A parallel evolutionary algorithm to optimize dynamic memory managers in embedded systems

Abstract

a b s t r a c tFor the last 30 years, several dynamic memory managers (DMMs) have been proposed. Such DMMs include first fit, best fit, segregated fit and buddy systems. Since the performance, memory usage and energy consumption of each DMM differs, software engineers often face difficult choices in selecting the most suitable approach for their applications. This issue has special impact in the field of portable consumer embedded systems, that must execute a limited amount of multimedia applications (e.g., 3D games, video players, signal processing software, etc.), demanding high performance and extensive memory usage at a low energy consumption. Recently, we have developed a novel methodology based on genetic programming to automatically design custom DMMs, optimizing performance, memory usage and energy consumption. However, although this process is automatic and faster than state-of-the-art optimizations, it demands intensive computation, resulting in a time-consuming process. Thus, parallel processing can be very useful to enable to explore more solutions spending the same time, as well as to implement new algorithms. In this paper we present a novel parallel evolutionary algorithm for DMMs optimization in embedded systems, based on the Discrete Event Specification (DEVS) formalism over a Service Oriented Architecture (SOA) framework. Parallelism significantly improves the performance of the sequential exploration algorithm. On the one hand, when the number of generations are the same in both approaches, our parallel optimization framework is able to reach a speed-up of 86.40Â when compared with other state-of-the-art approaches. On the other, it improves the global quality (i.e., level of performance, low memory usage and low energy consumption) of the final DMM obtained in a 36.36% with respect to two wellknown general-purpose DMMs and two state-of-the-art optimization methodologies.

DeSpErate++: An Enhanced Design Space Exploration Framework Using Predictive Simulation Scheduling

“…This approach, that is the core of the proposed solution, cannot be based simply on a job scheduler for parallel and distributed environments [45], since it is not enough to support a structured DSE with optimization purposes. Combined with parallel optimization algorithms, the usage of a parallel computing environment has been successfully exploited in the field of iterative compilation [46], dynamic memory management [47], and microprocessor optimization [48]. On top of these works, the proposed method combines a parallel DSE algorithm with quality and time prediction models to support the scheduling of simulations on the parallel resources.…”

Genetic Improvement of Programs

“…If these prove incorrect, the efficiency of the whole application can be affected. Thus, provided each DMM can be trained on representative patterns of memory use, they make a good candidate for bespoke automated software generation [29].…”

Multi-objective optimization of dynamic memory managers using grammatical evolution

“…Based on the method proposed by the authors of [13], our optimization process begins with a profile of the memory operations of the target application. This profile can be used to create a grammar, which is based on the previous DMM hierarchy, improved by including some application-specific data like the block sizes and the non-terminal symbols.…”

“…Based on these proposals, several recent works proposed optimization methods to automatically obtain optimal custom DMMs through grammatical evolution [14,13,7]. In these works, the authors describe different implementations of optimization algorithms and contribute with several approaches like parallel implementations or reliability-aware optimizations.…”