Scalable System Architecture Research Articles

FCUDA-NoC: A Scalable and Efficient Network-on-Chip Implementation for the CUDA-to-FPGA Flow

High-level synthesis (HLS) of data-parallel input languages, such as the Compute Unified Device Architecture (CUDA), enables efficient description and implementation of independent computation cores. HLS tools can effectively translate the many threads of computation present in the parallel descriptions into independent, optimized cores. The generated hardware cores often heavily share input data and produce outputs independently. As the number of instantiated cores grows, the off-chip memory bandwidth may be insufficient to meet the demand. Hence, a scalable system architecture and a data-sharing mechanism become necessary for improving system performance. The network-on-chip (NoC) paradigm for intrachip communication has proved to be an efficient alternative to a hierarchical bus or crossbar interconnect, since it can reduce wire routing congestion, and has higher operating frequencies and better scalability for adding new nodes. In this paper, we present a customizable NoC architecture along with a directory-based data-sharing mechanism for an existing CUDA-to-FPGA (FCUDA) flow to enable scalability of our system and improve overall system performance. We build a fully automated FCUDA-NoC generator that takes in CUDA code and custom network parameters as inputs and produces synthesizable register transfer level (RTL) code for the entire NoC system. We implement the NoC system on a VC709 Xilinx evaluation board and evaluate our architecture with a set of benchmarks. The results demonstrate that our FCUDA-NoC design is scalable and efficient and we improve the system execution time by up to $63\times $ and reduce external memory reads by up to 81% compared with a single hardware core implementation.

The Square Kilometre Array Science Data Processor. Preliminary compute platform design

The Square Kilometre Array is a next-generation radio-telescope, to be built in South Africa and Western Australia. It is currently in its detailed design phase, with procurement and construction scheduled to start in 2017. The SKA Science Data Processor is the high-performance computing element of the instrument, responsible for producing science-ready data. This is a major IT project, with the Science Data Processor expected to challenge the computing state-of-the art even in 2020. In this paper we introduce the preliminary Science Data Processor design and the principles that guide the design process, as well as the constraints to the design. We introduce a highly scalable and flexible system architecture capable of handling the SDP workload.

New CT system architectures for high temporal resolution with applications to improved geometric dose efficiency and cardiac imaging.

A new scalable computed tomography (CT) system architecture is introduced with the potential to achieve much higher temporal resolution than is possible with current CT designs while maintaining the flux per imaged slice near today's levels. The concept relies only on known technologies; in particular, effective rotation speeds several times higher than what is possible today can be achieved leveraging today's x-ray tube designs and capabilities. The new CT architecture comprises the following elements: (1) decoupling of the source rotation from the detector rotation through the provision of two independent, coaxial and coplanar rotating gantries (drums), (2) observation of a source at a range of azimuthal angles with respect to a given detector cell, (3) utilization of a multiplicity of x-ray sources, (4) use of a wide-angle isocentered detector mounted on the independent detector drum, (5) the detector drum presents a wide angular aperture allowing x-rays from the various sources to pass through, with the active detector cells occupying about 240° in one configuration, and the wide aperture the complementary 120°, (6) antiscatter grids with absorbing lamellas oriented substantially parallel to the main gantry plane, and (7) optional sparse view acquisition in "bunches," a sparse sampling pattern potentially enabling further data-acquisition speedup for specific applications. Temporal resolution gains are achieved when K multiple sources are simultaneously in view of the extended detector. Accordingly, projection data relate to the sum of up to K line-integral terms; recovery of the individual line-integral estimates that form the input to the usual image reconstruction methods necessitates the inversion of a sparse linear system. When data for a tomographic slice are acquired during a full effective gantry rotation, the linear system is amenable to inversion; when high temporal resolution is sought, the system is underdetermined and a-priori information is useful in regularizing the problem. A regularization method is proposed whereby each sampling time interval is subdivided and individual projection data are acquired for each source during a subinterval. Other approaches involve spectral multiplexing. Thus, the use of an energy-discriminating detector such as a photon-counting detector will be advantageous to the proposed design. Recently developed volume-based scatter correction methods have the potential to apply to the proposed architectures. Mathematical modeling indicates acquisition of complete data for a given transaxial slice could be achieved in 50 ms or less, while delivering an x-ray exposure commensurate with that delivered by a system acquiring complete data in 200 ms. Applications include cardiac CT and the design of a CT system with nearly 100% geometric dose efficiency, whereby the effective rotation speed enables the use of a relatively narrow z-aperture detector, with no antiscatter grids. This represents a 33% dose reduction versus a system with 75% geometric dose efficiency. A new, scalable CT system architecture has been described that can potentially lead to large increases in temporal resolution. Potential applications include cardiac CT and the design of a system with 100% dose geometric dose efficiency. Future investigations will address the feasibility of the proposed approach.

