Works in progress

• Hardware Design of Simulator for Deterministic Biochemical Systems (Since '02)
  Biochemical models are defined by their internal chemical reactions written in forms of differential equations. For accelerating simulations, we are exploiting a system that auto-generates an optimal hardware for each target models by combining several subcircuits that calculate reaction rate functions.
  
  
• Hardware Design of Stochastic Biochemical Simulator (Since Sept. '03)
  Stochastic simulation algorithm (SSA) is a quite rigorous method for simulating various biochemical systems. Calculation scheme itself is very simple, but vast number of repetitive calculation is required to achieve reasonable results. In order to relax severe calculation cost, we are developing stochastic biochemical simulator using reconfigurable devices such as FPGAs. We are exploring possibilites for providing higher performance for scientific calculations with lower cost with use of reconfigurable devices.
  
  
• Designing SBML Software Processor (Since Sept. '04)
  We have started a project to support the auto-generation of an optimal circuit for each biochemcal models written in SBML. Purpose of this project is to facilitate the use of ReCSiP for biochemists.
  
  

• Acceleration of Homology Searches on FPGA (Since Sept. '06)
  - More details (Japanese Only)
  
  

• Design and Evaluation of Dependable Systems on Commercial Accelerator Boards (Since Sept. '06)
  Dependability is an imperative factor for providing computer systems that are reliable and safe. In recent years, computing resources are applied for designing redundant systems besides using them for obtaining high throughput. Also, bio-inspired approaches are now drawing attention for designing robust and adaptive systems. We are devising methods for dependable designs with higher flexibility and efficiency, and evaluating them with commercial accelerator boards.