Analyzing Cross-layer Interaction in Overlay Networks Srinivasan Seetharaman September 2007
Jan 04, 2016
Analyzing Cross-layer Interaction
in Overlay Networks
Srinivasan SeetharamanSeptember 2007
22
Overlay Networks
Overlay networking helpsovercome functionalitylimitations of the Internetby forming a virtual networkover the native IP networkthat is:
Independent Customizable
33
Service Overlay Networks
Offer enhanced or new services by deploying intelligent routing schemes.
Relaying
Overlay link
44
Service Overlay Networks (contd.)
Characteristics: Nodes and links are persistent Perform overlay routing independent of native layer
routing Each Overlay path comprises one or more Overlay
links, based on a certain selfish objective
Many types of services can be offered Multicast (e.g. ESM, Overcast) QoS (e.g. OverQoS, SON) Security (e.g. DynaBone, SOS) Better routes (e.g. RON, Detour, X-Bone)… and much more
55
Cross-Layer Interaction
Performing dynamic routing at both overlay and native IP layers leads to:
Conflict due to mismatch or misalignment of routing objectives
Contention for limited physical resources
Functionality overlap (Both overlay layer and IP layer perform similar set of functions)
66
Cross-Layer Interaction (contd.)
These issues are amplified in the presence of
Selfish motives and aggressive behavior
Lack of information about other layer
Increasing impact ( #overlays |Traffic| )
77
Context
Native network topology Intra-domain Inter-domain
Attitude of native network Restrictive Oblivious Cooperative
88
Thesis Organization
. Oblivious Restrictive Cooperative
Intra-domain
Interaction with failure recovery
[Chapter III]
OR vs Load balancing
[Chapter V]
Overlay-friendly native
network[Chapter VII]
Inter-domain
OR vs Policy constraints[Chapter IV]
BT vs Load balancing
[Chapter VI]
INTERACTION BETWEEN FAILURE RECOVERY IN THE NATIVE
AND OVERLAY LAYERS
Chapter III
1212
Dual Rerouting
A
OVERLAY1
LAYER
NATIVE IP
LAYER
X
A
D
E F H
FH
GA
EC
C
BD
B
G
Each layer performs rerouting, with no knowledge of which layer leads to optimal restoration
Overlay reroutin
g
Native rerouting
Failure
1515
1. Overlap of functionality between layers causing Unnecessary route changes (esp when connectivity in
native network is very dynamic) Increased probing overhead
2. Unawareness of other layer’s decisions leading to Multiple simultaneous failures
3. Lack of flexibility and control
Downside to Dual Rerouting
1717
Tuning Dual Rerouting
Intra-domain (keepAlive-time = 1 sec, hold-time = 3 secs)
Dual Rerouting
Suppress overlay rerouting at 0.5 prob.
Defer overlay rerouting by 0.375 secs
Native-only rerouting
Average route changes 125.08% 101.59% 109.85% 1.567
Stabilized inflation 100% 108.32% 100% 1.202
Time when stable 113.7% 100.48% 107.33% 2.481
Peak inflation 114.22% 109.98% 110.73% 1.202
1919
Further Improving Recovery
Adjust the functioning of native layer:
Tuning the native layer keepAlive-time:
This produces the best tradeoff
between # of route changes,
stabilization time and recovery time
keep
Alive
-tim
e
keep
Alive
-tim
e
Tuning
INTERACTION BETWEEN OVERLAY ROUTING AND TRAFFIC ENGINEERING
Chapter V
2121
Repeated Non-Cooperative Game
Player1: Overlay Routing - Latency-optimized paths between nodes
Player2: Traffic Engineering - Optimal load-balanced routes
OverlayRoutingOverlay Link
Latencies
Overlay layertraffic
Overlay routes
TrafficEngineerin
g
Traffic on each overlay
link
Background traffic
Nativeroutes
Native linkdelays
TM
2626
Simulation Results
TEobjective
Overlayobjective
Overallstability
Round
2828
Our goal
.. is to propose strategies that
obtain the best possible performance for a particular layerwhile steering the system towards a stable state.
