![Page 1: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/1.jpg)
Tony McGregor
RIPE NCC Visiting Researcher
The University of [email protected]
Optimising Path DiscoveryDoubletree and Paris Traceroute
a b c d e
2 e
3 e
4 e
d
e
b
1 e
c
![Page 2: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/2.jpg)
About Finding The Path between Two Monitors
Traceroute
a b c d e
2 e
3 e
4 e
d
e
b
1 e
c
![Page 3: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/3.jpg)
Traceroute(the real story)
cb d f he ga
2 h
3 h
4 h
f
h
b
1 h
d
![Page 4: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/4.jpg)
Traceroute(the really real story)
cb d f he ga
b
1 h
b
1 h
b
1 h
2 h
d
2 h
d
![Page 5: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/5.jpg)
Traceroute(more reality)
a b
c
d e
f g
![Page 6: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/6.jpg)
Traceroute(But wait ... there's more)
RFC1918 addresses ICMP disabled ICMP filters False routes Load balancing Interface coalition ISPs hide topology
TTL 0 Tunnels
MPLS IPv6
![Page 7: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/7.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
Donnet B et al. “Efficient Algorithms for Large-Scale Topology Discovery” Sigmetrics 05
![Page 8: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/8.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 9: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/9.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 10: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/10.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 11: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/11.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 12: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/12.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 13: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/13.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 14: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/14.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 15: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/15.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 16: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/16.jpg)
Wasted Probes(doubletree)
S1
D1
D2
a b
d
c
S2
f
e
![Page 17: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/17.jpg)
Intra-monitor Redundancy
S1, a, b, c, D1
Stop set
S1
D1
D2
a b
d
c
S2
f
e
![Page 18: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/18.jpg)
Intra-monitor Redundancy
Stop set
S1, a, b, c, D1, e
S1
D1
D2
a b
d
c
S2
f
e
![Page 19: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/19.jpg)
Intra-monitor Redundancy
Stop set
S1, a, b, c, D1, e, d
S1
D1
D2
a b
d
c
S2
f
e
![Page 20: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/20.jpg)
Intra-monitor Redundancy
Stop set
S1, a, b, c, D1, e, d
S1
D1
D2
a b
d
c
S2
f
e
![Page 21: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/21.jpg)
Intra-monitor Redundancy
Stop set
S1, a, b, c, D1, e, d, D2
S1
D1
D2
a b
d
c
S2
f
e
![Page 22: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/22.jpg)
Extent of intra-monitor redundancy
Donnet et al. Estimate that in skitter 87% of probes were redundant.
skitter doubletree
![Page 23: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/23.jpg)
Inter-monitor redundancy
S1
D1
D2
a b
d
c
S2
f
e
![Page 24: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/24.jpg)
Inter-monitor redundancy
S1
D1
D2
a b
d
c
S2
f
e
![Page 25: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/25.jpg)
Inter-monitor redundancy
D1: a, b, c D2: a, b, d, e
global stop set
S1
D1
D2
a b
d
c
S2
f
e
![Page 26: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/26.jpg)
Inter-monitor redundancy
D1: a, b, c D2: a, b, d, e
global stop set
S1
D1
D2
a b
d
c
S2
f
e
![Page 27: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/27.jpg)
Inter-monitor redundancy
skitter doubletree
![Page 28: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/28.jpg)
Limitations
Doubletree finds fewer nodes that repeated traceroutes Path changes Load balancing
Global stop set requires communication Negates savings to some extent May be large if there are many destinations
Trade off between reducing intra- and inter- monitor redundancy
![Page 29: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/29.jpg)
Doubletree for DAR
The DAR project is investigating the design of a very large scale active measurement system Design should support in the order of 100.000
probes Want to find path to a small number of
targets from a large number of probes can't just probe double tree
A lot of redundancy so much potential for savings Need to share global stop set between many probes Global stop set will be smaller than for scamper
![Page 30: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/30.jpg)
Discovering Alternative Pathsload balancing and Paris traceroute
TCP traceroute study (July 2006, 5,000 addresses) 7 of 9 Tear one ISPs 17 of 64 top regional ISPs Over 80% of destinations appeared to have load
balancers in their path
Augustin et al. “Avoiding traceroute anomalies with Paris traceroute” Université of Pierre et Marie Curie
![Page 31: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/31.jpg)
Alternative Pathsnaïve traceroute
b d
e DS
c
a
![Page 32: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/32.jpg)
Alternative Pathsnaïve traceroute
b d
e DS
c
a
![Page 33: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/33.jpg)
Alternative Pathsnaïve traceroute
b d
e DS
c
a
![Page 34: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/34.jpg)
Alternative Pathsnaïve traceroute
b d
e DS
c
a
![Page 35: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/35.jpg)
Alternative Pathsnaïve traceroute
b d
e DS
c
a
![Page 36: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/36.jpg)
Alternative Pathsnaïve traceroute
d
e D
b d
e DS
c
c
S a
a
![Page 37: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/37.jpg)
Alternative Pathsnaïve traceroute
e D
b d
e DS
c
c
S a
a
![Page 38: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/38.jpg)
Load Balancing
Three types of load balancers Random (per-packet) Destination address based Flow ID based
Some or all of the classic 5-tuple TOS ICMP code and checksum
Traceroute deliberately varies flow information
Match TTL expired with probe packet
![Page 39: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/39.jpg)
Paris Traceroute
Goal: To discover all paths from a source to a destination Doesn't address per-destination load balancing
Create headers such that flow ID remains the same
Still need to match ICMP TTL expired packets to probes For UDP vary checksum
Requires manipulating the payload For ICMP vary sequence number and identifier
So that checksum remains constant For TCP vary the sequence number
Uses TCP half-open technique
![Page 40: Tony McGregor RIPE NCC Visiting Researcher tony.mcgregor@ripe.net The University of Waikato tonym@cs.waikato.ac.nz Optimising Path Discovery Doubletree](https://reader031.vdocuments.mx/reader031/viewer/2022013011/56649e4c5503460f94b415a7/html5/thumbnails/40.jpg)
Results
For a particular set of 5000 destinations from Université Pierre et Marie Curie
Loops 5.3% of traceroute paths 87% caused by per flow load balancing
Cycles 0.84% of traceroute paths 78% caused by per-flow load balancing
Diamonds 79% of traceroute paths 64% cause by per-flow load balancing