OHNO Kazuhiko

J-GLOBAL         Last updated: Feb 29, 2012 at 12:32
 
Avatar
Name
OHNO Kazuhiko
Affiliation
Mie University
Section
Graduate School of Engineering, Systems Engineering, Computer Archtecture Laboratory
Job title
Lecturer
Degree
(BLANK)(Kyoto University)
Other affiliation
Mie University

Research Areas

 
 

Academic & Professional Experience

 
Apr 2006
 - 
2011
associate professor, Graduate School of Engineering, Mie University
 
Apr 2003
 - 
Mar 2006
associate professor, Department of Information Engineering, Faculty of Engineering, Mie University
 
Apr 1998
 - 
Mar 2003
assistant, Toyohashi University of Technology
 

Education

 
 
 - 
1998
Graduate School, Division of Engineering, Kyoto University
 
 
 - 
1993
Faculty of Engineering, Kyoto University
 

Published Papers

 
Masaki Matsumoto, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo, Hiroshi Nakashima
4(3) 55-67   Jun 2011
Task scheduling is very important for efficient execution of large-scale workflows in distributed computing environments. Static scheduling schemes achieve high performance when executing workflows in stable environments. However, the scheduling c...
NAKABAYASHI Tomoyuki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
The IEICE transactions on information and systems (Japanese edetion)   94(4) 646-656   Apr 2011
MATSUMOTO Masaki, KATANO Satoshi, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio, NAKASHIMA Hiroshi
The IEICE transactions on information and systems (Japanese edetion)   93(6) 693-704   Jun 2010
Masaki Matsumoto, Satoshi Katano, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo, Hiroshi Nakashima
2(1) 1-17   Jan 2009
We are developing a task parallel script language MegaScript for parallel processing on large-scale widely-distributed environments. To achieve high performance, MegaScript requires a scheduling scheme, which efficiently schedules dependent tasks ...
SAKAGUCHI YUSUKE, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
47(12) 296-307   Sep 2006
We are developing a task-parallel script language named MegaScript for mega-scale parallel processing. MegaScript regards existing sequential/parallel programs as tasks, and controls them for massively parallel execution. Although MegaScript runti...
MIYAMAE Motoki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
The IEICE transactions on information and systems (Japanese edetion)   89(3) 482-491   Mar 2006
MARUYAMA MASAO, YAMAMOTO SHIGEHIRO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
45(3) 109-121   Mar 2004
MATSUO HARUYUKI, IMAFUKU SHIGERU, OHNO KAZUHIKO, NAKASHIMA HIROSHI
44(1) 114-127   Jan 2003
This paper describes our distributed architectural simulator of shared memory multiprocessors named Shaman. The simulator runs on a PC cluster that consists of multiple front-end nodes to simulate the instruction level behavior of the target multi...
MATSUO HARUYUKI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ Journal   43(4) 855-865   Apr 2002
In order to reduce the overhead of synchronizing operations of shared memory multiprocessors, we have proposed a mechanism named specMEM to excute memory accesses following a synchronizing operation speculatively before the completion of the synch...
OHNO KAZUHIKO, YAMAMOTO SHIGEHIRO, OKANO TAKANORI, NAKASHIMA HIROSHI
42(12) 95-110   Nov 2001
Automatic parallelization is much difficult in non-numerical field, because irregular and dynamic data structures are frequently used. Therefore, many parallelizing libraries and parallel programming languages have been proposed. However, the effi...
IMAFUKU SHIGERU, OHNO KAZUHIKO, NAKASHIMA HIROSHI
42(9) 93-105   Aug 2001
This paper proposes a method to reduce the amount of the memory references generated by the front-end of our distributed execution-driven simulator for shared memory multiprocessors named Shaman. The simulator consists of the front-end to execute ...
TANAKA KOTARO, OHNO KAZUHIKO, NAKAHSIMA HIROSHI
IPSJ Journal   42(4) 732-744   Apr 2001
This paper proposes a new parallel wire routing algorithm named Amaterous. As many successful parallel wire routers, Amaterous has global and detailed routers, but the feedback from the detailed to the global is removed so that both routers are pa...
A Parallel Programming Language based on Declarative Process Network Models(In Japanese)
IPSJ Transaction on High Performance Systems   Vol.42, No.12 (HPS4), 95-110    2001
SATO TAKAYUKI, MATSUO HARUYUKI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
41(5) 1-14   Aug 2000
In order to reduce the overhead of synchronizing operations of shared memory multiprocessors, this paper proposes a mechanism, named specMEM, to execute memory accesses following a synchronizing operation speculatively before the completion of the...
Efficient Mesage Communication of Concurrent Logic Programming Language KLI Based on Static Analysis(In Japanese)
IPSJ Transaction   38(8) 1638-1648   1997
Kubota Atsushi, Miyoshi Ikuo, Ohno Kazuhiko, Mori Shin-ichiro, Nakashima Hiroshi, Tomita Shinji
IPSJ Journal   35(4) 532-541   Apr 1994
ln this paper, we focus on parallelizing compiler techniques which generate SPMD codes for distributed memory computers. Parallelization of loops with indirect accesses with index arrays causes irregular access patterns. For such codes, a techniqu...

