Topic
Systems architecture
About: Systems architecture is a research topic. Over the lifetime, 17612 publications have been published within this topic receiving 283719 citations. The topic is also known as: system architecture.
Papers published on a yearly basis
Papers
More filters
••
TL;DR: The architecture of the open object-oriented database (OODB) management system, its requirements, and its computational model, which builds database functionality as an extensible collection of transparent extensions to existing programming languages, are described.
Abstract: The architecture of the open object-oriented database (OODB) management system, its requirements, and its computational model, which builds database functionality as an extensible collection of transparent extensions to existing programming languages, are described. The way in which open OODB's system architecture is decomposed into a kernal meta-architecture and a collection of modules implementing specific behavioral extensions is discussed. The potential risks of the open OODB architecture are reviewed. >
102 citations
••
TL;DR: In this article, the authors present and evaluate a framework aimed at finding the most appropriate deployment architecture for a distributed software system with respect to multiple, possibly conflicting QoS dimensions, and provide a set of tailorable algorithms for improving a system's deployment.
Abstract: A distributed system's allocation of software components to hardware nodes (i.e., deployment architecture) can have a significant impact on its quality of service (QoS). For a given system, there may be many deployment architectures that provide the same functionality, but with different levels of QoS. The parameters that influence the quality of a system's deployment architecture are often not known before the system's initial deployment and may change at runtime. This means that redeployment of the software system may be necessary to improve the system's QoS properties. This paper presents and evaluates a framework aimed at finding the most appropriate deployment architecture for a distributed software system with respect to multiple, possibly conflicting QoS dimensions. The framework supports formal modeling of the problem and provides a set of tailorable algorithms for improving a system's deployment. We have realized the framework on top of a visual deployment architecture modeling and analysis environment. The framework has been evaluated for precision and execution-time complexity on a large number of simulated distributed system scenarios, as well as in the context of two third-party families of distributed applications.
101 citations
•
22 Aug 1983TL;DR: The architecture is a declarative control language that allows one to write partial specifications of program behavior that facilitates incremental system development and the integration of disparate architectures like demons, object-oriented programming, and controlled deduction.
Abstract: One of the biggest problems in AT programming is the difficulty of specifying control. Meta-level architecture is a knowledge engineering approach to coping with this difficulty. The key feature of the architecture is a declarative control language that allows one to write partial specifications of program behavior. This flexibility facilitates incremental system development and the integration of disparate architectures like demons, object-oriented programming, and controlled deduction. This paper presents the language, describes an appropriate, and discusses the issues of compiling. It illustrates the architecture with a variety of examples and reports some experience in using the architecture in building expert systems.
101 citations
••
12 Oct 2019TL;DR: In this article, the authors present a vertically integrated hardware/software co-design, which includes a custom DIMM module enhanced with near-memory processing cores tailored for DL tensor operations.
Abstract: Recent studies from several hyperscalars pinpoint to embedding layers as the most memory-intensive deep learning (DL) algorithm being deployed in today's datacenters. This paper addresses the memory capacity and bandwidth challenges of embedding layers and the associated tensor operations. We present our vertically integrated hardware/software co-design, which includes a custom DIMM module enhanced with near-memory processing cores tailored for DL tensor operations. These custom DIMMs are populated inside a GPU-centric system interconnect as a remote memory pool, allowing GPUs to utilize for scalable memory bandwidth and capacity expansion. A prototype implementation of our proposal on real DL systems shows an average 6.2-17.6× performance improvement on state-of-the-art DNN-based recommender systems.
101 citations
••
01 Jul 1988
TL;DR: In this article, the authors propose an architecture for an expert system that can function as an operator's associate in the supervisory control of a complex dynamic system, called OFMspert (operator function model) expert system.
Abstract: The authors proposes an architecture for an expert system that can function as an operator's associate in the supervisory control of a complex dynamic system. Called OFMspert (operator function model (OFM) expert system), the architecture uses the operator function modeling methodology as the basis for the design. The authors put emphasis on the understanding capabilities, i.e. the intent referencing property, of an operator's associate. They define the generic structure of OFMspert, particularly those features that support intent inferencing. They also describe the implementation and validation of OFMspert in GT-MSOCC (Georgia Tech-Multisatellite Operations Control Center), a laboratory domain designed to support research in human-computer interaction and decision aiding in complex, dynamic systems. >
101 citations