multicast forwarding and application state scalability in the internet
DESCRIPTION
Multicast Forwarding and Application State Scalability in the Internet. Tina Wong Dissertation Seminar Computer Science Division University of California, Berkeley October 16, 2000. Challenge. - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/1.jpg)
1
Multicast Forwarding and Application State Scalability
in the Internet
Tina Wong
Dissertation SeminarComputer Science Division
University of California, BerkeleyOctober 16, 2000
![Page 2: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/2.jpg)
2
Challenge
“… in the long run, the biggest issue facing multicast deployment is likely to be the scalability of multicast forwarding state as the number of multicast groups increases.”
--Thaler and Handley 2000
The memory required to store multicast forwarding entries at a router with 32 interfaces is 1024 TB for IPv6, assuming 50% address space utilization
--Radoslavov, Govindan and Estrin 1999
![Page 3: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/3.jpg)
3
Outline
• Introduction, background, motivation• Multicast state scaling trends in Internet • Preference clustering protocol• Application-driven tunable reliability• Conclusions and future work
![Page 4: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/4.jpg)
4
IP Multicast
• Efficient point-to-multipoint delivery mechanism
• Packets travel on common parts of the network only once
S
R R R
R
R
![Page 5: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/5.jpg)
5
Multicast Routing
S
R R R
Broadcast
DVMRP• Per-source reverse shortest
path tree• Broadcast-and-prune• MBone
![Page 6: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/6.jpg)
6
Multicast Routing
S
R R R
Prune
DVMRP• Per-source reverse shortest
path tree• Broadcast-and-prune• MBone
![Page 7: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/7.jpg)
7
Multicast Routing
S
R R R
Forward Data
DVMRP• Per-source reverse shortest
path tree• Broadcast-and-prune• MBone
![Page 8: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/8.jpg)
8
Multicast Routing
• PIM-Dense Mode / Sparse Mode– Unidirectional shared tree– Explicit joins– Core location a problem
• Core Based Trees (CBT)– Bi-directional shared tree– More optimal data paths– Few routing vendors support
![Page 9: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/9.jpg)
9
Multicast Forwarding State
• Router maintains membership state to achieve forwarding
• State scales linearly with number of concurrent groups
• No natural aggregation
• Number of concurrent multicast groups limited by router memory
• Heartbeat messages to maintain state incur processing costs
oif0 oif1 oif2
iif0
s = 0.0.0.0/0G = 224.0.1.2/32iif0, oif1, oif2
![Page 10: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/10.jpg)
10
Motivation
Lots of simultaneously active multicast groups on the Internet?
• Many small, group-based applications– Few participants form a single multicast group– E.g. internet video conferencing, games, events
notifications, etc
• Few large-scale applications– Lots of users form many multicast groups– E.g. Content delivery, stock quotes, DIS, etc
![Page 11: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/11.jpg)
11
Related Work
• Multicast state reduction– Leaky and non-leaky state aggregation– Tunneling in backbone (MPLS, DCM)– Non-branching state elim (DTM, REUNITE)
• Application-level multicast– End Sytsem Multicast, YOID, Scattercast
![Page 12: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/12.jpg)
12
Contributions
• Comprehensive analysis on multicast state– Understand scaling trends in the Internet– Predict future growth– Estimate potentials for reduction– Apply to network provisioning, protocol and
application design
• Mechanisms for network and end-host state scalability in large-scale applications– Interest-based content delivery– Application-driven loss recovery
![Page 13: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/13.jpg)
13
Outline
• Introduction, background, motivation• Multicast state scaling trends in Internet• Preference clustering protocol• Application-driven tunable reliability• Conclusions and future work
![Page 14: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/14.jpg)
14
Questions: Scaling Trends
Much research and engineering effort into making IP multicast widely deployed...
• How do multiplying peering agreements among parallel backbone networks affect multicast state scalability?
• How do rising subscriptions to individual applications increase multicast state?
• What are the state scaling properties when more and more applications use multicast?
![Page 15: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/15.jpg)
15
Questions: State Concentration
An intuition: multicast state scalability is most critical at “core” routers…
• How concentrated is multicast state at “core” routers?
• How much benefit from tunneling?
“Core”
![Page 16: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/16.jpg)
16
Questions: State Reduction
An intuition: delivery trees of sparse multicast groups tend to have large number of non-branching routers...
• How prominent are non-branching routers?
• Are these routers stateful?
S
R
R
R R
![Page 17: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/17.jpg)
17
Basic Model
Local state• Fraction of concurrent
multicast groups
True local state• Local state with only
multicast forwarding
Independent of address space size and number of concurrent groups
5 concurrent groupsLocal state = 2/5True local state = 1/5
oif0 oif1 oif2
iif0
![Page 18: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/18.jpg)
18
Methodology
• Simulations– Extends upon SGB package
• Parameters– Topology – Session density– Membership model
![Page 19: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/19.jpg)
19
Topology
• 4 AS graphs from Nov97 to Jan00– Connectivity among Internet autonomous
systems– Study multicast state at inter-domain level– Over 3 year timespan
• Mbone graph from Feb99– Study multicast state at intra-domain level
• Generated graphs– TIERS– Transit-stub
![Page 20: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/20.jpg)
20
Session Density
• Graphs have different number of nodes, from 1000 to 6474– Session density instead of absolute size– 0.1% to 0.9%, 1% to 9%, 10% to 90%– E.g., session with 0.1% density in AS-Jan00
with 6500 nodes involves 7 domains– E.g., session with 10% density in Mbone
with 4200 nodes involves 420 routers
![Page 21: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/21.jpg)
21
Membership Taxonomy
Topological Correlationwithin one group
Subscription correlationacross multiple groups
NO
NO
YES
YES
1
random distrclusters
2
affinity/disaffinity
3
interest4
layeredinterest
5 6
![Page 22: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/22.jpg)
22
Experiments
• For each experiment, fix topology, session density and membership model– (1) Pick a set of nodes with these parameters– (2) Build shortest path tree rooted at a random
node from this set– Repeat (1) & (2) 1000 times– Calculate local state and true local state on each
node in topology
• All combinations of parameters used, yielding 945 experiments and results!