Misc

 
Tomoyuki Nakabayashi, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo
IPSJ SIG Notes   2011(15) 1-8   Jul 2011
Recently, the increase of the energy consumption of mobile computers caused by performance enhancement becomes one serious problem. Therefore, a lot of research for low energy and high performance computing is carried out. In order to reduce energ...
Dai Michiura, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo
IPSJ SIG Notes   2011(17) 1-9   Jul 2011
This paper presents a compiler which automatically generates data transfer code for GPGPU. The user does not need to specify data transfer functions of CUDA, reducing the burden of GPGPU programming. The data transfer time is often large overhead ...
Akihiro Mita, Takayuki Naka, Masaki Matsumoto, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo
IPSJ SIG Notes   2011(57) 1-8   Jul 2011
We are developing a parallel script programming language MegaScript for large-scale workflows. MegaScript is an OOPL and each task and communication channel called stream is represented as an object. Although this feature enables flexible descript...
Masaki Matsumoto, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo
IPSJ SIG Notes   2011(60) 1-8   Jul 2011
Task scheduling is very important for efficient execution of large-scale workflows. However, scheduling large-scale workflows using existing scheduling schemes is not practical because of the huge computational costs. To solve this problem, we hav...
Kazuhiko Ohno, Akihiro Mita, Masaki Matsumoto, Takahiro Sasaki, Toshio Kondo, Hiroshi Nakashima
4(3)    Jun 2011
We are developing a parallel script programming language MegaScript for large-scale workflows. MegaScript is an OOPL and each task and communication channel called stream is represented as an object. Although this feature enables flexible descript...
NAKABAYASHI Tomoyuki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
Technical report of IEICE. VLD   110(432) 51-56   Feb 2011
DFFs play important roles in the CMOS circuits because the performance of DFFs significantly affects the area and the power consumption of VLSI chips. In this paper, two types of Semi-static TSPC (True-single-phase clocking) DFFs using Split-outpu...
Masaki Matsumoto, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo, Hiroshi Nakashima
IPSJ SIG Notes   2010(6) 1-9   Jul 2010
Task scheduling is very important for efficient execution of large-scale workflows on distributed computing environments. Various scheduling schemes are proposed, including dynamic rescheduling schemes for environments which performance change dyn...
Yukitoshi Shirota, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo
IPSJ SIG Notes   2010(12) 1-8   Jul 2010
Power dissipation is a major concern not only for mobile computing but also high performance computing, and achieving both low energy and high performance at the same time is required. It is paticularly important to reduce leakage energy consumed ...
Tomoyuki Nakabayashi, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo
IPSJ SIG Notes   2010(22) 1-8   Jul 2010
Recently, the increase of the energy consumption of mobile computers caused by performance enhancement becomes one serious problem. So many researches for low energy and high performance computing are performed. In order to reduce the energy consu...
Masaki Matsumoto, Satoshi Katano, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo, Hiroshi Nakashima
3(1)    Mar 2010
We are developing a task parallel script language MegaScript for parallel processing on large-scale widely-distributed environment. To achieve high performance, MegaScript requires a scheduling scheme that efficiently schedules dependent tasks to ...
Keigo Matsushita, Kazuya Taniguchi, Masaki Matsumoto, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo, Hiroshi Nakashima
3(1)    Mar 2010
In recent years, the demand for parallel processing is increasing. However, coding and debugging parallel programs are still large burden because of the inter-process communications and nondeterministic behaviors. Therefore, the parallel programmi...
MATSUMOTO Masaki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio, NAKASHIMA Hiroshi
72(5) "5-101"-"5-102"   Mar 2010
MATSUBARA Nobuyuki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IEICE technical report. Computer systems   109(319) 7-12   Nov 2009
Power dissipation is a major concern for mobile computing, and achieving both low energy and high performance at the same time is required. Especially, it is important to reduce leakage energy consumed in a cache memory. Our proposed approach call...
