Pareto-optimal Configurations Research Articles

SAT-Based Exact Modulo Scheduling Mapping for Resource-Constrained CGRAs

Coarse-Grain Reconfigurable Arrays (CGRAs) represent emerging low-power architectures designed to accelerate Compute-Intensive Loops (CILs). The effectiveness of CGRAs in providing acceleration relies on the quality of mapping: how efficiently the CIL is compiled onto the platform. State-of-the-Art (SoA) compilation techniques utilize modulo scheduling to minimize the Iteration Interval (II) and use graph algorithms like Max-Clique Enumeration to address mapping challenges. Our work approaches the mapping problem through a satisfiability (SAT) formulation. We introduce the Kernel Mobility Schedule (KMS), an ad hoc schedule used with the Data Flow Graph and CGRA architectural information to generate Boolean statements that, when satisfied, yield a valid mapping. Experimental results demonstrate SAT-MapIt outperforming SoA alternatives in almost 50% of explored benchmarks. Additionally, we evaluated the mapping results in a synthesizable CGRA design and emphasized the runtime metrics trends, i.e., energy efficiency and latency, across different CILs and CGRA sizes. We show that a hardware-agnostic analysis performed on compiler-level metrics can optimally prune the architectural design space, while still retaining Pareto-optimal configurations. Moreover, by exploring how implementation details impact cost and performance on real hardware, we highlight the importance of holistic software-to-hardware mapping flows, as the one presented herein.

Mathematical Models for the Design of GRID Systems to Solve Resource-Intensive Problems

Artificial neural networks are successfully used to solve a wide variety of scientific and technical problems. The purpose of the study is to increase the efficiency of distributed solutions for problems involving structural-parametric synthesis of neural network models of complex systems based on GRID (geographically disperse computing resources) technology through the integrated application of the apparatus of evolutionary optimization and queuing theory. During the course of the research, the following was obtained: (i) New mathematical models for assessing the performance and reliability of GRID systems; (ii) A new multi-criteria optimization model for designing GRID systems to solve high-resource computing problems; and (iii) A new decision support system for the design of GRID systems using a multi-criteria genetic algorithm. Fonseca and Fleming’s genetic algorithm with a dynamic penalty function was used as a method for solving the stated multi-constrained optimization problem. The developed program system was used to solve the problem of choosing an effective structure of a centralized GRID system that was configured to solve the problem of structural-parametric synthesis of neural network models. To test the proposed approach, a Pareto-optimal configuration of the GRID system was built with the following characteristics: average performance–103.483 GFLOPS, cost–500 rubles per day, availability rate–99.92%, and minimum performance–51 GFLOPS.

Multiple Phase Change Material-Based Heat Sink for Cooling of Electronics: A Combined Experimental and Numerical Study

Abstract This paper reports an investigation of the thermal performance of an energy storage heat sink incorporated with multiple phase change materials (PCMs). A six-cavity cylindrical heat sink heated at the base is chosen for the investigations with Docosane, n-Eicosane, and Tetracosane as candidate PCMs. The phase transition of PCMs has been visualized with a digital camera and three-dimensional numerical simulations. The results show that the latent heat exploitation process of PCMs in a heat sink with multiple PCMs is different from the single PCM heat sink, where the PCMs in all cavities melt distinctly rather at a time, thereby opening up windows for obtaining deeper insights that can lead to better performing heat sinks. A trained artificial neural network (ANN) with 78 representative heat sink configurations based on the arrangement of the PCMs in the cavities as input and charging and discharging times as output is used to swiftly drive the optimization engine. Finally, multi-objective optimization is performed using the artificial bee colony algorithm with simultaneous consideration of two conflicting objectives (i.e., maximizing charging cycle time and minimizing discharging cycle time) of the heat sink. From the optimization study, best performing nondominated Pareto optimal heat sink configurations are obtained and validated with the in-house experimental results. From the investigations, it is found that the heat sink configurations with multiple PCMs perform on par with the single PCMs in the charging process and show a superiority of up to 24% in discharging process over a heat sink with single PCMs in terms of time to reach set point temperature.

