Tightly Coupled Accelerators アーキテクチャに基づく GPU クラスタの構築と性能予備評価

塙敏博, 児玉祐悦, 朴泰祐, 佐藤三久 - 情報処理学会論文誌 …, 2013 - ipsj.ixsq.nii.ac.jp
論文抄録 GPU などの演算加速装置を用いたクラスタが HPC システム向けに広く使われている.
しかしこのようなクラスタでは, ノード間をまたがる演算加速装置間の通信において, CPU …

Tightly coupled accelerators architecture for minimizing communication latency among accelerators

T Hanawa, Y Kodama, T Boku… - 2013 IEEE International …, 2013 - ieeexplore.ieee.org
In recent years, heterogeneous clusters using accelerators have been widely used in high
performance computing systems. In such clusters, inter-node communication among …

演算加速装置に基づく超並列クラスタ HA-PACS による大規模計算科学

朴泰祐, 佐藤三久, 塙敏博, 児玉祐悦… - 研究報告 …, 2011 - ipsj.ixsq.nii.ac.jp
論文抄録 筑波大学計算科学研究センターでは 2012 年 1 月の運用開始を目指し, 大規模 GPU
クラスタ HA-PACS の導入を進めている. HA-PACS は標準的技術による大規模 GPU …

A hybrid adaptation protocol for TCP-friendly layered multicast and its optimal rate allocation

J Liu, B Li, YQ Zhang - … Twenty-First Annual Joint Conference of …, 2002 - ieeexplore.ieee.org
Layered transmission has been proposed as a solution to video multicast over the Internet.
Existing protocols usually perform adaptation at the receiver and use static rate allocation …

Tightly Coupled Accelerators アーキテクチャのための通信機構

塙敏博, 児玉祐悦, 朴泰祐, 佐藤三久 - 研究報告計算機 …, 2012 - ipsj.ixsq.nii.ac.jp
論文抄録 筑波大学計算科学研究センターでは, 大規模 GPU クラスタである HA-PACS を導入し,
2012 年 2 月から運用を開始している. 素粒子・宇宙物理・生命科学等の, 極めて大量な演算を要求 …

Design and implementation of an alternate system interconnect based on PCI express

YW Kim, Y Ren, WH Choi - Journal of the Institute of Electronics …, 2015 - koreascience.kr
PCI Express is a well-known and widely used de-facto system bus standard for connecting
among a processor and IO devices. PCI Express is originated from old PCI standard, and its …

Optimising security, power consumption and performance of embedded systems

P Rahimi - 2024 - repository.essex.ac.uk
Increased interest in multicore systems has led to significant advancements in computing
power, but it has also introduced new security risks due to covert channel communication …

Pearl: Power-aware, dependable, and high-performance communication link using pci express

T Hanawa, T Boku, S Miura, M Sato… - 2010 IEEE/ACM Int'l …, 2010 - ieeexplore.ieee.org
We have proposed a power-aware, high-performance, dependable communication link
using PCI Express as a direct communication device, referred to as PEARL, for application …

Pearl and peach: A novel pci express direct link and its implementation

T Hanawa, T Boku, S Miura, M Sato… - … on Parallel and …, 2011 - ieeexplore.ieee.org
We have proposed PEARL, which is a power-aware, high-performance, dependable
communication link using PCI Express as a direct communication device, for application in a …

An 80 Gbps dependable multicore communication SoC with PCI express I/F and intelligent interrupt controller

S Otani, H Kondo, I Nonomura, A Ikeya… - 2011 IEEE Cool …, 2011 - ieeexplore.ieee.org
An 80 Gbps dependable communication SoC with four 4X PCIe Rev. 2.0 ports has been
developed that acts as a communication link with high transfer capability. By using the PCIe …