NOMURA Kazumasa, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IEICE technical report. Computer systems   109(319) 13-18   Nov 2009
Recently, specially in mobile processors, the increase of energy consumption by enhancing performance becomes serial problem. So achievement of both low energy and high-performance is demanded. Therefore various low energy techniques have been pro...
Tomoyuki Nakabayashi, Takahiro Sasaki, Kazuhiko Ohno, Toshio Kondo
IPSJ SIG Notes   2009(12) 1-8   Jul 2009
Recently, the increase of the energy consumption of mobile computers caused by performance enhancement becomes one serious problem. So many researches for low energy and high performance computing are performed. In order to reduce the energy consu...
Kazuya Taniguchi, Masaki Matsumoto, Kazuhiko Ohno, Takahiro Sasaki, Toshio Kondo, Hiroshi Nakashima
2(1)    Jan 2009
Large-scale parallel programming is quite difficult because conventional development environments do not have features enough to support from designing to debugging. Visual programming (VP) allows programming instinctively. However, dynamic code s...
SUGAYA Tomohiro, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
ITE Technical Report   32(56) 71-76   Dec 2008
The video coding standard H.264/AVC succeeds in providing both high quality and high compression efficiency with the widely used MPEG-2, but it requires large amount of computational power mainly consumed motion estimation. Many method has been pr...
SUGAYA Tomohiro, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IEICE technical report. Image engineering   108(344) 71-76   Dec 2008
The video coding standard H.264/AVC succeeds in providing both high quality and high compression efficiency with the widely used MPEG-2, but it requires large amount of computational power mainly consumed motion estimation. Many method has been pr...
SUGAYA Tomohiro, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IEICE technical report. Communication systems   108(343) 71-76   Dec 2008
The video coding standard H.264/AVC succeeds in providing both high quality and high compression efficiency with the widely used MPEG-2, but it requires large amount of computational power mainly consumed motion estimation. Many method has been pr...
SUGAYA Tomohiro, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IPSJ SIG Notes   2008(124) 71-76   Dec 2008
The video coding standard H.264/AVC succeeds in providing both high quality and high compression efficiency with the widely used MPEG-2, but it requires large amount of computational power mainly consumed motion estimation. Many method has been pr...
NAKABAYASHI Tomoyuki, SASAKI Takahiro, OHNO Kazuhiko, KONDO Toshio
IEICE technical report. Computer systems   108(273) 29-34   Oct 2008
Recently, the increase of the energy consumption of mobile computers caused by performance enhancement becomes one serious problem. So many researches for low energy and high performance computing are performed. In order to reduce the energy consu...
AKITA NAOKI, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2008(75) 43-48   Jul 2008
Recently, in the field of not only mobile computing but also high performance computing, the achivement of low energy computing and high performance computing is required, so various techniques are examined. Authors are proposing changeable pipeli...
NOMURA KAZUMASA, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2008(75) 49-54   Jul 2008
In late years, with the mobile processor, the increase of the consumption energy with the performance enhancement of the processor becomes the problem, and low consumption energy and high-performance coexistence are demanded. Therefore it is sugge...
OHNO Kazuhiko, Chou TETSUGUN, SASAKI Takahiro, KONDO Toshio, NAKASHIMA Hiroshi
70(5) "5-105"-"5-106"   Mar 2008
YAMAMOTO Yoshiya, KONDO Toshio, SASAKI Takahiro, OHNO Kazuhiko
IPSJ SIG Notes   2007(125) 19-24   Dec 2007
Recently, the delay time of video coding tends to increase together with the improvement of coding efficiency. Espcially, the delay time about four seconds becomes serious problem in digital terrestrial broadcasting. Then, we proposed a delayed ID...
YAMAMOTO Yoshiya, KONDO Toshio, SASAKI Takahiro, OHNO Kazuhiko
IEICE technical report. Image engineering   107(380) 19-24   Dec 2007
Recently, the delay time of video coding tends to increase together with the improvement of coding efficiency. Especially, the delay time about four seconds becomes serious problem in digital terrestrial broadcasting. Then, we proposed a delayed I...
Katano Tomotaka, Johnen Saburo, Sasaki Takahiro, Ohno Kazuhiko, Kondo Toshio
Technical report of IEICE. ICD   107(382) 71-76   Dec 2007
