reconfigurable computing has the potential for providing significant performance increases to a number of computing applications. However, realizing these benefits has traditionally required extensive digital design e...
详细信息
ISBN:
(纸本)1601320647
reconfigurable computing has the potential for providing significant performance increases to a number of computing applications. However, realizing these benefits has traditionally required extensive digital design experience and knowledge of hardware description languages (HDL). While a number of academic and commercial products have focused on translation of highlevel languages (HLL) to HDL, the tools do not create optimized digital designs that are competitive with handcoded solutions for all instances of input HLL code. In this paper, we describe an optimization in the C-to-HDL transformation that automatically reorganizes operations between pipeline stages in order to reduce critical path lengths. This optimization results in significant improvement in design frequency and throughput, especially for algorithms that cannot be pipelined effectively. The effects of this optimization are examined on SHA-1 and Smith-Waterman algorithms. Results show that the automatically generated hardware implementations perform comparably with hand-coded implementations.
For two decades, reconfigurable computing systems have provided an attractive alternative to fixed hardware solutions. reconfigurable computing systems have demonstrated the low cost and flexibility of a software solu...
详细信息
ISBN:
(纸本)1601320647
For two decades, reconfigurable computing systems have provided an attractive alternative to fixed hardware solutions. reconfigurable computing systems have demonstrated the low cost and flexibility of a software solution combined with the high performance of fixed hardware. For a variety of practical reasons, much of the work in this area focused on commercial FPGA devices as the underlying hardware platform. Recently, several new designs have diverged from the bit-level, circuit-oriented architectures of FPGAs and produced a variety of architectures more suitable for computation and high level language programming. These new highly parallel architectures contain a relatively large number of programmable cores, each approaching the complexity of a traditional microprocessor. Today such devices can be found in popular consumer electronics including game consoles and desktop PC graphics controllers as well as a new generation of supercomputers. These new devices, often described using the generic term 'multicore' represent the latest phase in the evolution of reconfigurablesystems. Like earlier reconfigurablesystems they promise very high performance at relatively low power with high levels of programmability. These new systems also feature software development tools geared more toward traditional high level language programming than the hardware design orientation found in earlier generations of reconfigurablesystems.
Projected computational requirements for future space missions are outpacing technologies and trends in conventional embedded microprocessors. In order to meet the necessary levels of performance, new computing techno...
详细信息
ISBN:
(纸本)1601320647
Projected computational requirements for future space missions are outpacing technologies and trends in conventional embedded microprocessors. In order to meet the necessary levels of performance, new computing technologies are of increasing interest for space systems, such as reconfigurable devices and vector processing extensions. These new technologies can also be used in tandem with conventional generalpurpose processors in the form of multiparadigm computing. By using FPGA resources and AltiVec extensions, as well as MPI extensions for multiprocessor support, we explore possible hardware/software designs for a synthetic aperture radar application. Design of key components of the SAR application including range compression and azimuth compression will be discussed, and hardware/software performance tradeoffs analyzed. The performance of these key components will be measured individually, as well as in the context of the entire application. Fault-tolerant versions of range and azimuth compression algorithms are proposed and their performance overhead is evaluated. Our analysis compares several possible multiparadigm systems, achieving up to 18 speedup while also adding fault tolerance to a pre-existing SAR application.
We present the XF-BOARD, a prototyping platform for reconfigurable hardware operating system research. The platform is based on two tightly coupled FPGAs. The first (C-FPGA) implements a soft CPU core that controls th...
详细信息
ISBN:
(纸本)1932415424
We present the XF-BOARD, a prototyping platform for reconfigurable hardware operating system research. The platform is based on two tightly coupled FPGAs. The first (C-FPGA) implements a soft CPU core that controls the overall system;the second (R-FPGA) is used as dynamically reconfigurable hardware resource. Furthermore, a rich set of I/O and memory devices are available for implementing multimedia and networking applications.
Coarse-grained dynamically reconfigurable processor arrays (DRPAs) have been received an attention as a flexible and efficient off-loading engine for various types of System-on-Chips (SoCs). Interconnection in these a...
详细信息
ISBN:
(纸本)1601320647
Coarse-grained dynamically reconfigurable processor arrays (DRPAs) have been received an attention as a flexible and efficient off-loading engine for various types of System-on-Chips (SoCs). Interconnection in these architectures is one of the important factors to be evaluated. MuCCRA-1, the first prototype of MuCCRA(Multi-Core Configurable reconfigurable Architecture) project, uses a typical island-style interconnection in its PE array. Although the island-style interconnection is flexible, the large delay time caused by passing multiple switches and long wires often degrades its clock frequency. In this paper, MuCCRA-D, a dynamically reconfigurable processor which uses direct interconnection between neighboring PEs, is designed and evaluated. The evaluation results show that the required semiconductor area for MuCCRA-D is 12% smaller than that of MuCCRA-1 by reducing wiring resource in the interconnection. Since higher clock frequency can be used, DCT, α-Blending, Bubble-Sort and SHA-1 implemented on the MuCCRA-D are 3.84 times faster than MuCCRA-1 at maximum.
