NIBEDITA MAULIK GRAND SEMINAR PRESENTATION OCT 21 st 2002.

Slides:



Advertisements
Similar presentations
CLive Cloud-Assisted P2P Live Streaming
Advertisements

Playback delay in p2p streaming systems with random packet forwarding Viktoria Fodor and Ilias Chatzidrossos Laboratory for Communication Networks School.
1 1 Chord: A scalable Peer-to-peer Lookup Service for Internet Applications Dariotaki Roula
SplitStream: High- Bandwidth Multicast in Cooperative Environments Monica Tudora.
Network Coding in Peer-to-Peer Networks Presented by Chu Chun Ngai
Image Indexing and Retrieval using Moment Invariants Imran Ahmad School of Computer Science University of Windsor – Canada.
296.3: Algorithms in the Real World
1 Accessing nearby copies of replicated objects Greg Plaxton, Rajmohan Rajaraman, Andrea Richa SPAA 1997.
Common approach 1. Define space: assign random ID (160-bit) to each node and key 2. Define a metric topology in this space,  that is, the space of keys.
PROMISE: Peer-to-Peer Media Streaming Using CollectCast Mohamed Hafeeda, Ahsan Habib et al. Presented By: Abhishek Gupta.
Efficient and Flexible Parallel Retrieval using Priority Encoded Transmission(2004) CMPT 886 Represented By: Lilong Shi.
June 3, 2015Windows Scheduling Problems for Broadcast System 1 Amotz Bar-Noy, and Richard E. Ladner Presented by Qiaosheng Shi.
Cis510: internet multimedia Papers to be presented today  Distributed Video Streaming over the Internet T Nguyen and A. Zakhor  On Peer-to-Peer Media.
Peer-to-peer Multimedia Streaming and Caching Service Jie WEI, Zhen MA May. 29.
PROMISE A Peer-to-Peer Media Streaming System Using CollectCast CPSC Presentation by Patrick Wong.
Topics in Reliable Distributed Systems Lecture 2, Fall Dr. Idit Keidar.
2010/3/81 Lecture 8 on Physical Database DBMS has a view of the database as a collection of stored records, and that view is supported by the file manager.
Distributed Multimedia Streaming over Peer-to-Peer Network Jin B. Kwon, Heon Y. Yeom Euro-Par 2003, 9th International Conference on Parallel and Distributed.
On Peer-to-Peer Media Streaming Dongyan Xu Mohamed Heffeda Susanne Hamrusch Bharat Bhargava 2002 International Conference on Distributed Computing Systems.
Understanding Mesh-based Peer-to-Peer Streaming Nazanin Magharei Reza Rejaie.
On Peer-to-Peer Media Streaming by Dongyan Xu, Mohamed Hefeeda, Susanne Hambrusch, Bharat Bhargava Dept. of Computer Science, Purdue University, West Lafayette.
P2P Course, Structured systems 1 Introduction (26/10/05)
Project Mimir A Distributed Filesystem Uses Rateless Erasure Codes for Reliability Uses Pastry’s Multicast System Scribe for Resource discovery and Utilization.
On-Demand Media Streaming Over the Internet Mohamed M. Hefeeda, Bharat K. Bhargava Presented by Sam Distributed Computing Systems, FTDCS Proceedings.
CS4432: Database Systems II
PROMISE: Peer-to-Peer Media Streaming Using CollectCast Presented by: Randeep Singh Gakhal CMPT 886, July 2004.
Fast Subsequence Matching in Time-Series Databases Christos Faloutsos M. Ranganathan Yannis Manolopoulos Department of Computer Science and ISR University.
Other Structured P2P Systems CAN, BATON Lecture 4 1.
 B+ Tree Definition  B+ Tree Properties  B+ Tree Searching  B+ Tree Insertion  B+ Tree Deletion.
UNC Chapel Hill M. C. Lin Point Location Reading: Chapter 6 of the Textbook Driving Applications –Knowing Where You Are in GIS Related Applications –Triangulation.
Database Management 9. course. Execution of queries.
1 Reading Report 5 Yin Chen 2 Mar 2004 Reference: Chord: A Scalable Peer-To-Peer Lookup Service for Internet Applications, Ion Stoica, Robert Morris, david.
Chapter 11 Indexing & Hashing. 2 n Sophisticated database access methods n Basic concerns: access/insertion/deletion time, space overhead n Indexing 
Content Addressable Network CAN. The CAN is essentially a distributed Internet-scale hash table that maps file names to their location in the network.
HUAWEI TECHNOLOGIES CO., LTD. Page 1 Survey of P2P Streaming HUAWEI TECHNOLOGIES CO., LTD. Ning Zong, Johnson Jiang.
Resource Addressable Network (RAN) An Adaptive Peer-to-Peer Substrate for Internet-Scale Service Platforms RAN Concept & Design  Adaptive, self-organizing,
Chord: A Scalable Peer-to-peer Lookup Service for Internet Applications.
Indexing and hashing Azita Keshmiri CS 157B. Basic concept An index for a file in a database system works the same way as the index in text book. For.
P2P Streaming Protocol (PPSP) Requirements draft-zong-ppsp-reqs-03.
Peer to Peer A Survey and comparison of peer-to-peer overlay network schemes And so on… Chulhyun Park
Sets of Digital Data CSCI 2720 Fall 2005 Kraemer.
1. Efficient Peer-to-Peer Lookup Based on a Distributed Trie 2. Complex Queries in DHT-based Peer-to-Peer Networks Lintao Liu 5/21/2002.
Session 1 Module 1: Introduction to Data Integrity
BATON A Balanced Tree Structure for Peer-to-Peer Networks H. V. Jagadish, Beng Chin Ooi, Quang Hieu Vu.
Peer-to-Peer Networks 03 CAN (Content Addressable Network) Christian Schindelhauer Technical Faculty Computer-Networks and Telematics University of Freiburg.
1 CSIS 7101: CSIS 7101: Spatial Data (Part 1) The R*-tree : An Efficient and Robust Access Method for Points and Rectangles Rollo Chan Chu Chung Man Mak.
Bootstrapped Optimistic Algorithm for Tree Construction
Multimedia Retrieval Architecture Electrical Communication Engineering, Indian Institute of Science, Bangalore – , India Multimedia Retrieval Architecture.
1 Plaxton Routing. 2 History Greg Plaxton, Rajmohan Rajaraman, Andrea Richa. Accessing nearby copies of replicated objects, SPAA 1997 Used in several.
Geethanjali College Of Engineering and Technology Cheeryal( V), Keesara ( M), Ranga Reddy District. I I Internal Guide Mrs.CH.V.Anupama Assistant Professor.
CS Spring 2010 CS 414 – Multimedia Systems Design Lecture 24 – Introduction to Peer-to-Peer (P2P) Systems Klara Nahrstedt (presented by Long Vu)
Database Applications (15-415) DBMS Internals- Part III Lecture 13, March 06, 2016 Mohammad Hammoud.
Data Integrity & Indexes / Session 1/ 1 of 37 Session 1 Module 1: Introduction to Data Integrity Module 2: Introduction to Indexes.
Chapter 25 Domain Name System.
Indexing Structures for Files and Physical Database Design
CS522 Advanced database Systems
Net 323 D: Networks Protocols
Accessing nearby copies of replicated objects
Database Applications (15-415) DBMS Internals- Part III Lecture 15, March 11, 2018 Mohammad Hammoud.
Spatial Online Sampling and Aggregation
Indexing and Hashing Basic Concepts Ordered Indices
Chapter 11 Data Compression
Reachability on Suffix Tree Graphs
Advance Database System
Applied Combinatorics, 4th Ed. Alan Tucker
Indexing 1.
2018, Spring Pusan National University Ki-Joune Li
Faculty Seminar Series Blockchain Technology
Switching Lemmas and Proof Complexity
Path Oram An Extremely Simple Oblivious RAM Protocol
Presentation transcript:

NIBEDITA MAULIK GRAND SEMINAR PRESENTATION OCT 21 st 2002

On Peer-to-Peer Media Streaming Purdue University P2P data streaming Supplying Peers stream media file to requesting peers Requesting peers play back and store media data during streaming session Requesting peers become supplying peers after streaming session Characteristics self growing, server less, heterogeneous in outbound bandwidth contribution, supplying peers/requesting peer relation is many to one Problems Media data assignment for a multi-supplier P2P streaming session (Optimal Media Data Assignment) Fast amplification of P2P streaming capacity (Fast System Capacity Amplification) Streaming model Each supplying peer participates in at most one P2P streaming session at any time Capacity of the P2P system is the total number of P2P streaming sessions simultaneously provided by the system The in-bound bandwidth of the requesting peer Pr is related to Ro i.e. the playback rate of the media data as Rin( Pr ) = Ro, The out-bound bandwidth of the supplying peers Ps has values Ro/2, Ro/4, Ro/8…..Ro/2 N classified as class-n peers for the out-bound bandwidth of Ro/2 n

OPTIMAL MEDIA DATA ASSIGNMENT Buffering Delay : Time interval between the start of the transmission and the start of the playback at Pr

OPTIMAL MEDIA DATA ASSIGNMENT(Algorithm) Assignment of first 2n segments is computed ; the assignment repeats itself every 2n segments for the rest of the media file. OTSp2p is optimal

