Controllable QoS for Imprecise Computation Tasks on DVFS Multicores With Time and Energy Constraints

Abstract

Abstract-Multicore architectures have been used to enhance computing capabilities, but the energy consumption is still an important concern. Embedded application domains usually require less accurate, but always in-time, results. Imprecise Computation (IC) can be used to divide a task into a mandatory subtask providing a baseline QoS and an optional subtask that further increases the baseline QoS. This work aims at maximizing the system QoS by solving task mapping and DVFS for dependent IC-tasks under real-time and energy constraints. Compared with existing approaches, we consider the joint-design problem, where task-to-processor allocation, frequency-to-task assignment, task scheduling and task adjustment are optimized simultaneously. The joint-design problem is formulated as an N P-hard Mixed-Integer Non-Linear Programming and it is safely transformed to a MixedInteger Linear Programming (MILP) without performance degradation. Two methods (basic and accelerated version) are proposed to find the optimal solution to MILP problem. They are based on problem decomposition and provide a controllable way to tradeoff the quality of the solution and the computational complexity. The optimality of the proposed methods is proved rigorously, and the experimental results show reduced computation time (23.7% in average) compared with existing optimal methods.

Data Placement for Multi-Tenant Data Federation on the Cloud

“…Then, we have p i,j = p i,j * p i,j . Then, the problem defined in Formula 19 is a Mixed Integer Linear Programming (MILP) problem, which is a proven NP-hard problem [43], [44], [45]. In this case, the exhaustive search for an optimal solution for p i,j increases exponentially and the complexity is O(N M ), which cannot be solved within a polynomial time, with N representing the number of storage types and M represents the number of data sets.…”

Imprecise Computation Task Mapping on Multi-core Wireless Sensor Networks

“…The QoS-aware task mapping problem is similar to Quadratic Assignment Problem, a well-known NP-hard problem. Therefore, finding an optimal solution satisfying all the given constraints (e.g., energy efficiency, deadline, QoS, task dependency, and DVFS) is very difficult and time consuming (17). Existing works usually focus on different contexts, as summarized in Table 1.…”

Collaborative State Estimation and Actuator Scheduling for Cyber-Physical Systems Under Random Multiple Events

“…Note that the problems (4) and (5) are ILP, which can be solved by existing ILP algorithms [9]. As the proposed scheduling methods are event-driven, i.e., when a new event occur, the scheduling decision, which is given by the solution of problem (4) or (5), should be updated.…”