We proposed a block matching method using bit-truncated images composed of an high frequency 2-bit component and a low frequency 1-bit component for the accurate motion estimation. Combining this matching method with an effective different extende...
YAMAMOTO Yoshiya, KONDO Toshio, SASAKI Takahiro, OHNO Kazuhiko
IEICE technical report. Communication systems   107(379) 19-24   Dec 2007
Recently, the delay time of video coding tends to increase together with the improvement of coding efficiency. Especially, the delay time about four seconds becomes serious problem in digital terrestrial broadcasting. Then, we proposed a delayed I...
ONGA TAKUYA, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2007(115) 51-56   Nov 2007
Power dissipation is one major concern not only for mobile computing but also high performance computing, and achieving both low energy and high performance at the same time is required. Especially, it is important to reduce leakage energy consume...
KATANO SATOSHI, MORI EIICHIROU, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2007(80) 37-42   Aug 2007
We are developing a task-parallel script language named MegaScript for mega-scale computation. To obtain high performance in heterogeneous environment, a scheduling scheme considering load-balancing and communication cost is required. However, opt...
TANIGUCHI KAZUYA, MATSUMOTO MASAKI, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2007(80) 91-96   Aug 2007
Although integrated development environment (IDE) is common in sequential programming, such environment does not exist for parallel programming. This makes large parallel programs difficult to write. So we are developing an IDE for our parallel sc...
AKITA NAOKI, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2007(79) 109-114   Aug 2007
Recently, in the field of not only mobile computing but also high performance computing, the achivement of low energy computing and high performance computing is required simultaneously.The DVS (Dynamic Voltage Scaling) is one of the current major...
ONGA TAKUYA, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2007(79) 115-120   Aug 2007
Power dissipation is a major concern not only for mobile computing but also high performance computing, and achieving both low energy and high performance at the same time is required. Especially, it is important to reduce leakage energy consumed ...
Yamamoto Yoshiya, Sasaki Takahiro, Ono Kazuhiko, Kondo Toshio
Proceedings of the IEICE General Conference   2007(2)    Mar 2007
Katano Tomotaka, Kondo Toshio, Sasaki Takahiro, Ohno Kazuhiko
Proceedings of the IEICE General Conference   2007(2)    Mar 2007
OHARA KAZUKI, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2006(127) 25-30   Nov 2006
Recent years, SMT or CMP processors are used not only high performance computing but also general personal computer or embedded computer. In the near future, one computer system will have many processor cores. In this situation, it is required for...
IGA TAKAYUKI, SASAKI TAKAHIRO, OHNO KAZUHIKO, KONDO TOSHIO
IPSJ SIG Notes   2006(127) 79-84   Nov 2006
Today, multiprocessor systems have spread and Parallelism on the system are widely used in every usage. Particularly, Middle-fine grain parallelism, dividing program into a large number of threads, is paid attention to. Because it hardly relies on...
KATANO SATOSHI, MORI EIICHIROU, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2006(87) 61-66   Jul 2006
We are developing a task-parallel script language named MegaScript. To obtain high performance in mega-scale environment, scheduling scheme considering load-balancing and communication cost is required. However, optimal scheduling is difficult bec...
TAKAGI YUJI, NISHIKAWA TAKEHIKO, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2006(87) 79-84   Jul 2006
We are developing a task parallel script language named MegaScript for megascale computing. MegaScript regards independent sequential/parallel programs as tasks, and executes them in parallel. The current implementation of MegaScript is based on t...
Yamamoto Yoshiya, Sasaki Takahiro, Ono Kazuhiko, Kondo Toshio
Proceedings of the IEICE General Conference   2006(2)    Mar 2006
SAKAGUCHI YUSUKE, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2005(81) 151-156   Aug 2005
In large-scale parallel processing like grid computing, various characteristics of the applications and the environments largely affect the program behaviour. Therefore in the programming language, providing a mechanism for easy handling of such c...
TAKAGI YUJI, NISHIKAWA TAKEHIKO, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2005(81) 157-162   Aug 2005
