Lecture 5: FPGA Routing September 17, 2013 ECE 636 Reconfigurable Computing Lecture 5 FPGA Routing.
Lecture Routing ExOR
Transcript of Lecture Routing ExOR
-
7/29/2019 Lecture Routing ExOR
1/43
ExOR: Opportunistic Multi-Hop Routing
for Wireless Networks
Sanjit Biswas and Robert Morris
M.I.T. Computer Science and Artificial Intelligence Laboratory
Presented by Deepak Bastakoty
With slides from :
Sanjit Biswas, Robert Morris, (MIT)
Saurabh Gupta (WPI)
Yao Zhao (Northwestern)
-
7/29/2019 Lecture Routing ExOR
2/43
Multi-Hop Wireless Networks
Dense 802.11b-based mesh, all sorts of loss rates
Goal is efficiency and high-throughput
Gateway
s
2km
2km
Gateway
-
7/29/2019 Lecture Routing ExOR
3/43
packet
packet
packet
The Traditional View
Identify a route, forward over links
Use link level retransmissions
src
A B
dst
C
packet
packet
-
7/29/2019 Lecture Routing ExOR
4/43
How Radios Actually Work
Every packet is broadcast
src
A B
dst
C
1 2 3 4 5 6
12 3 6
3 51
42 3
4 5 6
1 2 4 5 6
No such thing as a link
-
7/29/2019 Lecture Routing ExOR
5/43
packet
packet
packet
packet
packet
packet
ExOR: Exploiting the Insight
src
A B
dst
C
packet
packet
packet
Figure out which nodes rxd broadcast
Node closest to destination forwards
-
7/29/2019 Lecture Routing ExOR
6/43
packet
packet
packet
packet
packet
packet
ExOR: Exploiting the Insight
src
A B
dst
C
packet
packet
packet
Figure out which nodes rxd broadcast
Node closest to destination forwards
-
7/29/2019 Lecture Routing ExOR
7/43
ExORs Assumptions
1. Many receivers hear every broadcast
2. Gradual distance-vs-reception tradeoff
3. Receiver losses are uncorrelated
src
A B
dst
C
-
7/29/2019 Lecture Routing ExOR
8/43
1. Multiple Receivers per
Transmission
Broadcast tests onrooftop network
Source sendspackets at max rate
Receivers recorddelivery ratios
Omni-directionalantennas
Multiple nodes inradio range
1km
S
100%
75%
50%
25%
0%
-
7/29/2019 Lecture Routing ExOR
9/43
2. Gradual Distance vs. Reception
Tradeoff
Wide spread of ranges, delivery ratios
Transmissions may get lucky and travel long distances
Distance (meters)
DeliveryRatio
Same Source
-
7/29/2019 Lecture Routing ExOR
10/43
3. Receiver Losses are Uncorrelated
Two 50% links dont lose the same 50% of packets
Losses not due to common source of interference
Example Broadcast trace:
Receiver 1 (38%):
Receiver 2 (40%):
Receiver 3 (74%):
Receiver 4 (12%):
-
7/29/2019 Lecture Routing ExOR
11/43
Extremely Opportunistic Routing (ExOR)
Design Goals
Ensure only one receiver forwards the packet
The receiver closest to the destination should forward
Lost agreement messages may be common
Lets not get eaten alive by overheads
-
7/29/2019 Lecture Routing ExOR
12/43
N1 N3 N5 N7N6N2 N4 N8S D
Traditional Path
Traditional routing must compromise between hops to choose ones that arelong enough to make good progress but short enough for low loss rate
With ExOR each transmission may have more independent chances of beingreceived and forwarded
It takes advantage of transmissions that reach unexpectedly far, or fallunexpectedly short
How ExOR might provide more throughput
-
7/29/2019 Lecture Routing ExOR
13/43
Traditional routing: 1/0.25 + 1 = 5 transmissions
ExOR: 1/(1 (1 0.25)4) + 1 = 2.5 transmissions
Assumes independent losses
N1
src dst
N2
N3
N4
How ExOR might provide more throughput (contd..)
