An analog circuit design environment based on cooperating blackboard systems

A novel approach to design automation for analog circuits is presented. The prototype implementation-OASE-has been realized as a set of cooperating expert systems with blackboard architecture. The circuit specific knowledge bases use hybrid representation schemes and are strictly separated from an execution engine containing the necessary control knowledge. This alleviates the knowledge acquisition process as well as the extension and maintenance of existing knowledge. OASE has been developed as a design assistant, featuring different levels of interactivity, a hierarchical design style and fully embedded algorithmic standard tools. In its current version it is able to design a broad range of different CMOS operational amplifiers.     

A review on queueing network models with finite capacity queues for software architectures performance prediction

A review is carried out on how queueing network models with blocking have been applied so far into the performance evaluation and prediction of Software Architectures (SA). Queueing network models with finite capacity queues and blocking have recently been introduced and applied as more realistic models of systems with finite capacity resources and population constraints. Queueing network models have been often adopted as models for the evaluation of software performance. Starting from our own experience, we observe the need of a more accurate definition of the performance models of SA to capture some features of the communication systems. We consider queueing networks with finite capacity and blocking after service (BAS) to represent some synchronization constraints that cannot be easily modeled with queueing network models with infinite capacity queues. We investigate the use of queueing networks with blocking as performance models of SA with concurrent components and synchronous communication. Queueing theoretic analysis is used to solve the queueing network model and study the synchronous communication and performance of concurrent software components. Our experience is supported by other approaches that also propose the use of queueing networks with blocking. Directions for future research work in the field are included.     

Keynote paper: Architectures for integrating manufacturing activities and enterprises

This paper is a summary of the major technical report (Williams et al., 1993) of the IFAC/IFIP Task Force on Architectures for Integrating Manufacturing Activities and Enterprises. It presents a synopsis of the investigations of pertinent architectures undertaken, and the findings generated relating to the suitability of various architectures for the integration task. It also presents the Task Force's recommendations for achieving a “complete” architecture in terms of the necessary capabilities by “completing” a currently available architecture. The Task Force also outlined how a “best” architecture could be achieved by selecting and combining the best features of the available architectures.     

Parallelization of WHILE loops on pipelined architectures

Modulo scheduling theory can be applied successfully to overlap Fortran DO loops on pipelined computers issuing multiple operations per cycle both with and without special loop architectural support. This paper shows that a broader class of loops—REPEAT-UNTIL, WHILE, and loops with more than one exit, in which the trip count is not known beforehand—can also be overlapped efficiently on multiple-issue pipelined machines. The approach is described with respect to a specific machine model, but it can be extended to other models. Special features in the architecture, as well as compiler representations for accelerating these loop constructs, are discussed. Performance results are presented for a few select examples.An earlier version of this paper was presented at Supercomputing '90.     

Fast finite difference Poisson solvers on heterogeneous architectures

In this paper we propose and evaluate a set of new strategies for the solution of three dimensional separable elliptic problems on CPU–GPU platforms. The numerical solution of the system of linear equations arising when discretizing those operators often represents the most time consuming part of larger simulation codes tackling a variety of physical situations. Incompressible fluid flows, electromagnetic problems, heat transfer and solid mechanic simulations are just a few examples of application areas that require efficient solution strategies for this class of problems. GPU computing has emerged as an attractive alternative to conventional CPUs for many scientific applications. High speedups over CPU implementations have been reported and this trend is expected to continue in the future with improved programming support and tighter CPU–GPU integration. These speedups by no means imply that CPU performance is no longer critical. The conventional CPU-control–GPU-compute pattern used in many applications wastes much of CPU’s computational power. Our proposed parallel implementation of a classical cyclic reduction algorithm to tackle the large linear systems arising from the discretized form of the elliptic problem at hand, schedules computing on both the GPU and the CPUs in a cooperative way. The experimental result demonstrates the effectiveness of this approach.     

Semantic-based QoS management in cloud systems: Current status and future challenges

Cloud Computing and Service Oriented Architectures have seen a dramatic increase of the amount of applications, services, management platforms, data, etc. gaining momentum for the necessity of new complex methods and techniques to deal with the vast heterogeneity of data sources or services. In this sense Quality of Service (QoS) seeks for providing an intelligent environment of self-management components based on domain knowledge in which cloud components can be optimized easing the transition to an advanced governance environment. On the other hand, semantics and ontologies have emerged to afford a common and standard data model that eases the interoperability, integration and monitoring of knowledge-based systems. Taking into account the necessity of an interoperable and intelligent system to manage QoS in cloud-based systems and the emerging application of semantics in different domains, this paper reviews the main approaches for semantic-based QoS management as well as the principal methods, techniques and standards for processing and exploiting diverse data providing advanced real-time monitoring services. A semantic-based framework for QoS management is also outlined taking advantage of semantic technologies and distributed datastream processing techniques. Finally a discussion of existing efforts and challenges is also provided to suggest future directions.     

A framework for simulating real-time multi-agent systems

In this paper, we describe an implementation of use in demonstrating the effectiveness of architectures for real-time multi-agent systems. The implementation provides a simulation of a simplified RoboCup Search and Rescue environment, with unexpected events, and includes a simulator for both a real-time operating system and a CPU. We present experimental evidence to demonstrate the benefit of the implementation in the context of a particular hybrid architecture for multi-agent systems that allows certain agents to remain fully autonomous, while others are fully controlled by a coordinating agent. In addition, we discuss the value of the implementation for testing any models for the construction of real-time multi-agent systems and include a comparison to related work.

Recognition of circular patterns on GPUs: Performance analysis and contributions

We develop a novel approach for computing the circle Hough transform entirely on graphics hardware (GPU). A primary role is assigned to vertex processors and the rasterizer, overshadowing the traditional foreground of pixel processors and enhancing parallel processing. Resources like the vertex cache or blending units are studied too, with our set of optimizations leading to extraordinary peak gain factors exceeding 358x over a typical CPU execution. Software optimizations, like the use of precomputed tables or gradient information and hardware improvements, like hyperthreading and multicores are explored on CPUs as well. Overall, the GPU exhibits better scalability and much greater parallel performance to become a solid alternative for computing the classical circle Hough transform versus those optimal methods run on emerging multicore architectures.