Applying Pay-Burst-Only-Once Principle for Periodic Power Management in Hard Real-Time Pipelined Multiprocessor Systems

Abstract

Pipelined computing is a promising paradigm for embedded system design. Designing a power management policy to reduce the power consumption of a pipelined system with nondeterministic workload is, however, nontrivial. In this article, we study the problem of energy minimization for coarse-grained pipelined systems under hard real-time constraints and propose new approaches based on an inverse use of the pay-burst-onlyonce principle. We formulate the problem by means of the resource demands of individual pipeline stages and propose two new approaches, a quadratic programming-based approach and fast heuristic, to solve the problem. In the quadratic programming approach, the problem is transformed into a standard quadratic programming with box constraint and then solved by a standard quadratic programming solver. Observing the problem is NP-hard, the fast heuristic is designed to solve the problem more efficiently. Our approach is scalable with respect to the numbers of pipeline stages. Simulation results using real-life applications are presented to demonstrate the effectiveness of our methods.

Minimizing Peak Temperature for Pipelined Hard Real-time Systems

“…4 shows those from SCC platform. From the figures we can see that: (1) In all the cases, the peak temperatures obtained from BS and GD are identical in value, which proves the effectiveness of our fast heuristic method. (2) For the two platforms, the temperature difference between our PBOO based algorithms and algorithm SDP gets bigger when stage number increases.…”

“…1 See Thm.1 in [1] The right hand side of the inequality can be upper-bounded by a set of minimum bounded-delay functions bd f min…”

“…This paper focuses on this issue and addresses the optimization problem by reversely using the Pay-Burst-OnlyOnce (PBOO) principle. Our work is inspired by the work of Chen et al [1], which minimizes the total power consumption for pipelined stage systems. However, their work cannot be directly transplanted to temperature optimization, due to the reasons: (1) although temperature is a strong function of power, power management techniques that are effective for energy saving may not be suitable for temperature managing [2], which has already been theoretically proved by [3].…”

“…However, their work cannot be directly transplanted to temperature optimization, due to the reasons: (1) although temperature is a strong function of power, power management techniques that are effective for energy saving may not be suitable for temperature managing [2], which has already been theoretically proved by [3]. (2) The quadratic programming problem formulation in [1] cannot be reused, since calculating temperature is based on convolution while computing the total power consumption is based on integral. Therefore, the problem of temperature minimizing demands new analysis and approach, which are the main contributions of this paper:…”

Pay-Burst-Only-Once in Real-Time Calculus

“…In RTC, the pay-burst-only-once property has been adopted to derive tighter delay bound, which is first proposed in [10]. Then it is used in [19] for the delay bound calculation under operator-finitary RTC, and also in [20] to minimize the energy consumption for pipelined multiprocessor embedded systems. However, none of these includes the proof of correctness.…”