Development of a CAD for Control Systems using Multicore Processors
| Project/Area Number |
17560396
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Control engineering
|
|---|
| Research Institution | Kyushu Institute of Technology |
|---|
Principal Investigator |
KOGA Masanobu Kyushu Institute of Technology, Faculty of Computer Science and Systems Engineering, Associate Professor (90251644)
|
|---|
| Project Period (FY) |
2005 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥4,160,000 (Direct Cost: ¥3,800,000、Indirect Cost: ¥360,000)
Fiscal Year 2007: ¥1,560,000 (Direct Cost: ¥1,200,000、Indirect Cost: ¥360,000)
Fiscal Year 2006: ¥1,100,000 (Direct Cost: ¥1,100,000)
Fiscal Year 2005: ¥1,500,000 (Direct Cost: ¥1,500,000)
|
|---|
| Keywords | CAD for Control Systems / Parallel Processing / Numerical Computation |
|---|
| Research Abstract |
In recent years, the CPU architecture makes an important transition. While the method which speeds up CPU by raising the operating frequency reaches the performance limits, the multicore CPU which contains multiple CPU cores in a package enters the mainstream for speed-up technique. However, the software which supports multicore CPUS is indispensable for enjoying the benefits of multicore CPUS. Therefore it is required to promote the software which support the parallel computation based on the technology. Meanwhile, since the advanced control theory imposes more time-consuming computation on us, there is a real need for the CAD for control systems which helps us to do the control system design more effectively at a low price. The purpose of this research is to propose a method for high speed computation by using multicore processors and to develop a CAD for control systems which improves the process of the design by utilizing multicore processors. We carried out the following.
1. We built a computer system which contains pseudo-octacore processor by using two quad-core processors, and examined the methods to elicit high performance of multicore processors. I found that the parallel processing by multithread achieves the high scalability for relatively-simple problems which require small memory.
2. We designed a framework for parallel computation which enables us to select the method for parallelization dynamically. The framework allows us to use the full power of CPU during the transitional phase of the architecture of CPU.
3. We developed a cad for control systems which utilizes the power of multicore processors, and applied it to the education of control systems and confirmed the validity of the software.
|
Report
(4 results)
Research Products
(101 results)
