RIKEN Advanced Institute for Computational Science

Menu
Menu
Events/Documents イベント・広報

AICS Cafe

AICS Cafe is a place where the researchers in AICS can frankly discuss about their researches beyond their own disciplinary wall in order to collaborate with each other. We plan to have it twice a month regularly. We welcome all people including the promotion office and administration division of K computer in AICS.

  • Purpose : In order to promote the research collaboration beyond each of existing research disciplines, this seminar provides the discussion field for exchanging information, understanding neighboring researchers, and collaboration between each other.
  • Place: Workshop room (6th floor) at AICS
  • Lang : presentation in Japanese or English, the slide in English
  • Etc.: Please give your presentations clearly to researchers in other fields. Please do not hesitate to ask a question to the speakers.

The 121st AICS Cafe
Date and Time: Wed. Sep. 20, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: TBD
Speaker: Ryuji Yoshida (Computational Climate Science Research Team)

Presentation Language: TBD
Presentation Material: TBD

    Abstract: TBD

The 120th AICS Cafe
Date and Time: Wed. Sep. 13, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: TBD
Speaker: Hiroki Kashimura (Center for Planetary Science (CPS), Kobe University)

Presentation Language: TBD
Presentation Material: TBD

    Abstract: TBD

The 119th AICS Cafe
Date and Time: Wed. Sep. 6, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: TBD
Speaker: TBD

Presentation Language: TBD
Presentation Material: TBD

    Abstract: TBD

The 118th AICS Cafe
Joint seminar (Large-scale Parallel Numerical Computing Technology Research Team)
Date and Time: Wed. Aug. 2, 2017, 14:00-16:30
Place: Workshop room (6th floor) at AICS

Speaker:
Keita Teranishi (Sandia National Laboratories, California)
Balazs Gerofi (System Software ResearchTeam)
Miwako Tsuji(Programming Environment Research Team)

Presentation Language: English
Presentation Material: English

    Program:
【14:00-15:00】Keita Teranishi (Sandia National Laboratories, California)
Title: Toward Resilient Asynchronous Many Task Programming
Abstract: As semiconductor technology reaches its physical limit, the performance improvement of high performance computing systems no longer follows the predictions by Moore’s law. One of the viable approaches to address this stagnation is to relax the reliability of computing systems, and leave the application users to manage it. To enable this idea, it is essential for programming model to embrace a resilience capability. Today, the major resilience framework is coordinated checkpoint and restart (C/R), which involves global coordination of processes and threads for accommodating consistent global application state. However, this global recovery model entails inherent scalability issues and disproportionate use of resources to respond to local failures. These issues are better handled through asynchronous many task (AMT) programming model that is intended for deriving good scalability from unprecedented parallelism and complexity of node architecture of future HPC systems. A runtime system with AMT enables abstractions for encapsulating streams of program execution (tasks) and organizing the application data as objects rather than a sequence of data mapped to the system address space. In the AMT model, resilience is achieved through task re-execution and replication facilitated with versioning and replication of data objects. However, extensive research on task-based resilience is still required to determine the roadmap of resilience in the context of the programming environment. We will discuss our ongoing activities on the resilience of high performance AMT programming model and the challenges for scalable HPC application resilience.

【15:00-15:30】Break

【15:30-16:00】Miwako Tsuji(Programming Environment Research Team)
Title: Fault tolerance features in an XMP-YML scientific workflow programming model
Abstract: Supercomputers in the exa-scale era would consist of a huge number of nodes arranged in a multi-level hierarchy. There are many important challenges to exploit such systems such as scalability, programmability, reliability etc... In this talk, we focus on the scalability, programmability and fault tolerance features of a multi SPMD programming model. We have developed a development and execution environment based on workflow and PGAS (Partitioned Global Address Space) . We have extended the environment by incorporating fault resilience scheduling policy into the workflow scheduler.

【16:00-16:30】Balazs Gerofi (System Software ResearchTeam)
Title: IHK/McKernel: A Lightweight Multi-kernel based Operating System for Extreme Scale Supercomputing
Abstract: RIKEN Advanced Institute for Computation Science leads the development of Japan's next generation flagship supercomputer, the successor of the K Computer. Part of this effort is to design and develop a system software stack that suits the needs of future extreme scale computing. In this talk, we focus on operating system research and discuss IHK/McKernel, our multi-kernel based operating system framework. IHK/McKernel runs Linux with a light-weight kernel side-by-side on compute nodes with the primary motivation of providing scalable, consistent performance for large scale HPC simulations, but at the same time to retain a fully Linux compatible execution environment. We present an overview of the system architecture, provide preliminary results on up to two thousand Intel Xeon Phi nodes and outline future research directions.

