Efficient Data Dissemination in OceanStore

1
Efficient Data Dissemination in OceanStore
Puneet Mehra and Satrajit Chatterjee
CS 262 Advanced Topics in Computer Systems
For more info http//www.cs.berkeley.edu/satraji
t/cs262/
Adaptation Mechanism

1. Each node periodically probes a subset of its
   siblings and grandparent. (B is probing in the
   picture)
2. A node switches parents if it can get a 10
   improvement in a certain metric (eg latency or
   bandwidth). (B has switched parents to C).

Exploiting Network Topology
D-Tree Algorithms
Simulation Framework
Baseline Naïve All receivers connect to root
(Inner Ring) Tapestry Nodes utilize Tapestry to
connect to the nearest node which has the data
object Topologically Aware Transit-Stub
Transit Nodes connect to root, stub nodes to
their closest transit, and all others to their
closest stub. Closest Propagation
delay Transit-Only Stub-Only Adaptive Adaptive
Tapestry Adaptive Transit-Stub - Adapt for
Latency, Bandwidth or Combination

• Simulator
• Built on top of ns-2
• Used a static Tapestry implementation
• Network Characteristics
• 196 nodes. 4 Transit Nodes. 24 Stub nodes. Domain
  size varied.
• 10 Mb/s links between nodes in stub
• 45 Mb/s links between stub and transit nodes
• 100 Mb/s links between all transit nodes
• D-tree Architecture
• D-tree structure maintained through heartbeats
• Cycles in Tree were detected using timestamp from
  root.
• Workload
• Best effort model for data delivery
• Single source streaming application. 3000 byte
  data packet per second. Fraction of nodes joined
  tree in random order.
• Multiple producer model. 3000 byte packet every 4
  seconds

• Model Internet as transit-stub network.
• Data goes through stub nodes into stub domain.
  Transit nodes pass data between domains.
• Placing overhead Replicas at transit or stub
  nodes can decrease network utilization.