Advanced Subsystems Research Articles

Potential of a modern vector supercomputer for practical applications: performance evaluation of SX-ACE

Achieving a high sustained simulation performance is the most important concern in the HPC community. To this end, many kinds of HPC system architectures have been proposed, and the diversity of the HPC systems grows rapidly. Under this circumstance, a vector-parallel supercomputer SX-ACE has been designed to achieve a high sustained performance of memory-intensive applications by providing a high memory bandwidth commensurate with its high computational capability. This paper examines the potential of the modern vector-parallel supercomputer through the performance evaluation of SX-ACE using practical engineering and scientific applications. To improve the sustained simulation performances of practical applications, SX-ACE adopts an advanced memory subsystem with several new architectural features. This paper discusses how these features, such as MSHR, a large on-chip memory, and novel vector processing mechanisms, are beneficial to achieve a high sustained performance for large-scale engineering and scientific simulations. Evaluation results clearly indicate that the high sustained memory performance per core enables the modern vector supercomputer to achieve outstanding performances that are unreachable by simply increasing the number of fine-grain scalar processor cores. This paper also discusses the performance of the HPCG benchmark to evaluate the potentials of supercomputers with balanced memory and computational performance against heterogeneous and cutting-edge scalar parallel systems.

Addressing Complexity Aspects in Conceptual Ship Design: A Systems Engineering Approach

This research examines the handling complexity aspects of conceptual design. Contemporary consensus suggests vessel design must consider new market requirements such as greater emphasis on environmental performance, a larger degree of uncertainty in terms of contract horizon, and the need for reliability of multiple operations assessed during early stages. Consequently, the industry has experienced development on many levels of ship design, from advanced subsystems (e.g., a wide range of machinery congurations), to vessels with demanding operations (e.g., modern oshore support vessels), to incorporation of eet assessment in early stages. Designers face a number of new technologies - usually representing greater investment - to obtain improved energy eciency and exibility regarding multi-faceted, future scenarios in which the vessel must operate. This large number of options results in an increase in the amount of information that should be considered to understand important aspects of the ship during the conceptual phase. This thesis is based on a systems engineering perspective to approach these kinds of developments, especially recent theories combining complexity theory in engineering.This thesis reviews current methods and approaches that deal with conceptual ship design and its complexity aspects. Based on this review, three research questions are proposed. First, which general complex systems theory premises can be used to dene complexity in conceptual ship design? Second, what general principles for organizing and simplifying complexity t the conceptual ship design task? Third, what methods eciently handle primary complexity aspects during conceptual ship design?The results of this study are the identification of the general principle of handling complexity, based on decomposition and encapsulation, as a strategy to manage relevant information during conceptual design, and proposing a five-aspect taxonomy to characterize and classify complexity in conceptual ship design. The taxonomy categorizes five aspects of conceptual ship design. The structural aspect relates to arrangement and interrelationships of the physical parts in the ship. The behavioral aspect derives from form-function mapping. The external circumstances to which the ship is subjected are captured in the contextual aspect. Uncertainties in future scenarios and expected/unexpected changes over time relate to the temporal aspect. The perceptual aspect relates to how various stakeholders perceive the value they receive from a design through the operational life cycle of the vessel.A discussion of both traditional and novel techniques to handle each of the aspects is presented. Focus is given to methods able to handle the three extended aspects (i.e., contextual, temporal, and perceptual). The goal of the study is to designate ship design as a complex system problem, developing and improving methods capable of handling primary complexity aspects during the conceptual phase.The primary contribution is characterization of conceptual ship design as a complex systems engineering task. Decomposition and encapsulation is presented as a general principle to handle complexity during the conceptual phase of ship design. More importantly, it identifies the intelligent encapsulation allowed by the five-aspect taxonomy, with implementation and development of methods to handle each aspect. Structural and behavioral aspects are investigated, merging traditional and novel techniques. Epoch-era analysis and a ship design deployment problem are used to tackle contextual and temporal aspects. The perceptual aspect is discussed through complex value robustness, and integration and concurrent assessment of all five aspects is handled theoretically through the responsive systems comparison method.This thesis consists of two parts. The first contains an introductory chapter presenting the background, the research questions, state-of-the-art conceptual ship design, ship as a complex system, information growth in ship design and complexity in a systems engineering framework, the research approach, a timeline of the research, initial results of a study of complexity aspects, results relevant to answering the three research questions, discussion of contributions, concluding remarks, and future research. The second part contains the five papers, in which individual results and contributions are discussed in more detail.

A software radio testbed for two-transmitter two-receiver space-time coding OFDM wireless LAN

A real-time testbed based on the technology of software radio is adopted to efficiently evaluate cutting-edge technologies in wireless communications, and thus becomes a valuable tool for both academic research and system prototyping. In this article we describe a software radio testbed, established in the software radio laboratory at Georgia Institute of Technology, used to implement a physical layer similar to IEEE 802.11a space-time coded orthogonal frequency-division multiplexing. The testbed consists of a 2 x 2 multiple-input multiple-output configuration with powerful digital signal processor chains, high-speed data exchange interfaces, and many advanced subsystem modules, such as high-speed high-resolution analog-to-digital and digital-to-analog converters, digital up/downconverters, and wideband RF transmit and receive front-ends with synchronous channels and programmable settings. The design methodology and implementation for key algorithms, such as time, sampling, and frequency synchronization, and channel estimation and compensation are discussed. The experimental data obtained from a typical indoor environment demonstrates that the prototype is capable of providing 30 Mb/s peak data rate, operating at the central frequency of 2.435 GHz with a spectral occupancy of 6.25 MHz. Spatial-temporal diversity gain associated with space-time coding is verified by the experimental results.

