| Project/Area Number |
18500051
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (C)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Computer system/Network
|
|---|
| Research Institution | Shiga University |
|---|
Principal Investigator |
MORI Masaaki Shiga University, Dept. of Economics, Professor (90024942)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
SHIBATA Naoki Shiga University, Dept of Economics, Associate Professor (40335477)
|
|---|
| Project Period (FY) |
2006 – 2007
|
| Project Status |
Completed (Fiscal Year 2007)
|
| Budget Amount *help |
¥3,980,000 (Direct Cost: ¥3,500,000、Indirect Cost: ¥480,000)
Fiscal Year 2007: ¥2,080,000 (Direct Cost: ¥1,600,000、Indirect Cost: ¥480,000)
Fiscal Year 2006: ¥1,900,000 (Direct Cost: ¥1,900,000)
|
|---|
| Keywords | P2P Networks / Video Multicast / Transcode / QoS / PlanetLab / PlanetLab / ネットワーク |
|---|
| Research Abstract |
Recently, P2P video delivery on the Internet is becoming popular as several services like Joost and Zatoo has been started. The advantage of P2P video delivery is that it can reduce server and network loads while providing excellent service quality to the users. These already started services only provide delivery of one fixed quality of video, but there are many types of computing terminals. Thus, we need efficient video delivery methods for many terminals with different computation powers, display sizes and bandwidths. There are already several research efforts on simultaneously delivering videos to multiple users with different quality requirements. In our research project, we have proposed a video delivery method named MTcast which achieves efficient video broadcast to multiple heterogeneous users by relying on user peers to transcode and forward video. There are several advantages of MTcast. (i) it can provide a video whose quality is very close to the request of each user to each user ; (ii) it can withstand with node failure and abrupt departure ; (iii) the delivery delay is proportional to the logarithm of the number of nodes ; and (iv) newly joined nodes need not to wait long until it first receives the video stream. MTcast constructs a tree whose root is the sender of the original video content as a variation of a perfect n-ary tree, where peers with higher quality requirements are located near the root of the tree. The parent-child relationship between peers is decided according to bandwidth and topology of physical path. We also constructed a prototype system of MTcast system using Java language, which can be used on PlanetLab, and validated practicability and effectiveness of MTcast in the real Internet environment. With the prototype system, we confirmed that MTcast achieves short start-up latency, small channel switching time, and small recovery time from peer leaving/failure
|
