Block-switched%20Networks:%20A%20New%20Paradigm%20for%20Wireless%20Transport

1
Block-switched NetworksA New Paradigm for
Wireless Transport

• Ming Li, Devesh Agrawal, Deepak Ganesan and Arun
  Venkataramani
• presented by zhen qin, marcel flores

2
Motivation
3
How TCP works

• E2E rate control is error-prone

4
How TCP works

• E2E retransmissions are wasteful

5
How TCP works

• Link layer ARQ

6
How TCP works

• Link-layer ARQs/backoffs hurt TCP rate control

7
Hop Contribution

• A clean-slate design and implementation of a
  wireless transport protocol
• Using reliable per-hop block transfer as a
  building block

8
Hop Design
9
Reliable Block Transfer

• Structure of a block and Timeline of TCP vs Hop

10
Virtual Retransmission

• Exploit caching at intermediate node
• Hop routers store all packets they overhear
• Transmit BSYN packet when block dropped

11
Backpressure

• Limits outstanding blocks per-flow at forwarder

12
Backpressure

• Limits outstanding blocks per-flow at forwarder

13
Ack Withholding

• acknowledging only one BSYN packet
• withholding BACKs to other concurrent BSYN packet
  until outstanding block complete
• Mitigating impact of hidden terminals

14
Micro-block Prioritization

• Senders piggybacks small blocks to BSYN
• Receivers prioritizes small blocks BACK
• Low delay for small blocks

15
Evaluation

• A 20 node wireless mesh testbed
• Linux 2.6 kernel
• 802.11a/b/g Athero/MadWiFi card
• Spread around the CS building

16
Comparisons

• End-to-end
• UDP
• TCP with CUBIC congestion control
• Hop-by-hop
• Hop-by-hop TCP
• TCP with backpressure
• DTN2.5
• Always subtract TCP setup time

17
Results

• Single hop microbenchmarks
• 100 Random links, transferred 10 MB file
• Shows robust performance on poor links

18
Graceful Degradation

• Sorted by TCP goodput
• Tried artificially dropping packets, examined
  goodput for different rates

19
Multi-Hop

• 100 random pairs - varying lengths
• TCP slightly inflated, OSLR only picks good links
  (Hop does well on lossy)

20
Hop Components

• Compare Hop against different versions of itself

21
Other Comparisons

• High Load
• WLAN access points
• Small file transfers
• Single hop transfer delay
• Multi hop transfer delay
• Robustness to partitions
• Affect on VOIP
• Network and link layer dynamics
• 802.11g

22
Conclusion for Hop vs. TCP

• It looks like Hop would do well in this
  environment
• Blocks seem effective as a paradigm
• Do not claim TCP cant be made better
• Instead, have shown starting from the ground up
  has potential