We are developing a task parallel script language MegaScript for megascale computing. MegaScript regards independent programs as tasks, and executes them in parallel. Single scheduler controls all tasks in current implementation, but it will cause...
Kondo Toshio, Kobayashi Jin, Hiramatsu Takahiro, Sasaki Takahiro, Ohno Kazuhiko
Proceedings of the IEICE General Conference   2005(2)    Mar 2005
Kobayashi Jin, Hiramatsu Takahiro, Sasaki Takahiro, Ohno Kazuhiko, Kondo Toshio
Proceedings of the IEICE General Conference   2005(2)    Mar 2005
MORI EIICHIROU, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2004(128) 19-24   Dec 2004
We are developing a task-parallel script language named MegaScript, which is designed for mega-scale environment. In MegaScript programs, huge number of tasks are executed in parallel. Thus, although load-balancing among hosts and reducing inter-h...
Hashimoto Kentaro, Sasaki Takahiro, Ohno Kazuhiko, Kondo Toshio
3(3) 263-264   Aug 2004
Okuno Tsuyoshi, Kondo Toshio, Ono Kazuhiko, Sasaki Takahiro
3(3) 371-372   Aug 2004
Mamiya Yasuhiro, Kondo Toshio, Ohno Kazuhiko, Sasaki Takahiro
3(3) 377-378   Aug 2004
MIYAMAE Motoki, KONDO Toshio, OHNO Kazuhiko, SASAKI Takahiro
3(3) 379-380   Aug 2004
NISHIZATO HITOSHI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2004(81) 7-12   Jul 2004
We propose a task-parallel script language named MegaScript for mcgascalc computing. A MegaScript program has a tow-tier parallelism; the lower is an ordinary parallelism such as SPMD and the upper is task-level parallelism in which thousands of l...
NISHIKAWA TAKEHIKO, SAKAGUCHI YUSUKE, TANAKA KAZUKI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2004(81) 13-18   Jul 2004
Some applications, such as physical computations, requires great computing power over Pflops. A parallel script language MegaScript is under development to meet such demands. MegaScript provides very flexible and powerful faculty for parallelizati...
SAKAGUCHI YUSUKE, OHNO KAZUHIKO, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2004(81) 19-24   Jul 2004
In large-scale parallel distributed processing, the rate of the communication time to the whole processing time tends to become large. MegaScript is a programming language designed for mega-scale computing, and adopts a stream communication scheme...
OHNO KAZUHIKO, MORI EIICHIROU, SASAKI TAKAHIRO, KONDO TOSHIO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2004(81) 211-216   Jul 2004
We are developing a task-parallel script language named MegaScript for mega-scale parallel processing. For the efficient execution of this language, tasks must be scheduled considering load balancing and minimization of inter-host messages. So we ...
OHTSUKA YASUNORI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2003(83) 113-118   Aug 2003
We proposese a task-parallel script language named MegaScript for megascale computing A MegaScript program has a tow-tier parallelism, the lower is an ordinary parallelism such as SPMD and the upper is task-level parallelism in which thousands of ...
NISHIZATO HITOSHI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2003(83) 119-124   Aug 2003
We are designing and implementing a task parallel script language named MegaScript. MegaScript is developed for megascale computation. MegaScript providesa hierarchical library module for lower level parallel task management, which is easy enough ...
SUZUKI TAKEO, SHIBATA SYUNSUKE, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2003(83) 155-160   Aug 2003
We are pursuing a research project on megascale computing in which more than one million processors are involved. Simultaneously, we propose a simulator named Anastasia, which provides a virtual megascale environment This paper describes detailed ...
ARAI NAOHISA, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2003(29) 55-60   Mar 2003
This paper proposes an improved terminal connection algorithm of the Amaterous parallel wire router, in which the global and detailed routers work independently without the feedback. In Amaterous, a set of rapacity paths (c-paths) to maximize loca...
FUKANO TASUKU, YAMAMOTO SHIGEHIRO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
44(2)    Feb 2003
We are developing a parallel language called Orgel. In the execution model of Orgel, a set of agents are connected with abstract communication channels called streams. The agents run in parallel sending asynchronous messages through the streams. I...
NAKADA TAKASHI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2002(81) 157-162   Aug 2002
Microarchitectural simulation is an essential tool in the research and design of processors, compilers, and other system software. However, existing simulators of out-of-order processors run programs thousands of times slower than actual hardware....