Learning from the Past: Efficient High-level Synthesis Design Space Exploration for FPGAs

The quest to democratize the use of Field-Programmable Gate Arrays (FPGAs) has given High-Level Synthesis (HLS) the final push to be widely accepted with FPGA vendors strongly supporting this VLSI design methodology to expand the FPGA user base. HLS takes as input an untimed behavioral description and generates efficient RTL (Verilog or VHDL). One major advantage of HLS is that it allows us to generate a variety of different micro-architectures from the same behavioral description by simply specifying different combination of synthesis options. In particular, commercial HLS tools make extensive use of synthesize directives in the form pragmas. This strength is also a weakness as it forces HLS users to fully understand how these synthesis options work and how they interact to efficiently set them to get a hardware implementation with the desired characteristics. Luckily, this process can be automated. Unfortunately, the search space grows supra-linearly with the number of synthesis options. To address this, this work proposes an automatic synthesis option tuner dedicated for FPGAs. We have explored a larger number of behavioral descriptions targeting ASICs and FPGAs and found out that due to the internal structure of the FPGA a large number of synthesis options combinations never lead to a Pareto-optimal design and, hence, the search space can be drastically reduced. Moreover, we make use of large database of DSE results that we have generated since we started working in this field to further accelerate the exploration process. For this, we use a technique based on perceptual hashing that allows our proposed explorer to recognize similar program structures in the new description to be explored and match them with structures in our database. This allows us to directly retrieve the pragma settings that lead to Pareto-optimal configurations. Experimental results show that the search space can be accelerated substantially while leading to finding most of the Pareto-optimal designs.

MultiETSC: automated machine learning for early time series classification

Early time series classification (EarlyTSC) involves the prediction of a class label based on partial observation of a given time series. Most EarlyTSC algorithms consider the trade-off between accuracy and earliness as two competing objectives, using a single dedicated hyperparameter. To obtain insights into this trade-off requires finding a set of non-dominated (Pareto efficient) classifiers. So far, this has been approached through manual hyperparameter tuning. Since the trade-off hyperparameters only provide indirect control over the earliness-accuracy trade-off, manual tuning is tedious and tends to result in many sub-optimal hyperparameter settings. This complicates the search for optimal hyperparameter settings and forms a hurdle for the application of EarlyTSC to real-world problems. To address these issues, we propose an automated approach to hyperparameter tuning and algorithm selection for EarlyTSC, building on developments in the fast-moving research area known as automated machine learning (AutoML). To deal with the challenging task of optimising two conflicting objectives in early time series classification, we propose MultiETSC, a system for multi-objective algorithm selection and hyperparameter optimisation (MO-CASH) for EarlyTSC. MultiETSC can potentially leverage any existing or future EarlyTSC algorithm and produces a set of Pareto optimal algorithm configurations from which a user can choose a posteriori. As an additional benefit, our proposed framework can incorporate and leverage time-series classification algorithms not originally designed for EarlyTSC for improving performance on EarlyTSC; we demonstrate this property using a newly defined, “naïve” fixed-time algorithm. In an extensive empirical evaluation of our new approach on a benchmark of 115 data sets, we show that MultiETSC performs substantially better than baseline methods, ranking highest (avg. rank 1.98) compared to conceptually simpler single-algorithm (2.98) and single-objective alternatives (4.36).

A comparison between central- and self-dispatch storage management principles in island systems

This paper presents a comparative evaluation of central and self-dispatch management concepts for battery energy storage (BES) facilities in island power systems with a high renewable energy source (RES) penetration. BES facilities deployed to support the integration of additional wind capacity can be either centrally dispatched by the island System Operator or they can be self-dispatched within a Virtual Power Plant entity comprising renewables and storage, called a Hybrid Power Station (HPS). To explore the anticipated benefits of each BES management paradigm, annual simulations are performed for an example island system, employing a three-layer mixed integer linear programming (MILP) method to simulate the unit commitment and economic dispatch processes. The levelized cost of energy (LCOE) of combined BES and renewables investments is calculated and the achieved RES penetration levels and island system generation cost are evaluated, allowing the identification of Pareto optimal storage configurations, leading to lowest LCOE for a given RES penetration target. Overall, the centrally dispatched BES systems prove to be substantially more cost-effective, compared to the self-dispatched alternative, for achieving similar RES penetration levels.

