Design Space Exploration

Source-code: https://code.ornl.gov/3d_cache_modeling_tool/destiny. PDF of extended technical report http://goo.gl/qzyWFE]

The continuous drive for performance has pushed the researchers to explore novel memory technologies (e.g. nonvolatile memory) and novel fabrication approaches (e.g. 3D stacking) in the design of caches. However, a comprehensive tool which models both conventional and emerging memory technologies for both 2D and 3D designs has been lacking. We present DESTINY, an architecture-level tool for modeling 3D (and 2D) cache designs using SRAM, embedded DRAM(eDRAM), spin transfer torque RAM (STT-RAM), resistive RAM (ReRAM) and phase change RAM (PCM). DESTINY facilitates design-space exploration across several dimensions, such as optimizing for a target (e.g. latency or area) for a given memory technology, choosing the suitable memory technology or fabrication method (i.e. 2D v/s 3D) for a desired optimization target etc. DESTINY has been validated against industrial cache prototypes. We believe that DESTINY will drive architecture and system-level studies and will be useful for researchers, designers and marketing professionals.

A lo largo de la historia de la arquitectura, los espacios se han construido mediante una agenda donde los objetos, los fenómenos físicos y las situaciones sociopolíticas, han tomado el protagonismo sobre el que se conducen las necesidades y utilidades de sus habitantes. A esta “agenda”, es a lo que llamamos el programa arquitectónico: un sistema de operaciones y funciones determinadas por las actividades de cada individuo. Hoy en día, dichos usos e intereses han variado. Y la ambición de un futuro automatizado y dependiente de extensiones corporales, tales como los dispositivos móviles y algunos objetos manipulados biométricamente; posee mayor prioridad e inmediatez en su aplicación, que la preocupación por desempeñar otras actividades naturales: biológicas y psicológicas.

Al arquitecto le ha sido conferida la tarea de organizar dichas actividades, transformarlas en operaciones, de nir sus alcances y, a partir de estas, construir la idea de un espacio. ¿Pero, qué sucede cuando los métodos de programación habitual no logran un vínculo con los nuevos intereses y actividades de los individuos que habitan un espacio? ¿Se puede diseñar un espacio sin determinar su uso, dejando en manos del habitante la transformación de las operaciones en dicho espacio? ¿Cómo se entendería entonces el concepto del “programa arquitectónico”, como re ejo de una sociedad de cambios tan inmediatos, en el siglo XXI? ¿Cuál es la agenda futura del programa arquitectónico?

For sustainable building design, performance-based optimization incorporating parametric modelling and evolutionary optimization can allow architects to leverage building massing design to improve energy performance. However, two key challenges make such applications of performance-based optimization difficult in practice. First, due to the parametric modelling approaches, the topological variability in the building massing variants is often very limited. This, in turn, limits the scope for the optimization process to discover high-performing solutions. Second, for architects, the process of creating parametric models capable of generating the necessary topological variability is complex and time-consuming, thereby significantly disrupting the design processes. To address these two challenges, this paper presents a parametric massing algorithm based on the subtractive form generation principle. The algorithm can generate diverse building massings with significant topological variabi...

In performance-based architectural design optimization, the design of building massings and façades is commonly separated, which weakens the effectiveness in performance improvement. In response, this study proposes a hybrid massing-façade integrated design generation and optimization workflow to integrate the two elements in an evolutionary design process. Compared with the existing approaches, the proposed workflow emphasizes the diversity of building design generation, with which various combinations of building massing forms and façade patterns can be systematically explored. Two case studies and a corresponding comparison study are presented to demonstrate the efficacy of the proposed workflow. Results show that the optimization can produce designs coupling the potential of building massings and façades in performance improvement. In addition, the optimization can provide information that supports early-stage architectural design exploration. Such information also enables the architect to understand the performance implications associated with the synergy of building massing and façade design.

This work introduces Evolutionary Architectural Space layout Explorer (EASE), a design tool that facilitates the optimization of 3D space layouts. EASE addresses architectural design exploration and the need to attend to many alternatives simultaneously in layout design. For this, we use evolutionary optimization to find a balance between divergent exploration and convergent exploitation. EASE comprises a novel sub-heuristic that constructs valid spatial layouts, a mathematical framework to quantify the satisfaction of constraints, and evolutionary operators to improve alternative layouts' fitness. We test EASE on the design of a library building. We evaluate EASE's performance for different building forms and different evolutionary algorithm parameters. The results suggest that EASE can generate valid layouts, quantify the constraints' degree of satisfaction and find a number of optimal layout solutions. The layouts that EASE generates are intended not as end results but design artifacts that provide insight into the solution space for further exploration.

