FPGA-based Implementation of Sliding-Window Aggregates over Disordered Data Streams

スポンサーリンク

概要

• 論文の詳細を見る

• This paper proposes an order-agnostic implementation of sliding-window aggregate queries on an FPGA. Instead of buffering and reordering input tuples, the proposed approach can handle bounded disorder by using punctuations. This significantly reduces the input-to-output latency as all tuples can be processed in arrival order. The proposed approach is used to implement an example query on an FPGA, and it is demonstrated that the proposed implementation can achieve over 150M tuples/sec processing performance with low latency.

• 2013-01-09

著者

• Yoshinaga Tsutomu

  University Of Electro-communications

• Irie Hidetsugu

  Graduate School Of Information Science And Technology The University Of Tokyo

• KAWASHIMA Hideyuki

  University of Tsukuba

• OGE Yasin

  Graduate School of Information Systems, The University of Electro-Communications, Japan

• MIYOSHI Takefumi

  e-trees.Japan, Inc

• YOSHIMI Masato

  Graduate School of Information Systems, The University of Electro-Communications, Japan

• YOSHINAGA Tsutomu

  University of Tsukuba, Japan

• KAWASHIMA Hideyuki

  University of Tsukuba, Japan

関連論文

• Pipelined round-robin broadcast algorithm in homogeneous clusters of SMP (計算機アーキテクチャ・ハイパフォーマンスコンピューティング 「ハイパフォーマンスコンピューティングとアーキテクチャの評価」に関する北海道ワークショップ(HOKKE-2008))
• Computer aided detection system implementation for recognize cancer in mammograms over an FPGA (VLSI設計技術)
• Computer aided detection system implementation for recognize cancer in mammograms over an FPGA (コンピュータシステム)
• Computer aided detection system implementation for recognize cancer in mammograms over an FPGA (リコンフィギャラブルシステム)
• Pipelined Round-Robin Broadcast Algorithm in Homogeneous Clusters of SMP
• D-6-8 Hybrid Compiler-Controlled Self-Adjustable Parallelism-Independent Scheduling Algorithm for Cluster of Workstations
• Ultra Dependable Processor
• Computation-Communication Overlap of Linpack on a GPU-Accelerated PC Cluster
• Design and Implementation of a Handshake Join Architecture on FPGA
• Using Cacheline Reuse Characteristics for Prefetcher Throttling
• FPGA-based Implementation of Sliding-Window Aggregates over Disordered Data Streams
• FPGA-based Implementation of Sliding-Window Aggregates over Disordered Data Streams
• FPGA-based Implementation of Sliding-Window Aggregates over Disordered Data Streams

もっと見る 閉じる

スポンサーリンク