A flexible and scalable component-based system architecture for video surveillance as a service, running on infrastructure as a service

There are many proposals for moving traditional video surveillance systems into the cloud, commonly known as Video Surveillance as a Service (VSaaS). Most systems use Hadoop technology for storing ...

An Efficient Scalable RNS Architecture for Large Dynamic Ranges

This paper proposes an efficient scalable Residue Number System (RNS) architecture supporting moduli sets with an arbitrary number of channels, allowing to achieve larger dynamic range and a higher level of parallelism. The proposed architecture allows the forward and reverse RNS conversion, by reusing the arithmetic channel units. The arithmetic operations supported at the channel level include addition, subtraction, and multiplication with accumulation capability. For the reverse conversion two algorithms are considered, one based on the Chinese Remainder Theorem and the other one on Mixed-Radix-Conversion, leading to implementations optimized for delay and required circuit area. With the proposed architecture a complete and compact RNS platform is achieved . Experimental results suggest gains of 17 % in the delay in the arithmetic operations, with an area reduction of 23 % regarding the RNS state of the art. When compared with a binary system the proposed architecture allows to perform the same computation 20 times faster alongside with only 10 % of the circuit area resources.

Achieving a Scalable E/E-Architecture Using AUTOSAR and Virtualization

ABSTRACTToday's automotive software integration is a static process. Hardware and software form a fixed package and thushinder the integration of new electric and electronic features once the specification has been completed. Usually softwarecomponents assigned to an ECU cannot be easily transferred to other devices after they have been deployed. The mainreasons are high system configuration and integration complexity, although shifting functions from one to another ECU isa feature which is generally supported by AUTOSAR. The concept of a Virtual Functional Bus allows a strict separationbetween applications and infrastructure and avoids source code modifications. But still further tooling is needed toreconfigure the AUTOSAR Basic Software (BSW). Other challenges for AUTOSAR are mixed integrity, versioning andmulti-core support.The upcoming BMW E/E-domain oriented architecture will require all these features to be scalable across all vehiclemodel ranges. Entire vehicle E/E clusters, with heterogeneous software environments, must be encapsulated and shiftedfrom one ECU to another while still meeting all postulated requirements. Virtualization technology could be a reasonablecompromise between freedom from interference among integrated systems, scalable system architecture and efficientruntime environment.This work gives a brief vision of domain oriented architecture within the automotive domain by means of AUTOSARtechnology and discusses the requirements for such a scalable system. It is discussed, whether the requirements can be metwith AUTOSAR or if additional abstraction or separation layers are needed. Finally we outline the possible usage of ahypervisor in combination within the AUTOSAR Basic Software using multicore technology and propose a mixed systemarchitecture supporting the domain oriented architecture.CITATION: Reinhardt, D., Kaule, D. and Kucera , M., Achieving a Scalable E/E-Architecture Using AUTOSAR andVirtualization, SAE Int. J. Passeng. Cars – Electron. Electr. Syst. 6(2):2013, doi:10.4271/2013-01-1399.____________________________________

Configurable hardware integrate and fire neurons for sparse approximation

Sparse approximation is an important optimization problem in signal and image processing applications. A Hopfield-Network-like system of integrate and fire (IF) neurons is proposed as a solution, using the Locally Competitive Algorithm (LCA) to solve an overcomplete L1 sparse approximation problem. A scalable system architecture is described, including IF neurons with a nonlinear firing function, and current-based synapses to provide linear computation. A network of 18 neurons with 12 inputs is implemented on the RASP 2.9v chip, a Field Programmable Analog Array (FPAA) with directly programmable floating gate elements. Said system uses over 1400 floating gates, the largest system programmed on a FPAA to date. The circuit successfully reproduced the outputs of a digital optimization program, converging to within 4.8% RMS, and an objective cost only 1.7% higher on average. The active circuit consumed 559 μA of current at 2.4 V and converges on solutions in 25 μs, with measurement of the converged spike rate taking an additional 1 ms. Extrapolating the scaling trends to a N=1000 node system, the spiking LCA compares favorably with state-of-the-art digital solutions, and analog solutions using a non-spiking approach.

