| Budget Amount *help |
¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
Fiscal Year 2014: ¥1,430,000 (Direct Cost: ¥1,100,000、Indirect Cost: ¥330,000)
Fiscal Year 2013: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
Fiscal Year 2012: ¥1,950,000 (Direct Cost: ¥1,500,000、Indirect Cost: ¥450,000)
|
|---|
| Outline of Final Research Achievements |
This research develops a method of global speculative multithreading based on the frequent program execution paths in order to highly speed up non-numerical programs by fully utilizing the potential performance of multi-core processors that recently have becomes widespread. For speculative multithreading on entire program codes globally, major essential technologies are developed by extending our previous path-base speculative multithreading method. The three major outcomes of this research are as follows: (1)development of a global path-based thread decomposition method that can parallelize entire program codes across the subroutine/procedure boundaries within program code, (2)development of a program profiler that can collect and analyze the global execution information of target programs, (3)development of an execution environment to support the global path-based speculative multithreading.
|