Quantification of sustainable animal manure utilization strategies in Hangzhou, China

ContextChina's livestock and poultry industries have been experiencing a transformation over recent decades, transitioning from family-size farms to larger, confined animal feeding operations. This development has significantly improved animal production capacity and reduced costs but has also created new challenges to manure management. One important concern is the conflicting interests of environmental protection and economic welfare between policymakers and manure utilization practitioners. ObjectiveIn this study, a regional manure utilization chain (RMUC) model was developed by recognizing optimal logistic configurations for manure and manure-based products between animal feeding operations, centralized processing facilities, and crop farms. We then use RMUC model to quantify the impact of management practices to the animal manure utilization chain of Hangzhou, China in the context of sustainable development. MethodThe RMUC model implemented an analytical target cascading structure with a multi-objective optimization algorithm to generate a set of Pareto-optimal configurations for discussing the regional economic costs and greenhouse gas (GHG) emission considering the practitioners' operational decisions to the designated manure management practices. Results and conclusionA comparative analysis quantified and prioritized the manure management practices (solid/liquid separation, manure reduction strategies); estimated economic and GHG emission credits of manure composition measurements; and indicated economic and GHG emission benefits of electric vehicles and the secondary infrastructures on manure distribution. The results showed sustainable metrics of the manure utilization improvement, including private costs, regional benefits, and the global impact of GHG emissions. SignificanceThe RMUC model demonstrated the compromise between practitioners' interests and public sustainability benefits given a certain level of constraints in decision process. Our analysis is an example of implementing computational models to deal with agricultural systematic problems with social, environmental, and economic concerns.

Performance and Energy Trade-Offs for Parallel Applications on Heterogeneous Multi-Processing Systems

This work proposes a methodology to find performance and energy trade-offs for parallel applications running on Heterogeneous Multi-Processing systems with a single instruction-set architecture. These offer flexibility in the form of different core types and voltage and frequency pairings, defining a vast design space to explore. Therefore, for a given application, choosing a configuration that optimizes the performance and energy consumption is not straightforward. Our method proposes novel analytical models for performance and power consumption whose parameters can be fitted using only a few strategically sampled offline measurements. These models are then used to estimate an application’s performance and energy consumption for the whole configuration space. In turn, these offline predictions define the choice of estimated Pareto-optimal configurations of the model, which are used to inform the selection of the configuration that the application should be executed on. The methodology was validated on an ODROID-XU3 board for eight programs from the PARSEC Benchmark, Phoronix Test Suite and Rodinia applications. The generated Pareto-optimal configuration space represented a 99% reduction of the universe of all available configurations. Energy savings of up to 59.77%, 61.38% and 17.7% were observed when compared to the performance, ondemand and powersave Linux governors, respectively, with higher or similar performance.

Flood Control Versus Water Conservation in Reservoirs: A New Policy to Allocate Available Storage

The aim of this study is to contribute to solving conflicts that arise in the operation of multipurpose reservoirs when determining maximum conservation levels (MCLs). The specification of MCLs in reservoirs that are operated for water supply and flood control may imply a reduction in the volume of water supplied with a pre-defined reliability in the system. The procedure presented in this study consists of the joint optimization of the reservoir yield with a specific reliability subject to constraints imposed by hydrological dam safety and downstream river safety. We analyzed two different scenarios by considering constant or variable initial reservoir level prior to extreme flood events. In order to achieve the global optimum configuration of MCLs for each season, we propose the joint optimization of three variables: minimize the maximum reservoir level (return period of 1000 years), minimize the maximum released outflow (return period of 500 years) and maximize the reservoir yield with 90% reliability. We applied the methodology to Riaño Dam, jointly operated for irrigation and flood control. Improvements in the maximum reservoir yield (with 90% reliability) increased up to 10.1% with respect to the currently supplied annual demand (545 hm3) for the same level of dam and downstream hydrological safety. The improvement could increase up to 26.8% when compared to deterministic procedures. Moreover, dam stakeholders can select from a set of Pareto-optimal configurations depending on if their main emphasis is to maintain/increase the hydrological safety, or rather to maintain/increase the reservoir yield.