3131
Assume: Each layer has a general notion of the other layer’s selfish objective
Designate leader / follower
Operate leader such thata. Follower has no desire to change Friendlyb. Follower has no alternative to pick Hostile
Use history to learn desired action gradually.
Resolving Conflict – Our Approach
3636
Performance of Preemptive Strategies
We proposed four strategies that improve performance for one layer and achieve a stable operating point
Inflation factor= Steady state obj value with strategy
Best obj value achievedLeader Strategy Overlay TE
Overlay Friendly: Load-constrained LPHostile: Dummy traffic injection
1.0821.023
1.1221.992
Native Friendly: Hopcount-constrained LPHostile: Load-based Latency tuning
1.0271.938
1.1841.072
Inflation
Chapter VI
CROSS-LAYER INTERACTION OF PERFORMANCE-AWARE OVERLAY
APPLICATIONS
3939
BitTorrent File-Sharing
Popular file-sharing application that generates a large volume of Internet traffic
Characteristics: Service capacity increases with demand Centralized tracker regulating neighborhood Dynamically change active peers by
choke/unchoke protocol
4040
Comparison to Overlay Routing
AX
Data1
BY
Data2
A2
A1
B2B1A3
4141
BitTorrent Protocol
Tit-for-tat based incentive for uploading decisions Leecher: Unchoke the fastest uploaders Seed: Unchoke the fastest downloaders
Popular strategy to improve performance Optimistic unchoke: periodically look for faster
peers
4343
UnchokeRequest
BitTorrent Dynamics
B
X
A
L1
L2
ED
C
When bottlenecked on link L1
ChokeChoke
Load distribution across links is balanced
BitTorrent apps use all available b/wPeer TFT
Upload stats Download stats
Status Opt? Interested Pieces? Status My interest
Unchoke
4444
BitTorrent Dynamics
B
X
A
L1
L2
ED
C
When NOT bottlenecked on link L1
ChokeChoke
Load distribution across links is unbalanced
Unchoke
4545
Cross-Layer Interaction
Operating BitTorrent disrupts load balance and can result in high max util: This can be a problem for background traffic
Objective of native layer: Minimize ( Max Util.)
Objective of BitTorrent: Minimize (Overall finish time)
4949
Simulation Setup
Pick 100 ASes with 60% of them being non-stub ASes
5050
Simulation Setup
A
B
C
D
E
F
Generate 1-50 peers. Each associating with 1-3 torrents
F
L1
L2
5353
Simulation Performance Metrics
Max util across access links
= MaxaE ( Xa/Ca ), E is set of all links
X is the load, C is the capacity
Average finish time inflation of leechers
= 1/Nl ( ’i / i ) Nl is # of leechers
’ is finish time after strategy
Nl
i=1
5454
Reducing Impact – Traffic Engg
TE can be performed across inter-domain access links, in order to minimize (Max util)
Two flavors: Ingress / Egress
Determines which access link to pick for a certain destination or source IP address
5555
Reducing Impact – Traffic Engg (contd.)
Applying TE does not make much difference
Performance of a random AS (Focus AS)
5757
Alter certain BitTorrent protocol components or tune the associated parameters Minimal reduction of the max util Significant inflation of finish time
Specifically, we tried each of the following: Make peer selection random Make piece selection random Reduce duration of optimistic unchoking Freeze list of unchoked peers after 10 mins Tune the unchoking timers
Reducing Impact – Tuning BitTorrent
5959
Reducing Impact – Locality-awareness
Locality-based traffic management Give priority to peers within AS No change to BitTorrent clients Also try caching of requests sent outside AS
6060
Reducing Impact – Bandwidth Throttling
Limit bandwidth consumed by BT traffic Popular strategy among most ASes Involves lesser infrastructure cost
6161
Cross-layer Conflict
Native layer and BitTorrent layer constantly retaliate to other layer’s disruptive behavior
Peers deploy BitTorrent Protocol Encryption to avoid detection by native layer
We develop two “friendly” BitTorrent strategies that achieve a mutually agreeable point by reducing peak load
