International Technology Roadmap for Semiconductors 2003の要求清浄度について － シリコンウエハ表面と雰囲気環境に要求される清浄度, 分析方法の現状について －

Datacenter workloads demand high computational capabilities, flexibility, power efficiency, and low cost It is challenging to improve all of these factors simultaneously To advance datacenter capabilities beyond what commodity server designs can provide, we designed and built a composable, reconfigurable hardware fabric based on field programmable gate arrays (FPGA) Each server in the fabric contains one FPGA, and all FPGAs within a 48-server rack are interconnected over a low-latency, high-bandwidth networkWe describe a medium-scale deployment of this fabric on a bed of 1632 servers, and measure its effectiveness in accelerating the ranking component of the Bing web search engine We describe the requirements and architecture of the system, detail the critical engineering challenges and solutions needed to make the system robust in the presence of failures, and measure the performance, power, and resilience of the system Under high load, the large-scale reconfigurable fabric improves the ranking throughput of each server by 95% at a desirable latency distribution or reduces tail latency by 29% at a fixed throughput In other words, the reconfigurable fabric enables the same throughput using only half the number of servers

/pdf/a-reconfigurable-fabric-for-accelerating-large-scale-4075xboyh9.pdf

A reconfigurable fabric for accelerating large-scale datacenter services

Datacenter workloads demand high computational capabilities, flexibility, power efficiency, and low cost. It is challenging to improve all of these factors simultaneously. To advance datacenter capabilities beyond what commodity server designs can provide, we have designed and built a composable, reconfigurablefabric to accelerate portions of large-scale software services. Each instantiation of the fabric consists of a 6x8 2-D torus of high-end Stratix V FPGAs embedded into a half-rack of 48 machines. One FPGA is placed into each server, accessible through PCIe, and wired directly to other FPGAs with pairs of 10 Gb SAS cablesIn this paper, we describe a medium-scale deployment of this fabric on a bed of 1,632 servers, and measure its efficacy in accelerating the Bing web search engine. We describe the requirements and architecture of the system, detail the critical engineering challenges and solutions needed to make the system robust in the presence of failures, and measure the performance, power, and resilience of the system when ranking candidate documents. Under high load, the largescale reconfigurable fabric improves the ranking throughput of each server by a factor of 95% for a fixed latency distribution--- or, while maintaining equivalent throughput, reduces the tail latency by 29%

/pdf/a-reconfigurable-fabric-for-accelerating-large-scale-1gqzfuqd5z.pdf

Due to recent advances in digital technologies, and availability of credible data, an area of artificial intelligence, deep learning, has emerged and has demonstrated its ability and effectiveness in solving complex learning problems not possible before. In particular, convolutional neural networks (CNNs) have demonstrated their effectiveness in the image detection and recognition applications. However, they require intensive CPU operations and memory bandwidth that make general CPUs fail to achieve the desired performance levels. Consequently, hardware accelerators that use application-specific integrated circuits, field-programmable gate arrays (FPGAs), and graphic processing units have been employed to improve the throughput of CNNs. More precisely, FPGAs have been recently adopted for accelerating the implementation of deep learning networks due to their ability to maximize parallelism and their energy efficiency. In this paper, we review the recent existing techniques for accelerating deep learning networks on FPGAs. We highlight the key features employed by the various techniques for improving the acceleration performance. In addition, we provide recommendations for enhancing the utilization of FPGAs for CNNs acceleration. The techniques investigated in this paper represent the recent trends in the FPGA-based accelerators of deep learning networks. Thus, this paper is expected to direct the future advances on efficient hardware accelerators and to be useful for deep learning researchers.

FPGA-Based Accelerators of Deep Learning Networks for Learning and Classification: A Review

The past few years a lot of effort has been put in the development and fabrication of III-V semiconductor waveguiding devices with monolithic integrated mode size converters (tapers). By integrating a taper with a waveguide device, the coupling losses and the packaging cost of OEICs in future fiber-optical networks can be much reduced. This paper gives an overview of different taper designs, the possible fabrication technologies and performances of tapered devices.