SHIBATA SYUNSUKE, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2002(80) 173-178   Aug 2002
We propose a simulator named Anastasia for large scale distributed computation system as a easily usable development and evaluation tool. Ananstasia simulates a large scale distributed computation environment on a client/server system. It also cap...
ARAI Naohisa, OHNO Kazuhiko, NAKASHIMA Hiroshi
IEICE technical report. Computer systems   102(276) 47-52   Aug 2002
This paper proposes an improved terminal connection algorithm of the Amaterous parallel wire router, in which the global and detailed routers work independently without the feedback. In Amaterous, a set of capacity paths (c-paths) to maximize loca...
MARUYAMA MASAO, YAMAMOTO SHIGEHIRO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
43(3)    Mar 2002
In debugging a parallel program with conventional replay based method, the programmer has to rerun the program repeatedly from its beginning, because the code the programmer wants to examine next might have already gone beyond the breakpoint. To p...
TONOSAKI YURIKO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
43(3)    Mar 2002
The spread of PC clusters and multi-CPU machines makes multiprocessors environment available not only for the reseachers but also for the end users. On the uniprocessor machines, we can use both efficient languages such as C and simple script lang...
MATSUO HARUYUKI, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2001(76) 1-6   Jul 2001
This paper proposes distributed execution-driven simulator for shared memory multiprocessors named Shaman. The simulator consists of the front-end to execute programs in parallel and the back-end, driven by the memory references from the front-end...
TASHIRO TOMOSHIGE, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2001(77) 147-152   Jul 2001
We have proposed the distributed computation system WDC, that has the following features; (1)free join style; (2)widely applicable problem domain; and(3)dynamic computation amount adjustment. In this paper, we discuss about the improvement of WDC ...
Arai Naohisa, Ohno Kazuhiko, Nakashima Hiroshi
IEICE technical report. Computer systems   101(216) 23-29   Jul 2001
This paper proposes an improved terminal connection algorithm of the Amaterous parallel wire router, in which the global and detailed routers work independently without the feedback. In Amaterous, a set of capacity paths (c-paths) to maximize loca...
TANI HISATO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2001(39) 19-24   May 2001
We are now pursuing a research and development project of a new projection system for dome theaters such as planetaria. The system consists of a set of general purpose LCD pro-jectors and a PC-cluster. Since a projector is located on the spherical...
YAMAMOTO SHIGEHIRO, OHNO KAZUHIKO, NAKASHIMA HIROSHI
42(3)    Mar 2001
We are developing a parallel programming language named Orgel based on multi-agent paradigm.In the execution model of Orgel, a set of agents are connected with abstract communication channels called streams.The agents run in parallel sending async...
OHNO KAZUHIKO, OKANO TAKANORI, YAMAMOTO SHIGEHIRO, NAKASHIMA HIROSHI
41(9)    Nov 2000
Automatic Parallelization is much difficult in non-numerical field, because irregular and dynamic data structures are frequently used. Therefore many researches adopts the approaches such as using MPI/Pthreads on sequential language of designing a...
MATSUO HARUYUKI, NAKASHIMA HIROSHI, OHNO KAZUHIKO
IPSJ SIG Notes   2000(74) 13-18   Aug 2000
In order to reduce the overhead of synchronizing operations of shared memory multiprocessors, we have proposed a mechanism named specMEM to excute memory accesses following a synchronizing operation speculatively before the completion of the synch...
TASHIRO TOMOSHIGE, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2000(73) 143-148   Aug 2000
Recently, free-join style distributed computation is so attractive that various systems based on this paradigm have been developed. These systems, however, are too specialized for their own applications to use as general purpose frameworks. Thus, ...
IMAFUKU SHIGERU, OHNO KAZUHIKO, NAKASHIMA HIROSHI
IPSJ SIG Notes   2000(1) 37-42   Jan 2000
This paper proposes an efficient techinique for distributed simulation targeting shared memory parallel machines. Our approach to solve the problem of distributed clock management, which is the well-known bottleneck of distributed simulators, is a...
Tanaka Kotaro, Ohno Kazuhiko, Nakashima Hiroshi
IEICE technical report. Computer systems   99(252) 55-62   Aug 1999
