Keichi Takahashi
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Keichi Takahashi has a total of 99 co-authors. Here are the top co-authors by number of shared publications:
- Hiroyuki Takizawa
Tohoku University, Sendai, Japan
17 shared publications - Shinji Shimojo
Aomori University, Japan
11 shared publications Susumu Date
D3 Center University of Osaka Osaka Japan
9 shared publicationsDashdavaa Khureltulga
Osaka University Osaka, Japan
7 shared publications- Yoichi Shimomura
Tohoku University
7 shared publications Yoshiyuki Kido
Osaka University Osaka, Japan
7 shared publications
Latest Publications
- Open Access
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-8
SmartNIC-based Memory Monitoring for Lightweight Privilege Escalation Detection in HPC Systems
- 26 July 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-8
Open AccessSmartNIC-based Memory Monitoring for Lightweight Privilege Escalation Detection in HPC Systems
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 370-377
Semantic Equivalence Verification of HPC Codes Using LLMs
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 370-377
Open AccessSemantic Equivalence Verification of HPC Codes Using LLMs
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 177-185
CityScaleCast: Spatiotemporal GNN for City-Scale Weather Prediction with GraphCast-Guided Parallel Modeling and Multi-Step Forecasting in Sendai
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 177-185
Open AccessCityScaleCast: Spatiotemporal GNN for City-Scale Weather Prediction with GraphCast-Guided Parallel Modeling and Multi-Step Forecasting in Sendai
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 186-197
Explainable AI-Guided Genetic Algorithms for Efficient Software Automatic Tuning
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 186-197
Open AccessExplainable AI-Guided Genetic Algorithms for Efficient Software Automatic Tuning
- Open Access
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 31-41
TRIOS: Reducing File-System Contention through Predictive Time-Resolved I/O Simulation in Job Scheduling
- 25 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 31-41
Open AccessTRIOS: Reducing File-System Contention through Predictive Time-Resolved I/O Simulation in Job Scheduling
- Open Access
- 13 December 2025
- Published by Springer Nature in SN Computer Science
- Vol. 6 (8) , 1021
mdx II: A High-Performance Cloud Infrastructure for Data-Intensive and Cross-Disciplinary Science
Mdx II is an Infrastructure-as-a-Service (IaaS) cloud developed to accelerate data science research and promote interdisciplinary collaboration among Japanese universities and research organizations. In contrast to conventional high-performance computing (HPC) systems, mdx II utilizes OpenStack to deliver flexible, isolated computing resources, including virtual machines, virtual networks, and storage systems. This article provides a thorough performance assessment of mdx II, benchmarking it against public cloud and bare metal server. We analyzed the performance of a 16-vCPU virtual machine in several aspects: floating-point computation, memory bandwidth, network throughput, file system and object storage access performance, and real-world application performance. When compared to an AWS 16-vCPU instance, mdx II was found to excel in many aspects, highlighting its strong potential for high-performance data analytics (HPDA) workloads. Additionally, we investigated the effects of virtualization overhead using a 224-vCPU virtual machine that fully occupies a physical host. The results revealed that virtualization overhead is negligible for compute-bound tasks, but more pronounced for memory-intensive workloads. We also evaluated the performance of a virtual machine equipped with GPUs and demonstrated that the impact of virtualization on GPU performance is negligible. These insights are anticipated to help mdx II users in achieving optimal performance for data science tasks and to guide the development of future data-centric cloud infrastructures.
- 13 December 2025
- Published by Springer Nature in SN Computer Science
- Vol. 6 (8) , 1021
Open Accessmdx II: A High-Performance Cloud Infrastructure for Data-Intensive and Cross-Disciplinary Science
Mdx II is an Infrastructure-as-a-Service (IaaS) cloud developed to accelerate data science research and promote interdisciplinary collaboration among Japanese universities and research organizations. In contrast to conventional high-performance computing (HPC) systems, mdx II utilizes OpenStack to deliver flexible, isolated computing resources, including virtual machines, virtual networks, and storage systems. This article provides a thorough performance assessment of mdx II, benchmarking it against public cloud and bare metal server. We analyzed the performance of a 16-vCPU virtual machine in several aspects: floating-point computation, memory bandwidth, network throughput, file system and object storage access performance, and real-world application performance. When compared to an AWS 16-vCPU instance, mdx II was found to excel in many aspects, highlighting its strong potential for high-performance data analytics (HPDA) workloads. Additionally, we investigated the effects of virtualization overhead using a 224-vCPU virtual machine that fully occupies a physical host. The results revealed that virtualization overhead is negligible for compute-bound tasks, but more pronounced for memory-intensive workloads. We also evaluated the performance of a virtual machine equipped with GPUs and demonstrated that the impact of virtualization on GPU performance is negligible. These insights are anticipated to help mdx II users in achieving optimal performance for data science tasks and to guide the development of future data-centric cloud infrastructures.
- Open Access
Performance analysis of mdx II: A next-generation cloud platform for cross-disciplinary data science research
mdx II is an Infrastructure-as-a-Service (IaaS) cloud platform designed to accelerate data science research and foster cross-disciplinary collaborations among universities and research institutions in Japan. Unlike traditional high-performance computing systems, mdx II leverages OpenStack to provide customizable and isolated computing environments consisting of virtual machines, virtual networks, and advanced storage. This paper presents a comprehensive performance evaluation of mdx II, including a comparison to Amazon Web Services (AWS). We evaluated the performance of a 16-vCPU VM from multiple aspects including floating-point computing performance, memory throughput, network throughput, file system and object storage performance, and real-world application performance. Compared to an AWS 16-vCPU instance, the results indicated that mdx II outperforms AWS in many aspects and demonstrated that mdx II holds significant promise for high-performance data analytics (HPDA) workloads. We also evaluated the virtualization overhead using a 224-vCPU VM occupying an entire host. The results suggested that the virtualization overhead is minimal for compute-intensive benchmarks, while memory-intensive benchmarks experienced larger overheads. These findings are expected to help users of mdx II to obtain high performance for their data science workloads and offer insights to the designers of future data-centric cloud platforms.
