasks of higher-or-equal priority (with respect to tsk) in any interval of length delta.

      Definition total_hep_request_bound_function_FP :=
        \sum_(tsk_other <- ts | is_hep_task tsk_other )
         task_request_bound_function tsk_other delta.

We also define a bound for the total workload of higher-or-equal priority tasks other than tsk in any interval of length delta.

      Definition total_ohep_request_bound_function_FP :=
        \sum_(tsk_other <- ts | is_other_hep_task tsk_other )
         task_request_bound_function tsk_other delta.

    End AllTasks.

  End RequestBoundFunction.

In this section we prove some lemmas about request bound functions.

Section ProofWorkloadBound.

Consider a task set ts...

Variable ts : list Task.

...and let tsk be any task in ts.

Variable tsk : Task.
Hypothesis H_tsk_in_ts : tsk \in ts.

Assume that the job costs are no larger than the task costs.

Hypothesis H_job_cost_le_task_cost :
cost_of_jobs_from_arrival_sequence_le_task_cost arr_seq.

Next, we assume that all jobs come from the task set.

Hypothesis H_all_jobs_from_taskset : all_jobs_from_taskset arr_seq ts.

Let max_arrivals be any arrival bound for taskset ts.

Context `{MaxArrivals Task}.
Hypothesis H_is_arrival_bound : taskset_respects_max_arrivals arr_seq ts.

Let's define some local names for clarity.

    Let task_rbf := task_request_bound_function tsk.
    Let total_rbf := total_request_bound_function ts.
    Let total_hep_rbf := total_hep_request_bound_function_FP ts tsk.
    Let total_ohep_rbf := total_ohep_request_bound_function_FP ts tsk.

Next, we consider any job j of tsk.

    Variable j : Job.
    Hypothesis H_j_arrives : arrives_in arr_seq j.
    Hypothesis H_job_of_tsk : job_task j = tsk.

Next, we say that two jobs j1 and j2 are in relation other_higher_eq_priority, iff j1 has higher or equal priority than j2 and is produced by a different task.

Let other_higher_eq_priority j1 j2 := jlfp_higher_eq_priority j1 j2 && (~~ same_task j1 j2 ).

Next, we recall the notions of total workload of jobs...

Let total_workload t1 t2 := workload_of_jobs predT (arrivals_between t1 t2).

...notions of workload of higher or equal priority jobs...

Let total_hep_workload t1 t2 :=
workload_of_jobs (fun j_other ⇒ jlfp_higher_eq_priority j_other j) (arrivals_between t1 t2).

... workload of other higher or equal priority jobs...

Let total_ohep_workload t1 t2 :=
workload_of_jobs (fun j_other ⇒ other_higher_eq_priority j_other j) (arrivals_between t1 t2).

... and the workload of jobs of the same task as job j.

Let task_workload t1 t2 :=
workload_of_jobs (job_of_task tsk) (arrivals_between t1 t2).

In this section we prove that the workload of any jobs is no larger than the request bound function.

Section WorkloadIsBoundedByRBF.

Consider any time t and any interval of length delta.

Variable t : instant.
Variable delta : instant.

First, we show that workload of task tsk is bounded by the number of arrivals of the task times the cost of the task.

      Lemma task_workload_le_num_of_arrivals_times_cost:
        task_workload t (t + delta)
        ≤ task_cost tsk × number_of_task_arrivals arr_seq tsk t (t + delta).

As a corollary, we prove that workload of task is no larger the than task request bound function.

Corollary task_workload_le_task_rbf:
task_workload t (t + delta) ≤ task_rbf delta.

Next, we prove that total workload of other tasks with higher-or-equal priority is no larger than the total request bound function.

Lemma total_workload_le_total_rbf:
total_ohep_workload t (t + delta) ≤ total_ohep_rbf delta.

Next, we prove that total workload of tasks with higher-or-equal priority is no larger than the total request bound function.

Lemma total_workload_le_total_rbf':
total_hep_workload t (t + delta) ≤ total_hep_rbf delta.

Next, we prove that total workload of tasks is no larger than the total request bound function.

      Lemma total_workload_le_total_rbf'':
        total_workload t (t + delta) ≤ total_rbf delta.

    End WorkloadIsBoundedByRBF.

  End ProofWorkloadBound.

End TaskWorkloadBoundedByArrivalCurves.