This paper propose a new algorithm in which the global and detailed routers work independently. In this algorithm, a set of virtual paths called capacity path to maximize local wireability is obtained prior to global routing. Then global router wh...
IMAFUKU Shigeru, OHNO Kazuhiko, NAKASHIMA Hiroshi
IPSJ SIG Notes   99(67) 37-42   Aug 1999
Simulation of a large computer system such as a parallel machine takes much computation time. Parallelization will be a good solution, but it cause a serious performance bottleneck if we directly map inter-processor hardware events for synchroniza...
Sato Takayuki, Nakashima Hiroshi, Ohno Kazuhiko
IPSJ SIG Notes   99(41) 25-30   May 1999
In order to reduce the overhead of synchronizing operation of shared memory multiprocessors, we have proposed a mechanism to execute memory accesses following a synchronizing operation speculatively before the completion of the sysnchronization is...
Sugiyama Namiyo, Ohno Kazuhiko, Nakashima Hiroshi
56(1) 326-327   Mar 1998
OHNO KAZUHIKO, IKAWA MASAHIKO, MORI SHIN-ICHIRO, NAKASHIMA HIROSHI, TOMITA SHINJI
IPSJ Journal   38(8) 1638-1648   Aug 1997
In the execution of concurrent logic language KL1 on message-passing multiprocessors, frequent fine-grained communications cause a drastic inefficiency. We propose an optimization scheme which achieves high granularity of messages by packing data ...
Lecture Notes in Computer Science1345 Advances in Computing Scince-ASIAN'97   268-282   1997
Improvement of Message Communication in Concurrent Logic Language
Proceeding of the Second international Symposium on Parallel Sympolic Computation PASCO'97   156-164   1997
Orgel : A Parallel Programming Language with Declarative Communication Streams
Proceedings of the Third International Symposium ISHPC 2000   344-354   2000
IKAWA MASAHIKO, OHNO KAZUHIKO, GOSHIMA MASAHIRO, MORI SHIN-ICHIRO, NAKASHIMA HIROSHI, TOMITA SHINJI
IPSJ SIG Notes   96(107) 43-48   Oct 1996
In the conventional run-time system for the parallel logic programming language KLl, its fine-grained concurrency control frequently casues unnecessary goal switching which degrades its execution performance. We propose an optimization method to r...
Ikawa Masahiko, Ohno Kazuhiko, Nakashima Hiroshi, Tomita Shinji
IPSJ SIG Notes   95(82) 105-112   Aug 1995
In the execution of parallel logic programming language KL1 on message-passing multiprocessors, fine-grained communications occur frequently. This cause a drastic performance decline. To decrease overhead of communication, we proposed packing-send...
Ohno Kazuhiko, Ikawa Masahiko, Mori Shinichiro, Nakashima Hiroshi, Tomita Shinji
IEICE technical report. Computer systems   94(384) 87-94   Dec 1994
The KLIC system,a portable implementation of a concurrent logic programming language KL1,is being developed in ICOT.However,the current implementation is less efficient compared to procedural languages.We propose an optimization method using stati...
Ohno Kazuhiko, Ikawa Masahiko, Mori Shin-ichiro, Nakashima Hiroshi, Tomita Shinji
IPSJ SIG Notes   94(106) 183-190   Dec 1994
The KLIC system, a portable implementation of a concurrent logic programming language KL1, is being developed in ICOT. However, the current implementation is less efficient compared to procedural languages. We propose an optimization method using ...
Ohno Kazuhiko, Nakashima Hiroshi, Tomita Shinji
94(65) 17-24   Jul 1994
Although concurrent logic language is suitable for symbol processing, its execution is inefficient due to the dynamic overhead. We propose an optimization scheme using static analysis. Because each process is dealt independently, the analysis and ...
Ohno Kazuhiko, Mori Shin-ichiro, Nakashima Hiroshi, Tomita Shinji
93(73) 57-64   Aug 1993
For programmers of tree searching applications on loosely-coupled multiprocessors, it is troublesome to manage the tree structure spread across processors, and to balance the load of processors dynamically. As a solution of these problems, we prop...

Research Grants & Projects

 
Ministry of Education, Culture, Sports, Science and Technology: Grants-in-Aid for Scientific Research(基盤研究(C))
Project Year: 2003 - 2006    Investigator(s): Toshio KONDO
We proposed two delayed decoder refresh methods, named Delayed GDR and Delayed IDE, which reduce video decoding delay due to decoder buffer by double encoding of each refresh part and by allowing refresh delay. The Delayed GDR and the Delayed IDR ...
Design and Implementation of Agent-oriented Parallel Programming Language
Design and Implementation of Parallel Script Language
Design and Implementation of a Task Parallel Programming Language
JST Basic Research Programs (Core Research for Evolutional Science and Technology :CREST)
Project Year: 2002 - 2006