Multiple scenarios multi-objective salp swarm optimization for sizing of standalone photovoltaic system

The paper presents a new multiple scenario multi-objective salp swarm optimization (MS-MOSS) algorithm to optimally size a standalone PV system. An accurate estimation of the number of PV modules and storage battery is crucial as it affects the system reliability and cost. Three scenarios have been presented focusing on Pareto optimal solutions by minimizing two conflicting objectives. Loss of load probability (LLP) and life-cycle cost (LLC) are considered to obtain the Pareto front. The iterative method is employed for validation of the superiority results of the proposed MS-MOSS algorithm. The results show that the scenarios are able to find Pareto optimal configuration at a high level of accuracy and at a very low cost. The proposed three scenarios are faster than iterative approach approximately by 158, 194.2, and 141.6 times, respectively. The third scenario outperforms other scenarios in terms of coverage and convergence of the distribution of solution to the Pareto front. As a conclusion, The MS-MOSS algorithm is found to be very effective in sizing of SAPV system.

A Design-Time/Run-Time Application Mapping Methodology for Predictable Execution Time in MPSoCs

Executing multiple applications on a single MPSoC brings the major challenge of satisfying multiple quality requirements regarding real-time, energy, and so on. Hybrid application mapping denotes the combination of design-time analysis with run-time application mapping. In this article, we present such a methodology, which comprises a design space exploration coupled with a formal performance analysis. This results in several resource reservation configurations, optimized for multiple objectives, with verified real-time guarantees for each individual application. The Pareto-optimal configurations are handed over to run-time management, which searches for a suitable mapping according to this information. To provide any real-time guarantees, the performance analysis needs to be composable and the influence of the applications on each other has to be bounded. We achieve this either by spatial or a novel temporal isolation for tasks and by exploiting composable networks-on-chip (NoCs). With the proposed temporal isolation, tasks of different applications can be mapped to the same resource, while, with spatial isolation, one computing resource can be exclusively used by only one application. The experiments reveal that the success rate in finding feasible application mappings can be increased by the proposed temporal isolation by up to 30% and energy consumption can be reduced compared to spatial isolation.

An Analytical Cache Performance Evaluation Framework for Embedded Out-of-Order Processors Using Software Characteristics

Utilizing analytical models to evaluate proposals or provide guidance in high-level architecture decisions is been becoming more and more attractive. A certain number of methods have emerged regarding cache behaviors and quantified insights in the last decade, such as the stack distance theory and the memory level parallelism (MLP) estimations. However, prior research normally oversimplified the factors that need to be considered in out-of-order processors, such as the effects triggered by reordered memory instructions, and multiple dependences among memory instructions, along with the merged accesses in the same MSHR entry. These ignored influences actually result in low and unstable precisions of recent analytical models. By quantifying the aforementioned effects, this article proposes a cache performance evaluation framework equipped with three analytical models, which can more accurately predict cache misses, MLPs, and the average cache miss service time, respectively. Similar to prior studies, these analytical models are all fed with profiled software characteristics in which case the architecture evaluation process can be accelerated significantly when compared with cycle-accurate simulations. We evaluate the accuracy of proposed models compared with gem5 cycle-accurate simulations with 16 benchmarks chosen from Mobybench Suite 2.0, Mibench 1.0, and Mediabench II. The average root mean square errors for predicting cache misses, MLPs, and the average cache miss service time are around 4%, 5%, and 8%, respectively. Meanwhile, the average error of predicting the stall time due to cache misses by our framework is as low as 8%. The whole cache performance estimation can be sped by about 15 times versus gem5 cycle-accurate simulations and 4 times when compared with recent studies. Furthermore, we have shown and studied the insights between different performance metrics and the reorder buffer sizes by using our models. As an application case of the framework, we also demonstrate how to use our framework combined with McPAT to find out Pareto optimal configurations for cache design space explorations.