В настоящее время одним из основных средств уменьшения сроков проектирова- ния и снижения затрат на разработку различных образцов как гражданской, так и военной техни- ки является использование систем автоматизиро- ванного проектирования: трехмерного проекти- рования (CAD), моделирования и инженерного анализа (CAE), управления данными об изде- лии (PDM) и др. Как известно, CAD-системы дают возможность создать 3D-образ объекта, но в большинстве случаев не позволяют определить совокупности физических свойств проектируе- мого изделия. Эта задача решается с помощью CAE-систем, которые по имеющейся модели из- делия позволяют рассчитать его технические и эксплуатационные характеристики. Однако большинство CAE-систем не приближает облик изделия и его характеристики к оговоренному в техническом задании – конструктор совместно с инженером-расчетчиком на основе анализа ре- зультатов, полученных в CAE-системе, должны самостоятельно определить новый облик изде-
лия.
Таким образом, для эффективного решения
задач по созданию новых образцов техники не- обходимо автоматизировать сам процесс поиска оптимального облика и внутренних свойств изде- лия. Для этого нужно связать CAD- и CAE-сис- темы, создав единую среду, а также применить формализованные методики научного поиска, используя методы оптимизации и анализа дан- ных.
Программный комплекс pSeven позволяет решить эти задачи. В частности, с его помощью можно интегрировать различные CAПР и исполь- зовать их совместно с современным математичес- ким аппаратом – алгоритмами анализа данных и оптимизации, автоматизируя тем самым про- цессы инженерного анализа и оптимизации пара- метров изделия. Лежащая в основе pSeven алго- ритмическая библиотека MACROS включает в себя развитый инструментарий для проведения оптимизации и анализа данных:
• передовые математические методы интеллекту- ального анализа данных, снижения размерности, анализа чувствительности;
• современные высокоэффективные алгоритмы оптимизации, которые позволяют решать слож- нейшие оптимизационные задачи за минимальное время и число итераций;
• инструменты для построения метамоделей по данным как натурных, так и вычислительных экспериментов, а также для анализа построенных моделей. Использование программного комплекса pSeven влечет за собой улучшение технических характеристик проектируемых изделий и умень- шение числа дорогостоящих натурных и ресурсо- емких вычислительных экспериментов.
Пользовательский интерфейс пакета pSeven представлен как графической средой (pSeven GUI), так и командной оболочкой (pSeven Shell). Оба ин- терфейса полностью эквивалентны функционально и дают пользователю доступ ко всем возможностям комплекса. В дальнейшем речь будет идти только о графической среде pSeven GUI.

This paper reports on the process and outcomes of a digital design studio that integrates parametric design and generative systems in architectural and urban design projects. It explores the interrelationship between the emergence of innovative formal representations using parametric design systems, and design autonomy; more specifically the conscious process of generating and developing an architectural concept. Groups of undergraduate students working on an architectural project are asked to identify a specific conceptual parti that relates to an aspect of architectural quality, define strategies that satisfy those aspects, and computational methodologies to implement those strategies, such as rule-based systems, self-organization systems, and genetic algorithms. The paper describes the educational approach and studio outcomes, discusses implications for CAAD education and curricula, and addresses issues to be considered for parametric and generative software development.

Предоставлены широкие возможности программного комплекса PSE/MACROS автоматиза- ции инженерных расчетов, содержащего современные высокоэффективные алгоритмы опти- мизации, средства интеграции с различными CAD/CAE системами, разнообразные средства многодисциплинарного анализа и визуализации данных. Представлена совокупность применяе- мых в конкретном проекте инструментов в удобном виде во встроенной графической среде. Показано, что при решении сложнейших задач внедрение комплекса существенно сокращает сроки проектирования, что позволяет улучшить технические и эксплуатационные характери- стики проектируемых изделий.

Abstract—This paper presents Anahita Processor Description Language (APDL) for generation of retargetable processor de- sign tool sets. The emphasis,is onthe applicability of the gener- ated tools in the ,design space exploration (DSE) phase of de- signing a new,embedded,processor. APDL descriptions can be used for generating cycle-accurate instruction set simulators, assembler/disassembler tools, production quality compilers and architecture verification tools. The paper

Space is a fundamental issue in organizing yet remains unexplored by cultural historical activity theory. The growing interest for the theory to study organizations calls for a reflection on how to grasp space. The activity system model, a triangle that underpins the analysis of organizations, is evaluated against the issue of the production of space. The episodic experience of the author in applying the model to the design of a new healthcare facility is presented for reflection. The facility was supposed to produce a new multi-organization activity; however, the production of this shared space made contradictions emerge beyond the reach of the activity system model. Some possibilities are considered to ground the model with the dialectics of the production of space, including the triad of representations of space, space of representations, and spatial practices.

When dealing with computationally expensive simulation codes or process measurement data, surrogate modeling methods are firmly established as facilitators for design space exploration, sensitivity analysis, visualization, prototyping and optimization. Typically the model parameter (=hyperparameter) optimization problem as part of global surrogate modeling is formulated in a single objective way. Models are generated according to a single objective (accuracy). However, this requires an engineer to determine a single accuracy target and measure upfront, which is hard to do if the behavior of the response is unknown. Likewise, the different outputs of a multi-output system are typically modeled separately by independent models. Again, a multiobjective approach would benefit the domain expert by giving information about output correlation and enabling automatic model type selection for each output dynamically. With this paper the authors attempt to increase awareness of the subtleties involved and discuss a number of solutions and applications. In particular, we present a multiobjective framework for global surrogate model generation to help tackle both problems and that is applicable in both the static and sequential design (adaptive sampling) case.

The paper describes a model-integrated approach for embedded software development that is based on domain-specific, multiple-view models used in all phases of the development process. Models explicitly represent the embedded software and the environment it operates in, and capture the requirements and the design of the application, simultaneously. Models are descriptive , in the sense that they allow the formal analysis, verification, and validation of the embedded system at design time. Models are also generative, in the sense that they carry enough information for automatically generating embedded systems using the techniques of program generators. Because of the widely varying nature of embedded systems, a single modeling language may not be suitable for all domains; thus, modeling languages are often domain-specific. To decrease the cost of defining and integrating domain-specific modeling languages and corresponding analysis and synthesis tools, the model-integrated approach is applied in a metamodeling architecture, where formal models of domain-specific modeling languages-called metamodels-play a key role in customizing and connecting components of tool chains. This paper discusses the principles and techniques of model-integrated embedded software development in detail, as well as the capabilities of the tools supporting the process. Examples in terms of real systems will be given that illustrate how the model-integrated approach addresses the physical nature, the assurance issues, and the dynamic structure of embedded software.