-
7/29/2019 Lecture Routing ExOR
14/43
How often should ExOR run?
- Per packet is expensive- Use batches
Who should participate in the forwarding?
- Too many participants cause large overhead
When should each participant forward?
- Avoid simultaneous transmission
What should each participant forward?
- Avoid duplicate transmission
How and When does the process complete?
- Identify the convergence of the algorithm
ExOR: Protocol
Jump Ahead
-
7/29/2019 Lecture Routing ExOR
15/43
Who should participate?
The source chooses the participants (forwarder list) using ETX-likemetric
- Only considers forward delivery rate
The source runs a simulation and selects only the nodes whichtransmit at least 10% of the total transmission in a batch
- A background process collects ETX information via periodic link-stateflooding
-
7/29/2019 Lecture Routing ExOR
16/43
When should each participant forward?
Forwarders are prioritized by ETX-like metric to the destination
Receiving nodes buffer successfully received packets till the end ofthe batch
The highest priority forwarder transmits from its buffer when the batch
ends- These transmissions are called the nodes f ragmentof the batch
The remaining forwarders transmit in prioritized order
Question: How does each forwarder know it is its turn to transmit
- Assume other higher priority nodes send for five packet durationsif not hearing anything from them
-
7/29/2019 Lecture Routing ExOR
17/43
What should each participant forward?
Packets it receives yet not received by higher priority forwarders
Each packet includes a copy of the senders batch map, containing the
senders best guess of the highest priority node to have received each
packet in the batch
Question: How does a node know the set of packets received by
higher priority nodes?
- Using batch map
-
7/29/2019 Lecture Routing ExOR
18/43
How and When does the process complete?
If a nodes batch map indicates that over 90% of the batch has been
received by higher priority nodes, the node sends nothing when its
turn comes
When ultimate destinations turn comes to send, it transmits 10
packets including only its batch map and no data
Question: How is the remaining 10% data delivered?
- Using traditional routing
-
7/29/2019 Lecture Routing ExOR
19/43
Who Received the Packet?
Slotting prevents collisions (802.11 ACKs aresynchronous)
Only 2% overhead per candidate, assuming 1500 byteframes
payload ACK
payload ACK1cand1
src dest
cand2 cand3src ACK2 ACK3
src cand1
cand2
cand3
src dest
Standard unicast 802.11 frame with ACK:
ExOR frame with slotted ACKs:
-
7/29/2019 Lecture Routing ExOR
20/43
Slotted ACK Example
Packet to be forwarded by Node C But if ACKs are lost, causes confusion
payloadD C BA
A D
ACK
C
ACK
B
A B C D
X
-
7/29/2019 Lecture Routing ExOR
21/43
Agreeing on the Best Candidate
A: Sends frame with (D, C, B) as candidate set
A B C D
C: Broadcasts ACK C in second slot (not rxd by D)D: Broadcasts ACK D in first slot (not rxd by C, A)
B: Broadcasts ACK D in third slot
Node D is now responsible for forwarding the packet
XX
X
-
7/29/2019 Lecture Routing ExOR
22/43
ExOR: Packet Format
-HdrLen & PayloadLen indicate size of ExOR header and
payload respectively
-PktNum is current packets offset in the batch, corresponding
to the current batch-map entry
-FragSz is size of currently sending nodes fragment (in
packets)
-FragNum is current packets offset within the fragment
-FwdListSise is is number of forwarders in list
-ForwarderNum is current senders offset within the list
-Forwarder List is copy of senders local forwarder list
-Batch Map is copy of sending nodes batch map, where each
entry is an index into Forwarder List
-
7/29/2019 Lecture Routing ExOR
23/43
Transmission Timeline for an ExOR transfer
N24 not able tolisten to N5.
