Study of a High Performance Seamless Parallel I/O System between Parallel Computers with Load Balancing Awareness
| Project/Area Number |
21700063
|
|---|
| Research Category |
Grant-in-Aid for Young Scientists (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
Computer system/Network
|
|---|
| Research Institution | Kinki University |
|---|
Principal Investigator |
|
|---|
| Project Period (FY) |
2009 – 2011
|
| Project Status |
Completed (Fiscal Year 2011)
|
| Budget Amount *help |
¥4,810,000 (Direct Cost: ¥3,700,000、Indirect Cost: ¥1,110,000)
Fiscal Year 2011: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2010: ¥650,000 (Direct Cost: ¥500,000、Indirect Cost: ¥150,000)
Fiscal Year 2009: ¥3,510,000 (Direct Cost: ¥2,700,000、Indirect Cost: ¥810,000)
|
|---|
| Keywords | ネットワークコンピューティング / 並列入出力 / PCクラスタ / PC クラスタ / MPI / MPI-IO / Two-Phase I/O / PVFS2 / トレースツール |
|---|
| Research Abstract |
Parallel I/O interface in the MPI standard named MPI-IO provides collective parallel I/O. We have developed two pipelined processing implementations based on an existing original implementation by using(1) multithreaded technique with the help of Pthreads and (2) asynchronous I/O APIs, and both implementations have outperformed the original implementation. Furthermore, we have addressed to utilize a multiple buffering mechanism to minimize idle times of each process, and it has outperformed the original one.
|
Report
(4 results)
Research Products
(23 results)