This paper discusses the work done by the author and his coworkers in the area of self-checking and fault tolerant FPGA design. It also proposes an FPGA architecture that is composed of functional cells with built in ...
详细信息
ISBN:
(纸本)1601320647
This paper discusses the work done by the author and his coworkers in the area of self-checking and fault tolerant FPGA design. It also proposes an FPGA architecture that is composed of functional cells with built in error correction capability. A functional cell in the proposed architecture can be used to implement logic functions as well as to route signals to other functional cells. It is composed of three units: a logic block, a faulttolerant address generator and a director unit. The logic block uses a look-up table to implement logic functions. The fault-tolerant address generator corrects any single bit error in the incoming data to the functional cell. The director block can transmit output data from the logic block to another functional cell located at its South, North, East or West, or to cells in all four directions.
For the most part, reconfigurable architectures have been based on existing IC architectures or combinations of them. To date, they have lukewarm success at best. A fundamentally different thought process is needed to...
详细信息
ISBN:
(纸本)1601320647
For the most part, reconfigurable architectures have been based on existing IC architectures or combinations of them. To date, they have lukewarm success at best. A fundamentally different thought process is needed to unleash the value of adapting hardware to the problem at hand. One that delivers on the promise of super-computer performance on battery power at consumer price points to replace ASIC, SOC, FPGA and multi-processor solutions. Every engineer understands the definition of work as W=F*D Cos Θ. This is the definition used in physics. But, when asked what the definition of work is in electronic circuits you get a blank stare followed by, "it must have something to do with energy, MHz, bus width, etc." In other words, you do not receive a credible answer. This paper explores a few research concepts and opportunities in regards to a definition of work in electronics and its implication in the definition of reconfigurable architectures.
This paper presents a framework that improves the portability and ease-of-use issues of current reconfigurable Computers (RCs). These two drawbacks should be solved in order for RC to become a mainstream solution. Por...
详细信息
ISBN:
(纸本)1601320647
This paper presents a framework that improves the portability and ease-of-use issues of current reconfigurable Computers (RCs). These two drawbacks should be solved in order for RC to become a mainstream solution. Portability across platforms is difficult to achieve because RC systems have diverse hardware architectures and services. This lack of portability hinders reuse, and thus, ease-of-use. The framework proposed in this work is able to hide the architectural details of the systems, simplify the IP integration, and provide the portability across different RC platforms. User specifies IP requirements such as memory configuration, sequential or random access to the memory, or I/O registers using a graphical-user-interface (GUI) tool, which generates a hardware interface specification for the IP and the logic necessary to target the selected platform. The hardware interface remains the same regardless the targeted architecture. In addition, the tool generates a software library that includes services such as bitstream management and data exchange between microprocessor and IP. This framework has been demonstrated on two representative RCs: Cray XD1 and SGI RASC RC100.
One of the challenges of designing for coarse grain reconfigurable arrays is the need for mature tools. This is especially important because of the heterogeneity of the larger, more predefined (and hence more speciali...
详细信息
ISBN:
(纸本)1932415424
One of the challenges of designing for coarse grain reconfigurable arrays is the need for mature tools. This is especially important because of the heterogeneity of the larger, more predefined (and hence more specialized) array elements. This work describes the use of a genetic algorithm (GA) to automate the physical binding phase of kernel design. We identify the generalizable features of an example platform and discuss ways to harness the binding problem to a GA search engine.
Today's rapidly changing world has led travelers to seek customized experiences that align with their individual preferences and interests. With the continuous evolution of the digital realm, the expectations of u...
详细信息
ISBN:
(数字)9798331527495
ISBN:
(纸本)9798331527501
Today's rapidly changing world has led travelers to seek customized experiences that align with their individual preferences and interests. With the continuous evolution of the digital realm, the expectations of users interacting with online platforms for travel planning and reservations are also evolving. To cater to the demand for personalized recommendations and enhance the user experience, we propose the incorporation of a Content-Based Filtering algorithm into our tour website. This algorithm, in conjunction with a recommender system for Travel and tourism management, enables users to book tours nationwide through a single dynamic website that provides comprehensive information on various destinations and tour specifics. Developed with PHP as the front end and Microsoft SQL Server 2008 as the back end, the website is compatible with all browsers. Programming languages such as HTML, CSS, Bootstrap, and PHP were utilized in its creation. Administrators have the capability to add tour packages from selected travel agents and hotels by establishing a tour page. Subsequently, users can register and reserve each package, with confirmation managed by the admin in the booking management section. Users can easily view their confirmations in the “My Booking” page, making this platform user-friendly for all travelers seeking seamless bookings and detailed information. Furthermore, alongside implementing a Content-Based Filtering algorithm for personalized tour recommendations, we aim to promote the availability of accommo-dations in local rural areas. This endeavor not only presents travelers with cost-effective lodging choices but also contributes to the economic growth of the communities they visit.
暂无评论