The 117th AICS Cafe
Date and Time: Fri. July 28, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: How low can you go? Reducing the precision of data assimilation to improve weather forecast skill
Speaker: Mr. Samuel Hatfield (University of Oxford; Data Assimilation Research Team, Intern)

Presentation Language: English
Presentation Material: English

    Abstract:
Data assimilation, the process by which atmospheric data is combined with atmospheric models, is essential for skillful weather forecasts. Inserting data into models allows us to characterize more accurately the weather state at the start of the forecast, thereby extending the time in which the forecast is useful. However, data assimilation is a very computationally expensive process, and often costs as much as the actual weather forecasts. One way to reduce the cost is to lower the precision of the data assimilation computations. Lower precision computations use fewer bits to produce the answer to a calculation, and are therefore computationally cheaper. Lowering precision also introduces errors, but these errors may be acceptable, given that our models and observations are imperfect.
I will present some results from my PhD research on the subject of precision in data assimilation. I will show how lowering the precision of the data assimilation algorithm affects the quality of the output. I will demonstrate that the lowest precision that you can use is related to the overall quality of the model and the observations – the better the model, the more important precision becomes. Additionally, I will show that, by reusing the computational resources that we save when lowering precision, we can actually improve the data assimilation product and – ultimately – the skill of weather forecasts.

The 116th AICS Cafe
Date and Time: Wed. July 19, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: Leading Research on Data Assimilation in Numerical Weather Prediction
Speaker: Shunji Kotsuki (Data Assimilation Research Team)

Presentation Language: Japanese
Presentation Material: English

    Abstract: Detail

Estimating an accurate initial condition is very important for numerical weather prediction (NWP) because the weather system is known to be sensitive to the initial condition. Data assimilation plays an essential role in NWP to estimate the best initial condition using both simulation and observation data with their error statistics provided. Data Assimilation Research Team has been developing global and regional NWP systems known as the NICAM-LETKF and SCALE-LETKF, in which the ensemble Kalman filter, an advanced data assimilation method, is used. In this talk, I will present a brief introduction to data assimilation in NWP, and will convey some flavor of cutting-edge research such as precipitation data assimilation.

The 115th AICS Cafe
Date and Time: Wed. July 12, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: Development of FDPS (Framework for Developing Particle Simulators) on Sunway Taihulight
Speaker: Daisuke Namekata (Particle Simulator Research Team)

Presentation Language: English
Presentation Material: English

    Abstract: Detail

FDPS is a library for massively parallel particle simulations, which has been developed by our team. The current released version of FDPS can generate a parallel code that scales up to the K computer (Iwasawa et al. 2015, PASJ, 68, 54). However, it is not clear that FDPS generate a scalable code in the next generation of supercomputers. In order to identify potential problems of FDPS and to improve FDPS more, we need to analyze FDPS in a system larger than the K computer. In late January, we started the development of FDPS on Sunway Taihulight, which is the most powerful supercomputer in the world (as of Nov. 2016, 125.4 PFlops). In this talk, I'd like to introduce the current status of this project.

The 114th AICS Cafe
Date and Time: Wed. June 21, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: Optimization of communications on Large scale parallel computers
Speaker: Yoshiyuki Morie (System Software Research Team)

Presentation Language: Japanese
Presentation Material: English

    Abstract: Detail

In the recently, the scale of supercomputer has been increasing remarkably. This trend will have been going from now. On the other hand, in network, multiple communication devices are installed and fat-tree, mesh/torus as network topology are adopted. There are several problems to be solved for adopting these. One is the efficient utilization of multiple communication devices. Another one is shortening communication delay by reducing the number of hops or contentions on a such network topology. We will introduce the optimization of communication for these problems.

The 113th AICS Cafe
Date and Time: Wed. June 7, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: Study of Mercury's magnetosphere based on the MHD simulation and trajectory tracings
Speaker: Manabu Yagi (Programming Environment Research Team)

Presentation Language: Japanese
Presentation Material: English

    Abstract: Detail

Interaction between solar wind(supersonic plasma flow from the Sun) and intrinsic magnetic field of planets such as Earth forms plasma circulation system called "magnetosphere". A great number of studies have been done for the Earth magnetosphere using spacecraft observations, numerical simulations, and modelings. On the other hands, comparative studies of planetary magnetosphere such as Mercury or Jupiter are also important from a perspective of universal plasma physics. In this study, we try to discover the dynamics of Mercury's magnetosphere using magnetohydrodynamics(MHD) simulation, and trajectory tracings of test particles.

The 112th AICS Cafe
Date and Time: Wed. May 24, 2017, 15:30-16:30
Place: Workshop room (6th floor) at AICS

Title: Contribution of computational science for reducing the damage of flood disaster and activities among citizens, authorities and academia
Speaker: Satoru Oishi (Unit Leader for Computational Disaster Mitigation and Reduction Research Unit)

Presentation Language: Japanese
Presentation Material: Japanese

    Abstract: Detail

After flood disaster happened in Kinugawa river basin in 2015, citizens in Mukogawa river basin warried about torrential rain in the basin and they wanted to know what kind of damage they would suffer when it should be similar rain. The presentation explains the result of simulation with several introductions including Japanese river improvement plan and implementation, newer trial of computation science for dam utilization and contribution of science and technology to evacuation activity of citizens.