Large-Scale Systems Management

Research Staff

  • Professor Shoji KASAHARA

    Professor
    Shoji KASAHARA

  • Associate Professor Masahiro SASABE

    Associate Professor
    Masahiro SASABE

  • Assistant Professor Takanori HARA

    Assistant Professor
    Takanori HARA

  • Assistant Professor Yu NAKAHATA

    Assistant Professor
    Yu NAKAHATA

Research Areas

Fig.1 Distributed virtual currency and smart contract network

Fig.1 Distributed virtual currency and smart contract network

Fig.2 Hazard-area estimation and evacuation guidance using trajectories of mobile terminals

Fig.2 Hazard-area estimation and evacuation guidance using trajectories of mobile terminals

Fig.3 Ultra-scalable blockchain technology

Fig.3 Ultra-scalable blockchain technology

Fig.4 Cognitive radio networks

Fig.4 Cognitive radio networks

Fig.5 Blockchain-based access control

Fig.5 Blockchain-based access control

Fig.6 An example of a deep reinforcement learning framework for service function chaining

Fig.6 An example of a deep reinforcement learning framework for service function chaining

Fig.7 An example of indexing and querying using a ZDD

Fig.7 An example of indexing and querying using a ZDD

System analytics and simulation

  • Large-scale system modeling
  • Markov analysis
  • Queueing theory
  • Simulation tools and techniques for large-scale systems
  • Mechanism design
  • Distributed virtual currency and smart contracts

Human-behavior-aware network systems

  • Automation of hazard area estimation and evacuation guidance
  • Crowd guidance for congestion alleviation
  • Navigation for people with walking difficulty
  • Delay tolerant networking

Ultra-scalable Blockchain technology

  • Stochastic modeling and analysis of the fork mechanism of blockchains
  • P2P networking technologies for high-speed block synchronization
  • Block generation based on advanced data structure
  • Innovative applications of highly-scalable blockchain technologies

Network optimization

  • Next generation networks
  • Cloud computing
  • Controllable P2P contents distribution systems
  • Network function virtualization (NFV) networks
  • Mathematical optimization
  • Game-theoretic approach

IoT security

  • Blockchain-based access control
  • Physical layer security-based secure wireless communications

Machine Learning for Networking

  • Automated network operations
  • Topology-aware network optimization
  • Network function virtualization
  • Graph neural networks for networking
  • Deep reinforcement learning for networking

Algorithms for trustworthy AI

  • Issues in current AI: reliability, fairness, diversity
  • Vast and complex search space
  • Compressed data structures such as zero-suppressed binary decision diagrams (ZDDs)
  • Applications: network reliability evaluation, fair evacuation planning and political redistricting, enumerating diverse solutions

Key Features

The Large-Scale Systems Management Lab research aims to develop mathematical modeling and simulation techniques for design, control and architecture of large-scale systems such as computer/communication networks, with which the resulting systems achieve high performance, low vulnerability and highly efficiency energy. Our research focus is on network-science oriented design frameworks, fundamental technologies and highly qualified services, particularly for large-scale computer/communication network systems. The laboratory was established in June 2012, and we welcome students from abroad who have strong interest in theories and simulation skills for designing smart services over large-scale complex systems including Blockchains, data centers, cognitive radio networks, and energy-harvesting networks.