The parallel multigrid method is expected to play an important role in large-scale scientific computing on exa-scale supercomputer systems. Previously we proposed Hierarchical Coarse Grid Aggregation (hCGA), which dramatically improved the performance of the parallel multigrid solver when the number of MPI processes was O(10^4) or more. Because hCGA can handle only two layers of parallel hierarchical levels, the computation overhead due to coarse grid solver may become significant when the number of MPI processes reaches O(10^5)- O(10^6) or more. In the present work, we propose AM-hCGA (Adaptive Multilevel hCGA) that can take into account multiple layers of three or more levels, and show preliminary results using the Oakforest-PACS (OFP) system by JCAHPC. Additionally, we also examine the impact of a lightweight multi-kernel operating system, called IHK/McKernel, for parallel multigrid solvers running on OFP.

This is a joint work with Balazs Gerofi (RIKEN R-CCS), Yutaka Ishikawa (RIKEN R-CCS), and Masashi Horikoshi (Intel).

Recently, FPGA has been attracting attention as an alternative device to accelerate HPC applications. Data flow computing model is a popular abstraction of computing in both fine-grain and coarse-grain from decades, and this model is used as a programming model for FPGA such as Maxler DFE and SPGen (Stream Processor Generator). While it is a kind of “static” data flow model, it might be interesting to extend models by using “dynamic” data flow models as old dataflow architecture to handle the dynamic behavior of systems. On other hands, global programming models to integrate FPGA computing into parallel computing of host processors are also important. OpenMP task and target directives, which are recently introduced in OpenMP 4.5, can be extended to specify the interface to offloaded computation done by FPGA. And, the optimization for FGPA needs different metric such as hardware resources, which are very different from the optimization of CPU and GPU. In this talk, issues on programming models for FPGA accelerated HPC are presented.

The K computer shut down in August with many achievements. We are now working on the installation and preparation of the operation of the Fugaku supercomputer. In the spring of next year, an early access program that uses a part of Fugaku installed in R-CCS will start. To improve user services, we are now considering "cloud service," which includes collaboration with commercial service providers. The "cloud service" is expected to contribute to improve the usability of Fugaku and increase the number of users and application areas drastically. To achieve highly energy-efficient, we optimize the Fugaku system and its facility operation based on operation data analysis and have to motivate users to increase energy efficiency by using Power API. To steadily advance these new activities, it is more critical to collaborate with research teams and operation division tightly. In this talk, I give a vision of the operation improvement towards new service on Fugaku.

Every proton and neutron are made of elementary particles called quark. The dynamics of quarks is described by Quantum Chromo Dynamics (QCD) and it can be formulated on a structure lattice (Lattice QCD or simply LQCD). The bottle neck of LQCD simulation, which a Markov Chain Monte Carlo (MCMC), is solving large sparse linear equation. It appears in both generating configurations and calculating physical observables using configurations. In this talk, I will discuss linear solvers for LQCD application. I will mainly focus on our recent implementation of multigrid type solvers. I will also discuss usage of simd variables in LQCD.

ADP/ATP Carrier (AAC) is a membrane transport protein embedded in the inner mitochondrial membrane and mediates a 1:1 exchange of ADP and ATP. The transporter is known to alternate between the so-called c-state and the m-state in the course of the transport cycle. The crystallographic structure of the c-state has been known since 2003, while that of the m-state is not known until 2019. In 2015, we developed a molecular dynamics (MD) technique to simulate global conformational changes of protein. The method, linear response path following (LRPF), enables us to sample large conformational changes of protein without the prior knowledge of the target conformation. The method is suitable for the study of AAC, as no structural knowledge of the m-state is available when the project is started. Using LRPF simulations, we successfully predicted a model structure for the m-state structure (Tamura and Hayashi, 2017, PLOS ONE). The predicted structure was later validated by the experimentally determined m-state structure.

The High Performance Big Data Research Team at RIKEN Center for Computational Science (RIKEN R-CCS) has been researching and developing system software to facilitate extreme-scale big data processing, machine learning and deep learning for high performance computing (HPC) systems, i.e., convergence of AI/Big Data and HPC. In this talk, Kento Sato introduce several R&D activities for accelerating AI/Big data applications on HPC systems (HPC for AI/Big data) as well as ones for resolving HPC challenges by using these AI/Big data techniques (AI/Big data for HPC).

High resolution structures of biological molecules provide insights into their functions. Such structures are often derived from X-ray crystallographic experiments, but the information is not complete. Crystals are first cryo-cooled before collecting X-ray data. It has been suggested that, while backbone structures are usually very similar between room-temperature and cryo-temperature, cryo-cooling may hamper biologically relevant dynamics. In addition, X-ray crystallography requires crystals, which might lead to some artifacts as biomolecules might interact which each other, stabilizing specific conformations which might be different than in solution.
Furthermore, to fully understand function of biomolecules, dynamics need also to be considered. However, such X-ray crystallographic structure only represent one specific snapshot of the overall conformational space that biomolecules can cover. Therefore, to complement experimental data, Molecular Dynamics (MD) simulations can be utilized to connect structure, dynamics and function. We will be illustrating those points with studies we have been conducting on different biological systems.