A framework for modular analysis and exploration of heterogeneous embedded systems

Abstract

The increasing complexity of heterogeneous systems-on-chip, SoC, and distributed embedded systems makes system optimization and exploration a challenging task. Ideally, a designer would try all possible system configurations and choose the best one regarding specific system requirements. Unfortunately, such an approach is not possible because of the tremendous number of design parameters with sophisticated effects on system properties. Consequently, good search techniques are needed to find design alternatives that best meet constraints and cost criteria. In this paper, we present a compositional design space exploration framework for system optimization and exploration using SymTA/S, a software tool for formal performance analysis. In contrast to many previous approaches pursuing closed automated exploration strategies over large sets of system parameters, our approach allows the designer to effectively control the exploration process to quickly find good design alternatives. An important aspect and key novelty of our approach is system optimization with traffic shaping.

Learning early-stage platform dimensioning from late-stage timing verification

“…This powerful abstraction into few key parameters is the key enabler for both, the systematic and efficient analysis of timing and performance including full corner-case coverage, and the early application by means of "what-if" analyses for sensitivity analysis and design space exploration [18][2] [19].…”

A formal approach to robustness maximization of complex heterogeneous embedded systems

“…Note that both approaches utilize a framework for multi-dimensional design space exploration [5], which is based on evolutionary search techniques [2], and sensitivity analysis algorithms [7]. Figure 5(a) visualizes the separated robustness optimization approach.…”

Methods for power optimization in distributed embedded systems with real-time requirements

“…It is based on a previously published design space exploration framework [10], which uses multi-dimensional evolutionary search techniques [4,32].…”

“…We then give details on several important aspects: search space encoding (Section 4.2.2), the creation of an initial population used as starting point for the optimization (Section 4.2.3), and the variation operators used to guide the exploration (Sections 4.2.4 and 4.2.5). Figure 2 shows the compositional design space exploration framework [10] used in this work. It implements the standard exploration loop common to exploration frameworks based on evolutionary algorithms.…”

“…Its algorithms (creation of the initial population, crossover, and mutation) are tailored for power minimization with DVS. Note that the DVS chromosome can be combined with previously developed chromosomes for priority optimization [10], TDMA time slot optimization [9], traffic shaping [10], etc.…”