Low Power Sparse Approximation on Reconfigurable Analog Hardware

Compressed sensing is an important application in signal and image processing which requires solving nonlinear optimization problems. A Hopfield-Network-like analog system is proposed as a solution, using the locally competitive algorithm (LCA) to solve an overcomplete l1 sparse approximation problem. A scalable system architecture using sub-threshold currents is described, including vector matrix multipliers (VMMs) and a nonlinear thresholder. A 4 × 6 nonlinear system is implemented on the RASP 2.9v chip, a field programmable analog array with directly programmable floating gate elements, allowing highly accurate VMMs. The circuit successfully reproduced the outputs of a digital optimization program, converging to within 4.8% rms, and an objective value only 1.3% higher on average. The active circuit consumed 29 μA of current at 2.4 V, and converges on solutions in 240 μs. A smaller 2 × 3 system is also implemented. Extrapolating the scaling trends to a N=1000 node system, the analog LCA compares favorably with state-of-the-art digital solutions, using a small fraction of the power to arrive at solutions ten times faster. Finally, we provide simulations of large scale systems to show the behavior of the system scaled to nontrivial problem sizes.

A collaborative food safety service agent architecture with alerts and trust

With the recent advances in Internet and mobile technologies, there are increasing demands for ubiquitous access to food safety information for service integration and gathering first hand information. However, due to disparate food trading information among different food suppliers throughout the food supply chain such as food importers, food wholesalers, food retailers, it is still difficult for citizens to use them effectively during their marketplace shopping. To overcome this problem, we propose a Collaborative Food Safety Agent System (CFSAS) based on a scalable, flexible, and intelligent Multi-Agent Information System (MAIS) architecture for proactive aids and trust-based decision support on food purchasing to citizens. We formulate our MAIS architecture for CFSAS further with agent clusters based on a case study of the Center for Food Safety (CFS) in Hong Kong. Agent clusters may comprise several types of agents to achieve the goals involved in the major processes of a food safety mechanism. We show how agents help citizens better plan, understand, and specify their preferences collaboratively with the CFSAS. We further illustrate how this can be implemented with Web service technologies to integrate disparate food information resources along the food supply chain.

An Immersive VR System for Sports Education

The development of new technologies has undoubtedly promoted the advances of modern education, among which Virtual Reality (VR) technologies have made the education more visually accessible for students. However, classroom education has been the focus of VR applications whereas not much research has been done in promoting sports education using VR technologies. In this paper, an immersive VR system is designed and implemented to create a more intuitive and visual way of teaching tennis. A scalable system architecture is proposed in addition to the hardware setup layout, which can be used for various immersive interactive applications such as architecture walkthroughs, military training simulations, sports game simulations, interactive theaters, and telepresence exhibitions. Realistic interaction experience is achieved through accurate and robust hybrid tracking technology, while the virtual human opponent is animated in real time using shader-based skin deformation. Potential future extensions are also discussed to improve the teaching/learning experience.

A scalable multiagent system architecture for interactive applications

Interactive applications like crowd simulations need to properly render the virtual world while simulating the interaction of thousands of agents at the same time. The computational workload generated by these two tasks highly increases with the number of the simulated agents, requiring a scalable design of the multiagent system. In this paper, we present, in an unified manner, a distributed multiagent system architecture that can manage large crowds of autonomous agents at interactive rates while rendering multiple views of the virtual world being simulated. This architecture consists of a distributed multiagent system and a complementary distributed visualization subsystem. We also present a new scalability study of the distributed multiagent system architecture, as well as an application example. The scalability study shows that, since no system bottlenecks are present in the proposed multiagent system architecture, it can efficiently simulate population sizes of different orders of magnitude by simply adding more hardware, provided that the ratio of clients per server is appropriate for the computer platforms acting as servers. On the other hand, the application example shows that these improvements are achieved without affecting the interactivity of the simulation. Therefore, the scalability of the proposed multiagent system architecture is validated for interactive large-scale applications.

