Presentation is loading. Please wait.

Presentation is loading. Please wait.

後卓越子計畫報告 PLLAB 李政崑教授. Component Remoting Technology Map.

Similar presentations


Presentation on theme: "後卓越子計畫報告 PLLAB 李政崑教授. Component Remoting Technology Map."— Presentation transcript:

1 後卓越子計畫報告 PLLAB 李政崑教授

2 Component Remoting Technology Map

3 Research Result Streaming Support for Java RMI in Distributed Environment, C. C. Yang, Chung-Kai Chen, Yu-Hao Chang, Kai-Hsin Chung and Jenq- Kuen Lee, ACM International Conference on Principles and Practices of Programming In Java (PPPJ 2006), Mannheim, Germany, August 30 - September 1, 2006. Streaming Support for Java RMI in Distributed Environment Apply for a patent “ 提供遠端物件具備網路串流功能的機制 “ now.

4 Proposed Software Architecture of Streaming RMI Several important components are needed to support our mechanisms. Streaming data are pushed from server continuous buffer to client continuous buffer automatically. Streaming controller manages the content of continuous buffer. Streaming controller stores aggregated data in streaming buffer. The client application can consume the complete stream data from the streaming buffer. Continuous Buffer Continuous Buffer Stub RMI Client RMI Client Streaming Buffer Streaming Buffer Continuous Buffer Continuous Buffer Streaming Controller Streaming Controller RMI Server RMI Server Continuous Buffer Continuous Buffer Application Layer Streaming Layer RDMA-like Transportation

5 Features of Streaming Java RMI Pushing  The idea is the same as pre-fetching. Aggregation  This is for better manipulation of streaming data from multiple streaming servers. Forwarding  It provides bandwidth-sharing between clients.

6 Aggregation policy Notations  A set of streaming servers S = {s i | i = 1,.., n}  A set of data blocks D = {d j | j = 1,.., m}  For each streaming server s i : The supplying bandwidth b i of s i A set of data blocks that exists in s i : Blocks(s i ) The completeness of data in s i : Completeness(s i ) The amount of content: k i  The bandwidth requirement : Req(d j )  The bandwidth allocation table: BAT mxn

7 Scheduling Algorithm Weight evaluation and sorting Bandwidth allocation

8 Experiment Result 1 Compare the performances of standard RMI with streaming RMI To demonstrate the performance improvement brought by pushing mechanism

9 Experiment Result 2 Data overhead measurement  Overhead for sending a 5MB data stream

10 Simulation Conditions for Aggregation We observe the waiting time of each streaming task. Waiting time is defined as the time from a client issuing the request to the time ready for playback. Take the ratio of results using aggregation to those without aggregation. Bandwidth available -- (α) Completeness -- (β) Amount of content -- (γ )

11 Simulation Results (1)  Simulation A –α = 0.5, Number of Streams = 400  Simulation B –α = 0.5, Number of Streams = 200 While the number of streams decreasing, our algorithm can get better average waiting time than the algorithm without aggregation. Bandwidth available -- (α) Completeness -- (β) Amount of content -- (γ) (The lower the better)

12 Simulation Results (2)  Simulation C -α=0.5, Number of Streams= 100  Simulation D -α=0.1, Number of Streams= 100 Using different variable sets, the average waiting reduction will get better reduction. Bandwidth available -- (α) Completeness -- (β) Amount of content -- (γ)

13 On going research We will add mobility ability into our framework We will expend our framework by following SOA specification


Download ppt "後卓越子計畫報告 PLLAB 李政崑教授. Component Remoting Technology Map."

Similar presentations


Ads by Google