Component Remoting Technology Component Remoting Map Research Result

Component Remoting. Technology Component Remoting Map

Research Result l 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. l Apply for a patent “提供遠端物件具備網路串流功能的機制 “ now.

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

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

Aggregation policy l Notations ¡ A set of streaming servers S = {si | i = 1, . . , n} ¡ A set of data blocks D = {dj | j = 1, . . , m} ¡ For each streaming server si: l The supplying bandwidth bi of si l A set of data blocks that exists in si: Blocks(si) l The completeness of data in si: Completeness(si) l The amount of content: ki ¡ The bandwidth requirement : Req(dj) ¡ The bandwidth allocation table: BATmxn

Scheduling Algorithm Weight evaluation and sorting Bandwidth allocation

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

Experiment Result 2 l Data overhead measurement ¡ Overhead for sending a 5 MB data stream

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

Simulation Results (1) ¡ ¡ l Simulation A –α = 0. 5, Number of Streams = 400 Simulation B –α = 0. 5, Number of Streams = 200 Bandwidth available -- (α) Completeness -- (β) Amount of content -- (γ) While the number of streams decreasing, our algorithm can get better average waiting time than the algorithm without aggregation. (The lower the better)

Simulation Results (2) ¡ ¡ l Simulation C -α=0. 5, Number of Streams= 100 Simulation D -α=0. 1, Number of Streams= 100 Bandwidth available -- (α) Completeness -- (β) Amount of content -- (γ) Using different variable sets, the average waiting reduction will get better reduction.

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