Guest editorial

Guest editorial

Evaluation of scalable multi-agent system architectures for searching the Semantic Web

The Semantic Web (SW) is an open environment where heterogeneous and distributed knowledge is rendered machine-processable using ontologies, their extensions, and annotation metadata. The volume of resources available for semantic annotation indicates a requirement for scalable and robust mechanisms to search for resources using the encoded knowledge. SERSE is a multi-agent system to index, query, retrieve and aggregate Semantic Web resources. SERSE agents apply ontological knowledge to maintain a distributed semantic index, that is inter-connected by a semantic overlay network. Efficient retrieval of resources is achieved by the semantic routing mechanism, that locates those agents indexing resources matching a query. The paper presents the routing mechanism and an analytical evaluation of some of the autonomic features exhibited by the agents in order to ensure scalability.

MULTISENSORY DETECTION SYSTEM FOR DAMAGE CONTROL AND SITUATIONAL AWARENESS

A data fusion-based, multisensory detection system, called “Volume Sensor”, was developed under the Advanced Damage Countermeasures (ADC) portion of the US Navy's Future Naval Capabilities program (FNC) to meet reduced manning goals. A diverse group of sensing modalities was chosen to provide an automated damage control monitoring capability that could be constructed at a relatively low cost and also easily integrated into existing ship infrastructure. Volume Sensor employs an efficient, scalable, and adaptable design framework that can serve as a template for heterogeneous sensor network integration for situational awareness. In the development of Volume Sensor, a number of challenges were addressed and met with solutions that are applicable to heterogeneous sensor networks of any type. These solutions include: 1) a uniform, but general format for encapsulating sensor data, 2) a communications protocol for the transfer of sensor data and command and control of networked sensor systems, 3) the development of event specific data fusion algorithms, and 4) the design and implementation of modular and scalable system architecture. In full-scale testing on a shipboard environment, two prototype Volume Sensor systems demonstrated the capability to provide highly accurate and timely situational awareness regarding damage control events while simultaneously imparting a negligible footprint on the ship's 100 Mbps Ethernet network and maintaining smooth and reliable operation in a real-time fashion.

Towards ubiquitous tourist service coordination and process integration: A collaborative travel agent system architecture with semantic web services

With the recent advances in Internet and mobile technologies, there are increasing demands for ubiquitous access to tourist information systems for service coordination and process integration. However, due to disparate tourist information and service resources such as airlines, hotels, tour operators, it is still difficult for tourists to use them effectively during their trips or even in the planning stage. Neither can current tourist portals assist tourists proactively. To overcome this problem, we propose a Collaborative Travel Agent System (CTAS) based on a scalable, flexible, and intelligent Multi-Agent Information System (MAIS) architecture for proactive aids to Internet and mobile users. We also employ Semantic Web technologies for effective organization of information resources and service processes. We formulate our MAIS architecture for CTAS further with agent clusters based on a case study of a large service-oriented travel agency. Agent clusters may comprise several types of agents to achieve the goals involved in the major processes of a tourist's trip. We show how agents can make use of ontology from the Semantic Web to help tourists better plan, understand, and specify their requirements collaboratively with the CTAS. We further illustrate how this can be successfully implemented with Web service technologies to integrate disparate Internet tourist resources. To conclude, we discuss and evaluate our approach from different stakeholders' perspectives.

Trend and Challenge on System-on-a-Chip Designs

The success of system-on-a-chip (SoC) hinges upon a well-concerted integrated approach from multiple disciplines, such as device, design, and application. From the device perspective, rapidly improving VLSI technology allows the integration of billions of transistors on a single chip, thus permitting a wide range of functions to be combined on one chip. From the application perspective, numerous killer applications have been identified, which can make full use of the aforementioned functionalities provided by a single chip. From the design perspective, however, with greater device integration, system designs become more complex and are increasingly challenging to design. Moving forward, novel approaches will be needed to meet these challenges. This paper explores several new design strategies, which represent the current design trends to deal with the emerging issues. For example, recognizing the stringent requirements on power consumption, memory bandwidth/latency, and transistor variability, novel power/thermal management, multi-processor SoC, reconfigurable logic, and design for verification and testing have now been incorporated into modern system design. In addition, we look into some plausible solutions. For example, further innovations on scalable, reusable, and reliable system architectures, IP deployment and integration, on-chip interconnects, and memory hierarchies are all anticipated in the near future.