Clinical trial supply chain design based on the Pareto-optimal trade-off between time and cost

ABSTRACTLong duration and high cost are two key characteristics of clinical trials. Since the clinical trial duration is part of the limited patent life and determines the time to market, its reduction is of critical importance. Although the duration can be reduced through increasing the number of clinical sites in the supply chain, the total enrollment and operational costs are also increased. Hence, Pareto-optimal supply chain configurations are used to improve clinical trial efficiency. In this study, we propose a multi-objective clinical site selection model that considers the trade-off between time and cost of clinical trials. An efficiency curve representing the Pareto-optimal trade-off is provided for decision makers to design the supply chain. To identify all Pareto-optimal supply chain configurations efficiently, we develop an algorithm based on proposed propositions. The optimality cuts are derived to improve the solving efficiency of this problem.

DyPO

Modern multiprocessor systems-on-chip (MpSoCs) offer tremendous power and performance optimization opportunities by tuning thousands of potential voltage, frequency and core configurations. As the workload phases change at runtime, different configurations may become optimal with respect to power, performance or other metrics. Identifying the optimal configuration at runtime is infeasible due to the large number of workloads and configurations. This paper proposes a novel methodology that can find the Pareto-optimal configurations at runtime as a function of the workload. To achieve this, we perform an extensive offline characterization to find classifiers that map performance counters to optimal configurations. Then, we use these classifiers and performance counters at runtime to choose Pareto-optimal configurations. We evaluate the proposed methodology by maximizing the performance per watt for 18 single- and multi-threaded applications. Our experiments demonstrate an average increase of 93%, 81% and 6% in performance per watt compared to the interactive, ondemand and powersave governors, respectively.

Cloud-Supported Coverage Control for Persistent Surveillance Missions

A cloud-supported coverage control scheme is proposed for multi-agent, persistent surveillance missions. This approach decouples assignment from motion planning operations in a modular framework. Coverage assignments and surveillance parameters are managed on the cloud and transmitted to mobile agents via unplanned and asynchronous exchanges. These updates promote load-balancing, while also allowing effective pairing with typical path planners. Namely, when paired with a planner satisfying mild assumptions, the scheme ensures that (i) coverage regions remain connected and collectively cover the environment, (ii) regions may go uncovered only over bounded intervals, (iii) collisions (sensing overlaps) are avoided, and (iv) for time-invariant event likelihoods, a Pareto optimal configuration is produced in finite time. The scheme is illustrated in simulated missions.

Multi-Objective Optimization of Hybrid Renewable Energy System Using an Enhanced Multi-Objective Evolutionary Algorithm

Due to the scarcity of conventional energy resources and the greenhouse effect, renewable energies have gained more attention. This paper proposes methods for multi-objective optimal design of hybrid renewable energy system (HRES) in both isolated-island and grid-connected modes. In each mode, the optimal design aims to find suitable configurations of photovoltaic (PV) panels, wind turbines, batteries and diesel generators in HRES such that the system cost and the fuel emission are minimized, and the system reliability/renewable ability (corresponding to different modes) is maximized. To effectively solve this multi-objective problem (MOP), the multi-objective evolutionary algorithm based on decomposition (MOEA/D) using localized penalty-based boundary intersection (LPBI) method is proposed. The algorithm denoted as MOEA/D-LPBI is demonstrated to outperform its competitors on the HRES model as well as a set of benchmarks. Moreover, it effectively obtains a good approximation of Pareto optimal HRES configurations. By further considering a decision maker’s preference, the most satisfied configuration of the HRES can be identified.

Multi-objective shape optimization of a heat exchanger tube fitted with compound inserts

Abstract This paper reports a multi-objective optimization of a heat exchanger tube fitted with compound insert geometries to obtain a maximum convective heat transfer rate and minimum pressure loss. A series of experimental runs has been constructed using response surface methodology (RSM) to study the effect of geometrical and flow parameters on heat transfer and friction characteristics along with parameter optimization. The study involves the dimensionless parameters: Reynolds number (Re), pitch ratio (PR), twist ratio (TR), and twisted tape number (N). The characteristic numbers involving heat transfer characteristics (Nusselt number ratio, Nu / Nu s ) and friction characteristics (friction factor ratio, f / f s ) calculated from experimental data are served as the response variables for RSM. The objective functions for Nu / Nu s and f / f s are constructed using RSM and utilized in multi-objective genetic algorithm to obtain the Pareto-optimal fronts. The results show that the decrease in the PR, TR and increase in N, Re are measured in the augmentation of both the objective functions. It is concluded that the acquired Pareto optimal configurations provide principal understanding into the design parameters and allow autonomy of selection among the optimal solutions. A thermal enhancement factor is used to study the thermal-hydraulic performance of the Pareto optimal solutions.

