Reliable Multicast Using ReedSolomon and Gossip

1
Reliable Multicast Using Reed-Solomon and Gossip
Daniel Finchelstein and Noam Ivri Final
Presentation

• Continues Project to FEC using RS Project

2
Project Goals

• Search and determine the best approach to add
  reliability and performances to our
  Frame-Distribution system and implement it.
• Analyze the new system results Vs. first
  projects results.

3
Reed Solomon System
4
Same System With Gossip

• We inspire to get this graph

5
Recap on FEC using RS

• One lecturer (server) and many students (clients)

Server User Interface
FEC encoder
Server Application
Network
6
Client
from Data Channel
from Gossip Channel
7
What happens when we need more packets?

• Need more packets less then K were received
  after N were sent.
• Option1 (naïve) ask the server to send again
• Option2 change the (K,N) values for future
  packets
• Option3 get it from near-by clients!

8
Near-by clients?

• Reasonable assumption
• different clients loose different packets
• The clients have the resources to deal with low
  number of requests.

Using a background mechanism the clients can
exchange needed packets!
9
How does it work?

• Fixed redundant information set produced by the
  server (N-K rate)
• Clients join a Group Membership System (GMS)
• Gossip between sets of members
• Exchange needed packets from the redundant set.

10
Gossip, Design

• unsynchronized gossip rounds .
• In each round, 2 phases
• 1.detect message loss .
• 2 .correct such losses (runs only if needed) .

Members randomly choose other members ? send a
summary of their packets history ? ask for
packets they discover themselves to be lacking
11
Benefits of our approach

• Immediate correction of fault receives.
• Can work with low values of N,K.
• Does not bother the server.
• Extremely efficient for large number of clients.
  (epidemic rate of progress).
• Cost effective no need for complicated
  calculations.

12
Test Results

• The DFS without Gossip

13
Test Results

• The DFS with Gossip

14
A Note About the Tests

• The means in which we have tested the RS system
  are too simple for the Gossip system.
• A full scale simulation tool (like OPNET) should
  be used
• Performances

15
Project Summery

• Two separate system were created
• Reed-Solomon
• Gossip Channel
• Successful integration between the two system to
  create the Frame Distribution System
• Primary tests show the gossip enables better
  performances (under strict limitations)
• More tests should be done with a simulation tool

16
Acknowledgments

• Comnet Lab Yoram Orchen, Yoram Yihyie, Hai
  Vortman and all the wonderful stuff
• Gidon Gershinsky
• Justin Chapweske, Onion Networks' CEO
• Omer Gurewitz

17
Questions ?

• Thank You.