Open AccessPerformance analysis of mdx II: A next-generation cloud platform for cross-disciplinary data science research
mdx II is an Infrastructure-as-a-Service (IaaS) cloud platform designed to accelerate data science research and foster cross-disciplinary collaborations among universities and research institutions in Japan. Unlike traditional high-performance computing systems, mdx II leverages OpenStack to provide customizable and isolated computing environments consisting of virtual machines, virtual networks, and advanced storage. This paper presents a comprehensive performance evaluation of mdx II, including a comparison to Amazon Web Services (AWS). We evaluated the performance of a 16-vCPU VM from multiple aspects including floating-point computing performance, memory throughput, network throughput, file system and object storage performance, and real-world application performance. Compared to an AWS 16-vCPU instance, the results indicated that mdx II outperforms AWS in many aspects and demonstrated that mdx II holds significant promise for high-performance data analytics (HPDA) workloads. We also evaluated the virtualization overhead using a 224-vCPU VM occupying an entire host. The results suggested that the virtualization overhead is minimal for compute-intensive benchmarks, while memory-intensive benchmarks experienced larger overheads. These findings are expected to help users of mdx II to obtain high performance for their data science workloads and offer insights to the designers of future data-centric cloud platforms.
- 21 December 2024
- Published by Springer Nature
- p. 161-180
Maximizing Energy Budget Utilization Using Dynamic Power Cap Control
- 21 December 2024
- Published by Springer Nature
- p. 161-180
Maximizing Energy Budget Utilization Using Dynamic Power Cap Control
- Open Access
Modernizing an Operational Real-time Tsunami Simulator to Support Diverse Hardware Platforms
To issue early warnings and rapidly initiate disaster responses after tsunami damage, various tsunami inundation forecast systems have been deployed worldwide. Japan's Cabinet Office operates a forecast system that utilizes supercomputers to perform tsunami propagation and inundation simulation in real time. Although this real-time approach is able to produce significantly more accurate forecasts than the conventional database-driven approach, its wider adoption was hindered because it was specifically developed for vector supercomputers. In this paper, we migrate the simulation code to modern CPUs and GPUs in a minimally invasive manner to reduce the testing and maintenance costs. A directive-based approach is employed to retain the structure of the original code while achieving performance portability, and hardware-specific optimizations including load balance improvement for GPUs are applied. The migrated code runs efficiently on recent CPUs, GPUs and vector processors: a six-hour tsunami simulation using over 47 million cells completes in less than 2.5 minutes on 32 Intel Sapphire Rapids CPUs and 1.5 minutes on 32 NVIDIA H100 GPUs. These results demonstrate that the code enables broader access to accurate tsunami inundation forecasts.
Open AccessModernizing an Operational Real-time Tsunami Simulator to Support Diverse Hardware Platforms
To issue early warnings and rapidly initiate disaster responses after tsunami damage, various tsunami inundation forecast systems have been deployed worldwide. Japan's Cabinet Office operates a forecast system that utilizes supercomputers to perform tsunami propagation and inundation simulation in real time. Although this real-time approach is able to produce significantly more accurate forecasts than the conventional database-driven approach, its wider adoption was hindered because it was specifically developed for vector supercomputers. In this paper, we migrate the simulation code to modern CPUs and GPUs in a minimally invasive manner to reduce the testing and maintenance costs. A directive-based approach is employed to retain the structure of the original code while achieving performance portability, and hardware-specific optimizations including load balance improvement for GPUs are applied. The migrated code runs efficiently on recent CPUs, GPUs and vector processors: a six-hour tsunami simulation using over 47 million cells completes in less than 2.5 minutes on 32 Intel Sapphire Rapids CPUs and 1.5 minutes on 32 NVIDIA H100 GPUs. These results demonstrate that the code enables broader access to accurate tsunami inundation forecasts.
- Open Access
- 1 May 2024
- Published by IOP Publishing in Applied Physics Express
- Vol. 17 (5) , 052006
Towards sub-10 nm spatial resolution by tender X-ray ptychographic coherent diffraction imaging
As the first experiment at BL10U in NanoTerasu, tender X-ray ptychographic coherent diffraction imaging (PCDI) was conducted using a photon energy of 3.5 keV. The ptychographic diffraction patterns from a 200 nm thick Ta test chart and a micrometer-sized particle of sulfurized polymer were collected. Subsequently, phase images were reconstructed with resolutions of sub-20 nm and sub-50 nm, respectively. In the near future, tender X-ray PCDI with sub-10 nm resolution is anticipated to potentially revolutionize the visualization of nanoscale structures and chemical states in various functional materials composed of light elements.
- 1 May 2024
- Published by IOP Publishing in Applied Physics Express
- Vol. 17 (5) , 052006
Open AccessTowards sub-10 nm spatial resolution by tender X-ray ptychographic coherent diffraction imaging
As the first experiment at BL10U in NanoTerasu, tender X-ray ptychographic coherent diffraction imaging (PCDI) was conducted using a photon energy of 3.5 keV. The ptychographic diffraction patterns from a 200 nm thick Ta test chart and a micrometer-sized particle of sulfurized polymer were collected. Subsequently, phase images were reconstructed with resolutions of sub-20 nm and sub-50 nm, respectively. In the near future, tender X-ray PCDI with sub-10 nm resolution is anticipated to potentially revolutionize the visualization of nanoscale structures and chemical states in various functional materials composed of light elements.