N8 does
not send
N17 might have
missed somebatch-maps
-
7/29/2019 Lecture Routing ExOR
24/43
Preliminary Concept Evaluation
Strengths ExOR is nimble Efficient in total number of packet transmissions
Weaknesses Requires (partial) link-state graph Candidate selection is tricky
Requires changes to MAC
-
7/29/2019 Lecture Routing ExOR
25/43
1 kilometer
65 Roofnet node pairs
-
7/29/2019 Lecture Routing ExOR
26/43
Throughput (Kbits/sec)
1.0
0.8
0.6
0.4
0.2
0
0 200 400 600 800CumulativeFractionofNodeP
airs
ExOR
Traditional
ExOR: 2x Improvement in throughput
Median throughputs: 240 Kbits/sec for ExOR,
121 Kbits/sec for Traditional
Figure 8: The distribution of throughputs of ExOR and traditional routing between the 65 node
pairs. The plots shows the median throughput achieved for each pair over nine experimental
runs.
-
7/29/2019 Lecture Routing ExOR
27/43
25 Highest throughput pairs
Node Pair
Throughput
(Kbits/sec)
0
200
400
600
800
1000 ExOR
TraditionalRouting
1 Traditional Hop
1.14x
2 Traditional Hops
1.7x
3 Traditional Hops
2.3x
For single hop pairs ExOR provides the advantage of lower probability of source
resending packets, as theres higher probability of source receiving the
destinations 10 batch-map packets
Figure 9: The 25 highest throughput pairs, sorted by traditional routing throughput. The bars show each pair's median
throughput, and the error bars show the lowest and highest of the nine experiments.
-
7/29/2019 Lecture Routing ExOR
28/43
25 Lowest throughput pairs
Node Pair
4 Traditional Hops
3.3x
Longer Routes
Throughput(Kbits/s
ec)
0
200
400
600
800
1000 ExOR
TraditionalRouting
Figure 10: The 25 lowest throughput pairs. The bars show each pair's median throughput, and the error bars show the
lowest and the highest of the nine experiments. ExOR outperforms traditional routing by a factor of two or more.
As number of node pairs increases along a route, the likelihood of increased choice
of forwarding nodes and multiple ways to goss ip back batch-maps, increases
With greater routing length ExOR is able to take advantage of asymmetric links also
-
7/29/2019 Lecture Routing ExOR
29/43
Retransmissions affected by selection of hops
Traditional routing has to select the shortest path whichresults in compromise on selecting drop probability, thus
increasing the number of transmissions
ExOR has no limitations on number of nodes, from the forwarder list, that
can forward the packet. Hence it uses both nodes closer to source and
nodes closer to destination, irrespective of their drop probability
Figure 11: The number of transmissions made by each node during a 1000-packet transfer from N5 to N24. The X axis
indicates the sender's ETX metric to N24. The Y axis indicates the number of packet transmissions that node performs.
Bars higher than 1000 indicate nodes that had to re-send packets due to losses.
-
7/29/2019 Lecture Routing ExOR
30/43
ExOR moves packets farther
Figure 12: Distance traveled towards N24 in ETX space by each transmission. The X axis indicates the dierence in ETX
metric between the sending and receiving nodes; the receiver is the next hop for traditional routing, and the highest-priority
receiving node for ExOR. The Y axis indicates the number of transmissions that travel the corresponding distance. Packets
with zero progress are not received by the next hop (for traditional routing) or by any higher-priority node (for ExOR).