A generic theoretical framework for modeling gossip-based algorithms

During the past 30 years of the Internet revolution, the Internet has become a major force of change with an enormous effect on civilization. Consequently, computer networks have evolved into more complex system and become virtually ubiquitous. This in turn, has given raise to a growing demand for scalable and reliable computer system architectures. Thus far, there has been enormous effort by the research community to introduce decentralized, simple, and scalable distributed systems to solve a wide range of problems. In this paper we explore one promising solution, which was initially inspired by mathematical models that investigate two everyday life phenomena, epidemics and gossip, which we used interchangeably throughout the paper. During the last century, mathematicians developed models to predict the rate of diseases spread, namely epidemics, using differential equations. In addition, researches developed discrete mathematics models to predict what we already know; rumors spread fast, namely gossip. It was thus natural to harness these models in order to design distributed systems that mimic the basic behavior of such fast spreading everyday life paradigms.

Microfluidics-Based Biochips: Technology Issues, Implementation Platforms, and Design-Automation Challenges

Microfluidics-based biochips are soon expected to revolutionize clinical diagnosis, deoxyribonucleic acid (DNA) sequencing, and other laboratory procedures involving molecular biology. In contrast to continuous-flow systems that rely on permanently etched microchannels, micropumps, and microvalves, digital microfluidics offers a scalable system architecture and dynamic reconfigurability; groups of unit cells in a microfluidics array can be reconfigured to change their functionality during the concurrent execution of a set of bioassays. As more bioassays are executed concurrently on a biochip, system integration and design complexity are expected to increase dramatically. This paper presents an overview of an integrated system-level design methodology that attempts to address key issues in the synthesis, testing and reconfiguration of digital microfluidics-based biochips. Different actuation mechanisms for microfluidics-based biochips, and associated design-automation trends and challenges are also discussed. The proposed top-down design-automation approach is expected to relieve biochip users from the burden of manual optimization of bioassays, time-consuming hardware design, and costly testing and maintenance procedures, and it will facilitate the integration of fluidic components with a microelectronic component in next-generation systems-on-chips (SOCs).

Scalable monitoring support for resource management and service assurance

Continuous monitoring of network status and its resources are necessary to ensure proper network operation. Deployment of QoS-based value-added services in IP networks necessitates the employment of resource management techniques and specifically the use of traffic engineering. The latter typically relies on monitoring data for both offline proactive and dynamic reactive solutions. The variety of data to be collected and analyzed using different measurement methods and tools, and the extent of monitoring information to use demand a proper QoS monitoring infrastructure. A monitoring system should be scalable in terms of network size, speed, and number of customers subscribed to value-added services. This article investigates the requirements of scalable monitoring system architectures, proposes principles for designing such systems, and validates them through the design and implementation of a scalable monitoring system for QoS delivery in IP differentiated services networks. Experimental assessment results prove the accuracy and scalability of the proposed monitoring system.

Building Quality-of-Service Monitoring Systems for Traffic Engineering and Service Management

Deployment of quality-of-service (QoS) based value-added services in IP networks necessitates the use of traffic engineering. Traffic engineering allows service providers to use the network resources efficiently, according to the different quality levels associated with the range of services they offer. Traffic engineering relies typically on monitoring data for both “offline proactive” and “dynamic reactive” approaches. Monitoring data may be used for network provisioning, dynamic resource allocation, route management, and in-service performance verification for value-added IP services. A monitoring system should scale with the network size, the network speed, and the number of customers subscribed to use value-added IP services. This paper investigates the requirements of scalable monitoring system architectures, proposes principles for designing such systems and validates these principles through the design and implementation of a scalable monitoring system for traffic engineering and QoS delivery in IP Differentiated Services networks. Methods for assessing the relative merits of such monitoring systems are proposed. Experimental assessment results prove the scalability, accuracy, and also demonstrate the benefits of the proposed monitoring system.