Advanced Abrams Crew Station Design: A Human Factors Perspective

In 1994, the U.S. Army sponsored the design of a technologically advanced combat tank, the Advanced Abrams Crewstation (AACS). The goals of the AACS project were: (1) design a three-man crewstation that outperformed the four-man M1A2 tank; (2) incorporate advanced technology subsystems chosen by the Army; and (3) meet all of the Army's requirements for information management that it deemed essential to the digital synchronization of the battlefield. While aircraft have traditionally incorporated state-of-the-art equipment in their designs, main battle tanks have emphasized ruggedness and durability over sophistication. In the past decade, the M1 and M1A2 Abrams tanks have begun to modernize their crewstations with digital, electronic command and control (C2) systems. Initial feedback from field tests of these systems suggested that the addition of an electronic display to the Abrams crewstation, and the accompanying increase in information provided to the tank commander, did not improve the crew's ability to do their jobs. The difficulty of interacting with these systems limited the potential advantage of the underlying technology, and it sparked the initiation of a new soldier-centered design effort. The authors were among the human factors members of a multi-disciplinary team contracted to design a crewstation for a three-man battle tank. This paper discusses the system design and development process and specific lessons learned during the project.

Study of an Advanced Fuel Handling System

To reduce the cost of fast breeder reactor plants, advanced subsystem concepts are studied. The material quantities of the fuel handling system constituent subsystems are estimated and those with the largest quantities are selected for improvement. The advanced subsystem concepts studied include in-vessel fuel storage while cooling, an in-vessel fuel transfer machine with variable-reach arm, and a half-swing hinged-track fuel transfer machine. A significant reduction of overall system material quantity is promised by a suitable combination of the proposed advanced concepts.

Conceptual design of the advanced technology platform

This paper presents the concept of the ATP (Advanced Technology Platform) which is a co-orbiting platform planned to be launched in 1995. The ATP has two major objectives. One is to establish the advanced technologies essential to the future space platforms. The other is to conduct autonomous and in-orbit serviceable space experiments. To meet the various mission demands and to attain the high performance, the ATP adopts modular design, in-orbit servicing capability and advanced subsystem technologies. As subsystem technologies for the ATP, rendezvous-docking, two phase fluid loop thermal control and retractable flexible solar array are under development. The mission profile of the ATP is divided into two phases. Many space experiments such as material processing are conducted after the technology demonstration mission.

An Advanced Subsystem Status Monitor Designed to Reduce Crew Workload

An Advanced Subsystem Status Monitor Designed to Reduce Crew Workload

Barium Tetratitanate MIC Technology

Experimental results on BaTi/sub 4/O/sub 9/ microwave integrated circuit (MIC) characteristics, including microstrip dispersion and loss, have shown excellent agreement with the theoretical predictions. Precision measurements of temperature stability were conducted at 14 GHz. The high dielectric constant (epsilon/sub r/ = 37) and its negative temperature coefficient can be used in specialized MIC's for application to advanced microwave subsystems.

Review of some microwave integrated circuit components utilizing microstrip techniques

This paper briefly reviews some of the techniques involved in realizing the practical microwave integrated circuit for systems applications. The main technological considerations and their application to advanced receiver sub-systems are discussed with emphasis on superheterodyne operation.

SHIPBOARD DATA MULTIPLEX SYSTEM (SDMS) — A NEW DESIGN FOR LOWER SHIP COST AND IMPROVED FLEXIBILITY

ABSTRACTThe Shipboard Data Multiplex System (SDMS) is a general purpose information transfer system directed toward fulfilling the internal data intercommunication requirements of a variety of naval combatant ships and submarines in the 1980 time frame. The need for a modem interior communication system of the size and capability of SDMS has been increased in unison with the sophistication of shipboard electronic equipments and the associated magnitude of equipment‐to‐equipment signal traffic.Instead of the miles of unique cabling that must be specifically designed for each ship, SDMS will meet information transfer needs with general‐purpose multiplex cable that will be installed according to a standard plan that does not vary with changes to the ship's electronics suite (Figure 1). With standard multiplex cabling throughout the ship and standard SDMS interfaces to ship subsystems, adding new equipment will be almost as simple as plugging a TV into standard house wiring outlets. However, the greatest impact of SDMS will be the decoupling of ship subsystems from each other and from the ship. Standard multiplex interfaces will avoid the cost and delay of modifying subsystems to make them compatible. The ability to wire a new ship according to a standard multiplex cable plan, long before the ship subsystems are fully defined, frees both the ship and the subsystems to develop at their own pace, will allow compression of the development schedules and provide ships with more advanced subsystems.