Dynamically mapping tasks with priorities and multiple deadlines in a heterogeneous environment

Abstract

In a distributed heterogeneous computing system, the resources have different capabilities and tasks have different requirements. To maximize the performance of the system, it is essential to assign the resources to tasks (match) and order the execution of tasks on each resource (schedule) to exploit the heterogeneity of the resources and tasks. Dynamic mapping (defined as matching and scheduling) is performed when the arrival of tasks is not known a priori. In the heterogeneous environment considered in this study, tasks arrive randomly, tasks are independent (i.e., no inter-task communication), and tasks have priorities and multiple soft deadlines. The value of a task is calculated based on the priority of the task and the completion time of the task with respect to its deadlines. The goal of a dynamic mapping heuristic in this research is to maximize the value accrued of completed tasks in a given interval of time. This research proposes, evaluates, and compares eight dynamic mapping heuristics. Two static mapping schemes (all arrival information of tasks are known) are designed also for comparison. The performance of the best heuristics is 84% of a calculated upper bound for the scenarios considered.

Time Utility Functions for Modeling and Evaluating Resource Allocations in a Heterogeneous Computing System

“…The K-percent Best algorithm [1], [6], [7], [13], [23], [30], [31] combines both the MET and the Maximum Utility heuristics by choosing machines based on both these heuristics. When a new task arrives, the K-percent Best algorithm picks a machine in two stages.…”

Stochastic robustness metric and its use for static resource allocations

“…The CR heuristic uses the sufferage concept introduced in [29], and used in [24]. Like the Two-Phase heuristic, in every iteration this heuristic first determines the best assignment for each of the applications left unmapped.…”

Investigation of the Scheduler for Heterogeneous Distributed Computing Systems based on Minimal Cover Method

“…It is defined by such metrics as utilization resources, run time of all tasks on the resources, total time completion of the tasks set later than other ones, etc. Now large amount of papers is devoted to the comparative analysis batch scheduling methods [1][2][3][4][5][6][7][8][9][10][11][12][13][14]. For example, a detailed description and classification scheduling algorithms for Grid computing is given in [12,13,14].…”

“…We investigated articles that examine the most common method of FCFS, the main advantage of which is the lack of information on the requirements of task (a method scheduling the task, standing first in order in query for the first free resource), with batch modes [6][7][8][9]. However, FCFS has significant disadvantages: its efficiency decreases sharply with increasing intensity of tasks flows and the heterogeneity of computing environment: resource queues are formed, which greatly degraded the utilization of resources due to their inactivity [8,9].…”

“…However, FCFS has significant disadvantages: its efficiency decreases sharply with increasing intensity of tasks flows and the heterogeneity of computing environment: resource queues are formed, which greatly degraded the utilization of resources due to their inactivity [8,9]. More preferred is batch scheduling mode.…”