FAST SYSTEM CAPACITY AMPLIFICATION

Distributed Admission Control Protocol DACp2p Each supplying peer individually decides in a probabilistic fashion with different probability value applied to different classes of requesting peers, whether or not to participate in a streaming session A requesting peer may send a ‘reminder’ to a busy supplying peer, reminding it not to elevate its admission preferences to requesting peers of classes lower than that of the requesting peer FAST SYSTEM CAPACITY AMPLIFICATION(Algorithm) DACp2p – Supplying Peers Ps maintains an admission probability vector Pr[i] = probability with which a class-I requesting peer with be supplied in case Ps is not busy in another streaming session The vector is computed as follows 1. If Ps is a class-k peer, for 1<=i<=k Pr[i] =1.0 & for k<i <=N Pr[i] = 1/ 2 i-k 2. If Ps has been idle, the probability vector will be updated after Tout. For k<i<N, Pr[i]=Pr[i]*2 3. If Ps just finished a streaming session, it will update it’s probability vector as follows - if during the session it did not receive any request from it’s favored class then K<i<=N Pr[i]=Pr[i]*2 - if it received at least 1 request from it’s favored class, however it was not granted because Ps was busy and the requesting peer left a ‘reminder’ and the class of that requesting peer was L then 1<=i<=L Pr[i]=1.0 and L<i<=N, Pr[i] = 1/2 i-k

FAST SYSTEM CAPACITY AMPLIFICATION(Algorithm) DACp2p – Requesting Peer - Pr obtains a list of M randomly selected candidate supplying peers by some P2P lookup mechanisms. The class of each candidate is also obtained - Pr obtains enough permissions from supplying peers such that 1. They are neither down nor busy 2. The are willing to provide streaming service 3. Their aggregated out-bound bandwidth offered is Rsum = Ro - Pr executes algorithm OTSp2p to compute media data assignment, triggers the participating supplying peers, and the session begins - If Pr cannot get enough permission it leaves ‘reminder to a subset W of the busy candidates. Members of W are selected as follows 1. The candidate currently favors the class of Pr 2. The aggregated out-bound bandwidth offer of the candidates in W is equal to Ro – Rsum - If Pr is admitted, it becomes a supplying peer when the streaming session is over, if Pr is rejected it will backoff for at least T bkf before making the request again

THE NUMERIC INDEXING FOR MUSIC DATA Chaoyang University of Technology, Taiwan Goal Efficient Content based retrieval of audio & music data To transform the music data into numeric forms and develop an index structure for effective retrieval Ensure scalability in an efficient audio/music data retrieval system Suffix tree for Music Indexing Constructed from a symbol with length m symbols, consists of m leaf nodes numbered from 1 to m Any two branches from a non-leaf node should be labeled with different symbols The number of each node points out the start position of the sub-string which consists of the symbols labeled from the root to this leaf node of the tree

NUMERIC INDEXING Let music data consist of n notes Do, Re, Mi ….etc each note represented by a music symbol ‘a’, ‘b’, ……., ‘n’ respectively Each music symbol is mapped onto integer values 0, 1, ……., n-1 For a music segment with m adjacent notes x1x2x3….xm, the integer value of each note is to be represented by P(xi), this segment can be transformed into a numeric value by the function v(m) =  P(xi) X n i-1 Modified R-tree to construct numeric index structure Each non-leaf node of the modified R-tree consists of the upper bound and lower bound of numeric values of the sub-tree under it Each leaf node of the R-tree stores the transformed value of music feature string and a linked list of target music Each entry in the linked list consist of music information and a pointer Each music information consist of the music ID in the database and the start position of the music segment in the music The pointer points to the next entry with the same transformed value

EXAMPLE FOR NUMERIC INDEXING

EXACT MATCHING The music query segment has to be found the exact same pattern in a melody from the music database e.g Music query segment ccdbb. Two sub-segments of length 4 will be ccdb & cdbb of value 1322 & 1132 Leaf node > (S2, 2), (S3, 1) Leaf node > (S2, 3), (S3, 4) The intersection yields {S2, S3} -> both music feature strings consist of segments ‘ccdb’ & ‘cdbb’ Check if the starting symbols of sub-segments found in each music feature string are adjacent each other as the sequence of query segment. Here sub-segments ‘ccdb’ and ‘cdbb’ starting on 2 nd and 3 rd symbols of S2 have the identical sequence as the query. Hence S2 is the target music

APPROXIMATE MATCHING Transform the music query string into numeric value Compare the value with the values in the leaf node Iff the difference between the transformed value and the value of leaf node matches one of the conditions listed in Table 1, there will be a target node found and it will be the target music