Keynote Talks

Chair: Michihiro Koibuchi, National Institute of Informatics

Speaker: Akihiro Tabuchi, Fujitsu Ltd.

Title: Quantum Computer Simulators and Quantum Applications in the NISQ Era

Abstract: Quantum computing is attracting a lot of attention both in academia and industry, with many research institutes and companies developing a wide variety of quantum computers, emerging quantum applications, and quantum computer simulators. Fujitsu also announced the development and provision of a 64-qubit superconducting quantum computer in collaboration with RIKEN RQC in October 2023. Current quantum computers are called Noisy Intermediate-Scale Quantum (NISQ) devices, where computational results are strongly affected by noise errors. Thus, it is extremely difficult to perform large-scale accurate quantum computations. Therefore, the development of novel techniques to mitigate the impacts of noise errors is necessary for the practical use of current or near-future NISQ devices. For these developments, quantum computer simulators are also essential to understand the behavior of quantum computers with and without noise errors. This talk will introduce Fujitsu's recent work in the development of quantum computer simulators and applications for quantum chemistry calculations.

Chair: Jacir L. Bordim, University University of Brasilia

Speaker: Hideharu Amano, Keio University

Title:　Contribution of FPGA to Quantum Computing

Abstract: Quantum computing offers various applications for FPGAs. The controller for quantum computers is configured by FPGAs, and the presence of FPGAs is crucial for achieving future fault-tolerant quantum computation (FTQC). Also, the quantum computer simulator is an important application. Since state vector simulation of the machine with N qubits requires 2 ^ (N+4) byte memory, a supercomputer is required to simulate a system with more than 40 qubits. We have developed an FPGA board directly connected 32 SATA disks and are trying to realize a super-cost-efficient quantum computer simulator. The problem is that the access speed of SATA is much slower than that of supercomputer on-core memory. In this talk, various techniques to improve the simulation speed are introduced.

Chair: Hideyuki Kawashima, Keio University

Speaker: Michihiro Koibuchi, National Institute of Informatics

Title: Photonic Approximate Communication Highlighting Ultimate Nature of Light

Abstract:The concept of approximate computing, which fills the gap between the accuracy provided by a system and that required by an application for improving energy and performance, can be applied to communication technology in parallel and distributed computer systems. This talk focuses on photonic approximate communication that highlights the ultimate nature of light. With classical communication, each signal is processed one bit at a time. In contrast, this approximate computing performs multi-value processing on each wavelength, making it possible to improve processing speeds up to ten times the current speeds. Using multi-value processing may lead to low reliability and low technological maturity, which become a new research challenge. Approximate communication techniques are classified into lossy data compression, imperfect hardware usage, and approximate synchronization in parallel processing. This talk illustrates their case studies, including our research outputs in an FPGA cluster using an optical interconnection network.

Chair: Junya Nakamura, Toyohashi University of Technology

Speaker: Sébastien Tixeuil, Sorbonne University, CNRS, LIP6, Institut Universitaire de France

Title: A reality check on self-stabilization

Abstract: It is almost fifty years since Dijkstra coined the term self-stabilization to denote a distributed system able to recover correct behavior starting from any arbitrary (even unreachable) configuration. His seminal paper triggered many works since then, exploring over the years new variants of the original concept, new application domains, and new complexity results. While the huge majority of those contributions relates to theory, considering computability and worst case complexity issues, this talk revisits old and recent contributions from the prism of “realistic” distributed systems, aiming to address the following question: is self-stabilization relevant for actually deployed distributed systems?

Chair: Hideharu Amano, Keio University

Speaker:Junichiro Kadomoto, The University of Tokyo

Title: Wireless Chiplets-Based Shape-Changeable Computer System for Sustainable Computing

Abstract: The adoption of chiplet-based implementation techniques is advancing to address challenges in computer system design and manufacturing costs. With the miniaturization and enhanced integration of integrated circuits, computer systems are increasingly being embedded in various devices and interconnected wirelessly. The shape-changeable computer system proposed by the presenters comprises multiple tiny chiplets wirelessly connected, enabling diverse shapes and functionalities. Each chip is equipped with coils and a wireless transceiver circuit, facilitating wireless data exchange between adjacent chips. This presentation will discuss the current state of research and development of the shape-changeable computer system and offer insights into future prospects for sustainable computing.===