A Kriging framework for the efficient exploitation of the nanoscale junctioless DG MOSFETs including source/drain extensions and hot carrier effect

Recently, the Junctionless Double Gate MOSFET with source/drain extensions has proved competitive performance measures due to the reduction of the series resistance between the metal contacts and source/drain regions. However, the precise modeling of different response functions is still an important issue especially at nanoscale level because of the inclusion of complex phenomena such as quantum transport mechanisms and hot carrier degradation effect. In this perspective, our aim in the presented work is to investigate the efficiency of a novel approach based on Kriging metamodeling and multi-objective particle swarm optimization for the exploitation of the considered device in terms of some analog/RF performances. Data generated by ATLAS-2D simulator, are used to establish Kriging metamodels for the Gain and Transconductance Generation Factor. It is highlighted that the obtained models can be used accurately in a multi-objective context to offer several Pareto optimal configurations. As a result, a wide range of biasing configurations is available to the designer depending on imposed constraints.

5 by 5 Microstrip Antenna Array Design by Multiobjective Differential Evolution based on Fuzzy Performance Feedback

Differential evolution is often regarded as one of the most efficient evolutionary algorithms to tackle multiobjective optimization problems. The key to success of any multiobjective evolutionary algorithms (MOEAs) is maintaining a delicate balance between exploration and exploitation throughout the evolution process. In this paper, the authors develop an Improved version of the Fuzzy-based Multiobjective Differential Evolution (IFMDE) that exploits performance metrics, specifically hypervolume, spacing, and maximum spread, to measure the state of the evolution progress. They apply the fuzzy inference rules, derived from domain knowledge, to these metrics in order to dynamically adjust the associated control parameters of a chosen mutation and crossover strategy used in this algorithm. One mutation parameter controls the degree of greedy or exploitation, while another regulates the degree of diversity or exploration of the reproduction phase. On the other hand, crossover rate controls the fraction of trial vector elements inherited from the mutant vectors. In doing so collectively, the authors can appropriately adjust the degree of exploration and exploitation through performance feedback. A 5 by 5 microstrip antenna array design problem is formulated as a three-objective optimization problem. The proposed IFMDE is applied to tackle this problem under real-world complications. Since the objective evaluations of a 5 by 5 microstrip antenna array are computationally very expensive, a radial basis function neural network is trained as a surrogate model for the fitness function approximations. The experimental results demonstrate the ability of IFMDE that it can find not only one, but a set of Pareto optimal solutions, specifically in terms of side lobe level and reflection coefficient. These multiple Pareto-optimal configurations can then be chosen from by a decision maker given dynamic operating environments, constraints and uncertainties.

Multiple Criteria Analysis of the Airport Terminal Effectiveness by Multi-objective Optimization and Simulation

A rational approach to terminal airport management is not a trivial task due to relatively complex interactions between passengers and terminal infrastructure. Such infrastructure may be represented or modelled as a network of service nodes. To make a decision about such a network structure, one has to take into account not only the cost of terminal infrastructure, but also a set of quality indicators depicting passenger service level. Such decision problems may be formulated in the multiple criteria setting. We propose a bi-criteria decision making problem with a discrete-event simulation model of a terminal airport as a base model. The simulation model is used to evaluate a finite set of configurations representing a network of service nodes. To point out the most preferred Pareto optimal configuration, we propose to use an interactive decision making method to navigate Pareto optimal solutions with so-called vectors of concessions and reference points as preference carriers. Such versatile decision making scheme may be used to solve practical multiple criteria decision problems with values of criteria functions obtained by simulation runs.