Max. distance traveled by
hops in traditional routing
Distance traveled by
transmissions in ExOR
Big chunk of transmission, in
traditional routing, takes place
over shorter distances
Number of packets carried over
individual long distance links is
small
But cumulative
transmission is substantial
-
7/29/2019 Lecture Routing ExOR
31/43
ExOR moves packets farther
Delivery Probability decreases with distance
ExOR average: 422 meters/transmission
Traditional Routing average: 205 meters/tx
FractionofTra
nsmissions
0
0.1
0.2
0.6ExOR
Traditional Routing
0 100 200 300 400 500 600 700 800 900 1000
Distance (meters)
25% of ExOR transmissions
58% of Traditional Routing transmissions
-
7/29/2019 Lecture Routing ExOR
32/43
ExOR uses links in parallel
Traditional Routing
3 forwarders
4 links
ExOR
7 forwarders
18 links
-
7/29/2019 Lecture Routing ExOR
33/43
Batch Size
ExOr header grows with the batch size
Large batches work well for low-throughput pairs due to redundant batch map
transmissions
Small batches work well for high throughput pairs due to lower header
overhead
-
7/29/2019 Lecture Routing ExOR
34/43
Critical Analysis
Static No mobility
Small Scale Tens of nodes
Dense network- Maybe Only Rooftop Networks
File downloading application No voice, maybe not web (No reliable guarantee)
No Cross Traffic
-
7/29/2019 Lecture Routing ExOR
35/43
Static
Knowing the whole topology
In a mobile network, this is expensive
EXT is costly Measure link states of all possible links
Route change
During a batch, route may change
-
7/29/2019 Lecture Routing ExOR
36/43
Small Scale
Knowing the whole topology In a mobile network, this is expensive
EXT is costly Measure link states of all possible links
Large overhead in ExOR packet header All the forwarders are included in ExOR header Long vain waiting of forwarding timer The larger the network, the longer the average distance between
S and D, the more forwarders in the list
Traditional header (24~48 -> 8 if AODV) ExOR header (44~114 for 38-node network)
-
7/29/2019 Lecture Routing ExOR
37/43
Dense Networks
S
A
B
C
D
EX
-
7/29/2019 Lecture Routing ExOR
38/43
Dense Networks
-
7/29/2019 Lecture Routing ExOR
39/43
More Critical AnalysisYao Zhao (Northwestern)
No TCP and hence proxy
Voice Jitter
Web service Is batch good? May introduce large delay
Large file download Best for ExOR
-
7/29/2019 Lecture Routing ExOR
40/43
Cross TrafficYao Zhao (Northwestern)
Forwarding timer
Give higher-priority nodes enough time to send?
Assume 5 packets sent if a node cannot hear another node with
higher priority hard to justify heuristic. Also, forwarders couldbe consistently mutually inaccessible
802.11 use CSMA/CA, competition based MAC
If there is cross traffic, hard to estimate the transmission time ofother nodes
-
7/29/2019 Lecture Routing ExOR
41/43
Unfair Comparison ?Yao Zhao (Northwestern)
Bad Traditional routing (DSR) Dont think about link state changing Long packet header Send the entire file to the next node before the next node
starts sending
Bad MAC Selection Retransmit packet if ACK is lost Why not packet train?
A Paper in 2005 compared to some works before 1999 ?
-
7/29/2019 Lecture Routing ExOR
42/43
Questions?
-
7/29/2019 Lecture Routing ExOR
43/43
Acknowledgements
Many sketches, animated-diagrams, as well as some text have been sourced from the followingmaterials-
Course material on Net Centric Systems taught at TECHNISCHE UNIVERSITT DARMSTADT
Presentation on A High Throughput Route-Metric for Multi-Hop Wireless Routing by Eric Roznerof University of Texas, Austin
Presentation on ExOR: Opportunistic Multi-Hop Routing for Wireless Networks, by Sanjit Biswasand Robert Morris at Siggcomm
ExOR: Opportunistic Multi-Hop Routing for Wireless Networks - Sanjit Biswas and Robert Morris
Presentation on ExOR: Opportunistic Multi-Hop Routing for Wireless Networks, by Avijit ofUniversity of California, Santa Barbara
Presentation on ExOR: Opportunistic Multi-Hop Routing for Wireless Networks, by Yu Sun ofUniversity of Texas, Austin
Presentation on ExOR: Opportunistic Multi-Hop Routing for Wireless Networks, by GauravGupta, University of Southern California
Presentation on ExOR: Opportunistic Multi-Hop Routing for Wireless Networks, by Ao-Jan Su,Northwestern University