A review on fabrication technologies for the monolithic integration of tapers with III-V semiconductor devices

High-Performance Computing using FPGA covers the area of high performance reconfigurable computing (HPRC). This book provides an overview of architectures, tools and applications for High-Performance Reconfigurable Computing (HPRC). FPGAs offer very high I/O bandwidth and fine-grained, custom and flexible parallelism and with the ever-increasing computational needs coupled with the frequency/power wall, the increasing maturity and capabilities of FPGAs, and the advent of multicore processors which has caused the acceptance of parallel computational models. The Part on architectures will introduce different FPGA-based HPC platforms: attached co-processor HPRC architectures such as the CHRECs Novo-G and EPCCs Maxwell systems; tightly coupled HRPC architectures, e.g. the Convey hybrid-core computer; reconfigurably networked HPRC architectures, e.g. the QPACE system, and standalone HPRC architectures such as EPFLs CONFETTI system. The Part on Tools will focus on high-level programming approaches for HPRC, with chapters on C-to-Gate tools (such as Impulse-C, AutoESL, Handel-C, MORA-C++); Graphical tools (MATLAB-Simulink, NI LabVIEW); Domain-specific languages, languages for heterogeneous computing(for example OpenCL, Microsofts Kiwi and Alchemy projects). The part on Applications will present case from several application domains where HPRC has been used successfully, such as Bioinformatics and Computational Biology; Financial Computing; Stencil computations; Information retrieval; Lattice QCD; Astrophysics simulations; Weather and climate modeling.

High-Performance Computing Using FPGAs

With a growing interest in Patent Information Retrieval, there is a need to better understand the context associated with patent users, their tasks, needs and expectations of patent search systems and applications. Patent search is known to be a complex, difficult and challenging activity, usually requiring expert Patent Information Specialists to spend a substantial amount of time sourcing (or not) documents relevant to their particular task. Information Retrieval provides a whole array of possible techniques and tools which could be applied to ease the burden of such retrieval tasks, and also make searching patents more accessible to non-Patent Information Specialists. In this paper, we report the findings from a survey of patent users conducted to ascertain information about patent users and their search requirements with respect to Information Retrieval systems and applications.

/pdf/a-survey-of-patent-users-an-analysis-of-tasks-behavior-4bdcn6nfnn.pdf

A survey of patent users: an analysis of tasks, behavior, search functionality and system requirements

Networks on chip (NoC) emerged as a promising alternative to bus-based interconnect networks to handle the increasing communication requirements of the large systems on chip. Employing an appropriate topology for a NoC is of high importance mainly because it typically trade-offs between cross-cutting concerns such as performance and cost. The spidergon topology is a novel architecture which is proposed recently for NoC domain. The objective of the spidergon NoC has been addressing the need for a fixed and optimized topology to realize cost effective multi-processor SoC (MPSoC) development [7]. In this paper we analyze the traffic behavior in the spidergon scheme and present an analytical evaluation of the average message latency in the architecture. We prove the validity of the analysis by comparing the model against the results produced by a discrete- event simulator.

/pdf/an-analytical-performance-model-for-the-spidergon-noc-1qyrv6wgtq.pdf

An Analytical Performance Model for the Spidergon NoC

Patent search tasks are difficult and challenging, often requiring expert patent analysts to spend hours, even days, sourcing relevant information. To aid them in this process, analysts use Information Retrieval systems and tools to cope with their retrieval tasks. With the growing interest in patent search, it is important to determine their requirements and expectations of the tools and systems that they employ. In this poster, we report a subset of the findings of a survey of patent analysts conducted to elicit their search requirements.

Search system requirements of patent analysts

This paper presents an architecture and implementation details for MORA, a novel coarse grained reconfigurable processor for accelerating media processing applications. The MORA architecture involves a 2-D array of several such processors, to deliver low cost, high throughput performance in media processing applications. A distinguishing feature of the MORA architecture is the co-design of hardware architecture and low-level programming language throughout the design cycle. The implementation details for the single MORA processor, and benchmark evaluation using a cycle accurate simulator are presented.

/pdf/mora-an-architecture-and-programming-model-for-a-resource-2mnretuevb.pdf

Wim Vanderbauwhede

Papers

High-Performance Computing Using FPGAs

A survey of patent users: an analysis of tasks, behavior, search functionality and system requirements

An Analytical Performance Model for the Spidergon NoC

Search system requirements of patent analysts

MORA - An Architecture and Programming Model for a Resource Efficient Coarse Grained Reconfigurable Processor