6262
A. Limit # of parallel downloads
The unchoking protocol and their timeline is uncoordinated across neighborsA
vera
ge
6363
A. Limit # of parallel downloads (contd.)
Reduces peak load from 0.94 to 0.852Finish time inflation is 1.1501
6464
B. Avoiding common neighbors
Problem is that two peers in same AS often contact same peer outside AS
Algorithm Perform bilateral info exchange where each
peer A finds out if its neighbor B has a neighbor C inside its own AS
If yes, toss a coin to determine if we can download from this peer B (Randomization acts as a load balancing strategy)
6565
B. Avoiding common neighbors (contd.)
Reduces max util from 0.94 to 0.85Finish time inflation is 1.187
ANALYZING INTER-DOMAIN POLICY VIOLATIONS IN OVERLAY
ROUTES
Chapter IV
7070
Inter-Domain Policy Violations
Two types of violations exist
Provider 1
Provider 1
Client 1Client 1
A
Client 2
Client 2
BClient
3Client
3C
Provider 2
Provider 2Peer
Legitimate native route
Overlay route
Exit violation
$
Transitviolation
$$
7575
Each transit violation has a corresponding exit violation upstream
Extent of exit policy violations in multihop paths
Measurement Results
Violation Type % paths
Next hop AS violated 72.05
Exit point violated 15.63
Total 87.68
7676
Policy Enforcement by Native Layer
As ISPs become aware of the negative impact of overlays and commence filtering, this leads to
drastic deterioration in overlay route performance commensurate with the number of ASes enforcing policy
7777
Overlay Service Provider (OSP) adopts a combination of the following strategies for achieving good legitimate paths:
1. Obtain transit permit from certain AS for $T
2. Add new node to certain provider AS for $N
3. Obtain exit permit from certain AS for $E
Resolving Conflict
7878
With no filtering,
Illustration of Mitigation Strategy
31
21
32
22
11 13
23
33
Cust-Prov relation
Peering relation
Transitviolation
AS hosting overlay node
Tier-1 provider
Tier-2 provider
Stub customer
7979
With filtering, we have no multi-hop paths
Illustration of Mitigation Strategy (contd.)
31
21
32
22
11 13
23
33
Cust-Prov relation
Peering relation
AS hosting overlay node
Tier-1 provider
Tier-2 provider
Stub customer
8080
Option 1: Add new overlay node to provider AS 22
Option 2: Obtain transit permit from stub AS 32
Illustration of Mitigation Strategy (contd.)
31
21
32
22
11 13
23
33
Cust-Prov relation
Peering relation
AS hosting overlay node
Tier-1 provider
Tier-2 provider
Stub customer
22
8181
Objective of Mitigation Strategy
For a certain budget, determine which ASes to obtain transit permit from to add new node to to obtain exit permit from
… so as to achieve the best possible gainGain = Native route latency – Overlay path
latency Native route latency
8383
Mitigation Results
When all permit fee = P, new node fee = N
Add new node
Permit
9090
Summary of Cross-Layer Interaction
Overlays offer valuable services needed by end-systems. But, lead to complex cross-layer interaction with potentially detrimental effects
Layer awareness is essential to reduce negative effects and to improve performance of both layers. We propose simple strategies that achieve this goal in an effective manner.
9191
Contributions of Thesis
Knobs for better control over the cross-layer interaction
Analysis and mitigation of the conflict in objective between native and overlay layers:
inter-domain: OR vs Policy enforcement intra-domain: OR vs Traffic engg
Ways to improve coexistence between BitTorrent file-sharing and native layer
Framework for network layer support of overlay services
9292
Future of Overlays
Overlays are essential as… Means for end-systems to collaborate Environment for testing future innovations (GENI) Architecture for Future Internet in the form of Network
Virtualization
Cross-layer interaction will affect performance. How best to design protocols and services in the future?
9393
Future Work
Need to address the network impasse. How to tune the network for
.. the new breed of Internet applications? (e.g., file sharing) …and new paradigms of communication? (e.g., wireless)
Which layer to implement a service at? For example, a service like multicast can be performed at both native layer and overlay layer!
Which layer to use for a particular scenario? How can the other layer support this service?