![Page 23: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/23.jpg)
24
Answers: Scaling Trends 1
• How do multiplying peering agreements among parallel backbone networks affect multicast state scalability?
– More state at a handful of core routers– Offset by reduced state in majority of
routers
![Page 24: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/24.jpg)
25
Topological Properties
![Page 25: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/25.jpg)
26
Hypothesis
• In a more connected network– Trees have larger fanouts and shorter
heights– Only a few highly peered routers involved in
most concurrent multicast trees
![Page 26: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/26.jpg)
27
Hypothesis
• In a less connected network– Trees have smaller fanouts and taller
heights– Backbone routers share responsibility of
multicast forwarding -- “load balancing”?
![Page 27: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/27.jpg)
28
Path Lengths
![Page 28: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/28.jpg)
29
Node Degrees
AS-Nov97 MBone
![Page 29: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/29.jpg)
30
Past and Future ScalingTrends
• Implication– If Internet continues to evolve as it has been,
multicast memory requirements at most of border routers actually decline, all things remain equal
• Evidence– Peering increases for past 3 years– Maximum domain degree from 605 to 1459, roughly
50% expansion each year– Slight decrease in state for majority of nodes– Slight increase for rest of nodes
![Page 30: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/30.jpg)
32
Answers: Scaling Trends 2
• How do rising subscriptions to individual applications affect multicast state?
– Follows power law• fraction of stateful routers grows proportional to
some constant power of multicast group size
– Exponents within each membership for the Internet similar over past 3 years
– Predictive of future state growth
![Page 31: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/31.jpg)
35
Answers: State Concentration
• How concentrated is multicast state at the “core” routers?
– State concentration does not follow “10/90” rule even when session density is 0.1%
– Application-driven membership significantly impact state distribution and concentration
– Tunneling useful for multicast applications with very sparse and spread-out membership
![Page 32: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/32.jpg)
36
Answers: State Reduction
• How prominent are non-branching routers? Are these routers stateful?
– Very prominent– Up to 2 orders of magnitude reduction is
possible even at top 10% most stateful nodes
– Substantial even at 90% session density– Promising approach
![Page 33: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/33.jpg)
37
Outline
• Introduction, background, motivation• Multicast state scaling trends in Internet• Preference clustering protocol• Application-driven tunable reliability• Conclusions and future work
![Page 34: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/34.jpg)
38
Large-scale Applications
• Large-scale applications: many receivers, many sources, rich data types, UI
• Multicast uses one data stream to satisfy potentially heterogeneous receivers
• Lead to Preference Heterogeneity– Users differ in interest on application data– E.g. Content delivery, news dissemination,
stock quotes, network games, DIS, etc
![Page 35: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/35.jpg)
39
Example: Stock Quotes Service
...
AABC BACU CABL DAGR EACO FACO
www.StockCentral.com
Amy
INTCDELLCSCOMSFT
Bob
AAPLAMZNEWEBMSFTGABCQCOM
Cathy
PWBCSISIYHOO
...
![Page 36: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/36.jpg)
40
Example: Network GamesA player's position in virtual environmentdrives its preferences on entity updates
![Page 37: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/37.jpg)
41
Preference Heterogeneity
• Assign each logical data stream a unique multicast address ?
+No superfluous data
–Multicast routing state scalability
–Multicast address allocation and scarcity
–End-host connection maintenance
• 100% reliability not necessary– Different levels of reliability desired– Help to reduce NACK implosion
![Page 38: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/38.jpg)
42
The Clustering Concept
completeheterogeneity
completesimilarity
UNICAST MULTICAST
CLUSTER
approximately similar sources and receivers into like groups
many smallgroups
![Page 39: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/39.jpg)
43
Preference Clustering Protocol
• Clustering algorithm– On-line and adaptive to changes in preferences– Customizable to different application and data
types
• Signaling protocol– Coordinate clustering within an application– Scalable, fault tolerant and reliable through
decentralization, soft state and sampling
• API • Detailed evaluation
![Page 40: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/40.jpg)
44
App-Level Tunable Reliability
• Consider application semantics in loss recovery decisions– Meta-data to describe data content– Temporal: statistics on update frequency– Semantic: magnitude or importance of
change– Policy-driven by individual receivers
![Page 41: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/41.jpg)
45
Outline
• Introduction, background, motivation• Multicast state scaling trends in Internet• Preference clustering protocol• Application-driven tunable reliability• Conclusions and future work
![Page 42: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/42.jpg)
46
Conclusions
• Comprehensive study on multicast state scalability– Scaling trends confirmed with past 3 years– State distribution and concentration– Potentials for reduction
• Mechanisms to accommodate problem for large-scale applications– Customizable and adaptive preference
clustering protocol– Tunable reliable multicast protocol
![Page 43: Multicast Forwarding and Application State Scalability in the Internet](https://reader035.vdocuments.site/reader035/viewer/2022081603/56813f57550346895daa2091/html5/thumbnails/43.jpg)
47
Future Directions
• Compare and contrast methodologically IP multicast and application-level multicast– Params: Topology, session density,
membership– Apps: Few-to-few, one-to-many– Metric: Bandwidth, latency, complexity, etc
• Placement of service agents in Internet– Spawning of new agents – Coalescing based on topology, user
population, network measurements