Inflatable systems have been attracting attention in the field of interaction design. Conventional tabletop-sized pneumatic systems tend to be complex because they require bulky and noisy equipment. Therefore, several...
详细信息
fpgas have found their way into data centers as accelerator cards, making reconfigurablecomputing more accessible for high-performance applications. At the same time, new high-level synthesis compilers like Xilinx Vi...
详细信息
ISBN:
(纸本)9781665420105
fpgas have found their way into data centers as accelerator cards, making reconfigurablecomputing more accessible for high-performance applications. At the same time, new high-level synthesis compilers like Xilinx Vitis and runtime libraries such as XRT attract software programmers into the reconfigurable domain. While software programmers are familiar with task-level and data-parallel programming, fpgas often require different types of parallelism. For example, data-driven parallelism is mandatory to obtain satisfactory hardware designs for pipelined dataflow architectures. However, software programmers are often not acquainted with dataflow architectures-resulting in poor hardware designs. In this work we present FLOWER, a comprehensive compiler infrastructure that provides automatic canonical transformations for high-level synthesis from a domain-specific library. This allows programmers to focus on algorithm implementations rather than low-level optimizations for dataflow architectures. We show that FLOWER allows to synthesize efficient implementations for high-performance streaming applications targeting Systemon-Chip and FPGA accelerator cards, in the context of image processing and computer vision.
The slowdown of Moore's law and the end of Dennard scaling created a demand for specialized accelerators, including Field Programmable Gate Arrays (fpgas), in cloud data centers. Al the same time, compute resource...
详细信息
ISBN:
(纸本)9781665400602
The slowdown of Moore's law and the end of Dennard scaling created a demand for specialized accelerators, including Field Programmable Gate Arrays (fpgas), in cloud data centers. Al the same time, compute resources are increasingly consumed via public and private clouds and traditional applications are modernized using scalable microservices and Function-as-a-Service (FaaS) offerings. Nonetheless, true FaaS based on fpgas or other accelerators is virtually absent from the offering catalogs of all major cloud providers. In addition, FPGA applications are typically coded in a monolithic fashion, due to device and vendor specific dependencies, which reduces the portability and usability of FPGA cloud offerings further. However, FPGA-based FaaS can improve execution efficiency and minimize (tail-) latencies while decreasing costs. We propose a novel system architecture, called Mantle, that uses disaggregated fpgas to enable scalable, usable, portable and efficient FaaS offerings for fpgas. Our experimental results demonstrate a significant reduction of end-to-end service provisioning lime to below 7 seconds and an increase in execution efficiency by a factor of 4 with negligible overhead.
We present the design and implementation of an FPGA-based accelerator for bioinformatics applications in this paper. To process large amounts of data generated by next generation sequencing (NGS) technologies, bioinfo...
We present the design and implementation of an FPGA-based accelerator for bioinformatics applications in this paper. To process large amounts of data generated by next generation sequencing (NGS) technologies, bioinformatics applications require high-performance computing capabilities. Because of their parallel processing capabilities and reconfigurable hardware, fpgas are a promising solution for accelerating bioinformatics applications. To achieve high throughput and low latency, the proposed accelerator architecture employs parallelism and pipelining. The FPGA accelerator's performance is compared to software implementations on a CPU. The results of the experiments show that the FPGA accelerator outperforms the CPU implementation.
Approximate arithmetic modules are attractive solutions to save power consumption in error-resilient applications. The full adder (FA), as the core computing module, contributes a major part of arithmetic modules in d...
详细信息
Artificial Intelligence (AI) has been used in applications to alleviate specific problems in academia and industry. For instance, in healthcare, where edge-based computing platforms are heavily used, when it comes to ...
详细信息
Antimony sesquisulfide Sb2S3, an emerging low-loss phase change material, has attracted great interest for its unique properties, enabling its huge potential applications in programmable integrated optics. A reconfigu...
详细信息
With the rapid advancement of deep learning technology and the Internet of Things (IoT), new applications are continuously emerging, such as image classification and object tracking, which require deployment on edge c...
详细信息
The computational parallelism and energy efficiency inherent in reconfigurable hardware architectures like finegrained Field-Programmable Gate Arrays (fpgas) and Coarse-Grained reconfigurable Arrays (CGRAs) have been ...
详细信息
ISBN:
(纸本)9781665437592
The computational parallelism and energy efficiency inherent in reconfigurable hardware architectures like finegrained Field-Programmable Gate Arrays (fpgas) and Coarse-Grained reconfigurable Arrays (CGRAs) have been a subject of research mostly for Multimedia applications for many years [1]. The said strengths of reconfigurable systems are also beneficial for other application domains, e.g. High-Performance computing (HPC), since single-core and multicore systems may soon hit scaling limits.
The design of computation-intensive and complicated applications is limited by hardware resources. To realize efficient utilization of circuit area with lightweight computational logics and execute optimized operation...
详细信息
ISBN:
(纸本)9781665459716
The design of computation-intensive and complicated applications is limited by hardware resources. To realize efficient utilization of circuit area with lightweight computational logics and execute optimized operation, we propose an architecture of reconfigurable stochastic computing (SC) unit. The proposed SC unit decodes mathematical formula and reconfigures SC circuits to improve design productivity. Compared to the deterministic implementation, the proposed architecture reduces the circuit area by 39.7%.
暂无评论