Rethinking Traffic Management: Design Optimizable Networks - PowerPoint PPT Presentation

Loading...

PPT – Rethinking Traffic Management: Design Optimizable Networks PowerPoint presentation | free to download - id: aad9a-NjhiM



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Rethinking Traffic Management: Design Optimizable Networks

Description:

Source Rate xi. User. Utility. Ui(xi) Source-destination pair indexed by i. source rate. Utility function represents user satisfaction and elasticity. routing matrix ... – PowerPoint PPT presentation

Number of Views:114
Avg rating:3.0/5.0
Slides: 69
Provided by: jenn90
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Rethinking Traffic Management: Design Optimizable Networks


1
Rethinking Traffic ManagementDesign Optimizable
Networks
  • Jiayue He
  • May 9th, 2008

2
Approach Theory Meets Practice
  • Using optimization theory
  • Analyze system properties
  • Derive protocols and architectures
  • Practical solutions
  • Understand limitations of todays protocols and
    architectures
  • Propose new protocols and architectures
    implementable using existing technology

3
Traffic Management
  • Determines traffic rate along each path
  • Supports multiple Internet applications

Traffic Management
4
Traffic Management Today
Operator (hours) Traffic Engineering
Evolved organically without conscious design
Routers (seconds) Routing Protocols
User (RTTs) Congestion Control
5
Goal Redesign Traffic Management
Resource Allocation between Multiple Traffic
Classes (Part III 18 min)
Throughput-Sensitive Traffic Analysis (Part I
10 min) Design (Part II 22 min)
Other Traffic Classes
6
Scope of This Talk
  • Single Internet Service Provider backbone
  • Control and visibility of network
  • Traffic management of aggregate flows
  • No inter-network economics

Multipath with flexible splitting
7
PART ONE
  • Can Congestion Control and Traffic Engineering Be
    at Odds?

8
Motivation
  • Congestion Control
  • maximize user utility
  • Traffic Engineering
  • minimize network congestion

Given routing Rli, how to adapt end rate xi?
Given traffic xi, how to perform routing Rli?
9
Goal Understand Interaction
Congestion Control
xi
Rli
Traffic Engineering
  • Understand system properties
  • Convergence to a stable value?
  • What is a reasonable overall objective?

10
Congestion Control ImplicitlyMaximizes Aggregate
User Utility
Source-destination pair indexed by i
aggregate utility
User Utility Ui(xi)
max.?i Ui(xi) s.t. ?i Rlixi cl var. x
Fair rate allocation among greedy users
routing matrix
source rate
Source Rate xi
Utility function represents user satisfaction and
elasticity
Kelly98, Low03, Srikant04
11
Traffic Engineering ExplicitlyMinimizes Network
Congestion
aggregate congestion cost
Links are indexed by l
Cost f(ul)
min. ?l f(ul) s.t. ul ?i Rlixi/cl var. R
To avoid bottlenecks in the network
ul 1
link utilization
Link Utilization ul
Cost function represents penalty for approaching
capacity and approximates average queuing delay
FortzThorup04
12
Model of Interaction
Assume the TCP session is between two customers
of same ISP
Congestion Control (RTTs) max ?i Ui(xi), s.t.
?i Rlixi cl
xi
Rli
Traffic Engineering (hours) min ?l f(ul), s.t.
ul ?i Rlixi/cl
f is controlled by the operators and can be
modified
13
Numerical Experiments
  • MATLAB experiments
  • Different topologies and capacity distributions
  • Benchmark
  • Observations
  • System converges
  • Utility gap exists between the joint system and
    benchmark

max. ?i Ui(xi) s.t. Rx c Var. x, R
14
Backward Compatible Design
  • Simulation of the joint system suggests that it
    is stable, but suboptimal
  • Gap reduced if we change f to red curve

Cost f
f(ul)
ul 1
0
Link utilization ul
15
Theoretical Results
Master Problem min. g(x,R) - ?iUi(xi)
??lf(ul)
Gauss-Siedel
Congestion Control argminx g(x,R)
Traffic Engineering argminR g(x,R)
  • Theorem the joint system model converges if
  • Replace the capacity constraint in congest
    control with a penalty function
  • Ui(xi) -Ui(xi) /xi holds for all TCP
    variants

16
Pros and Cons of Changing f
  • Pros
  • Backwards compatible
  • Can maximize aggregate user utility
  • Cons
  • Creates bottleneck links
  • Fragile to high volume traffic bursts
  • Motivation for redesign in Part II

17
Contributions and Related Work
  • Related Work
  • Separate analysis of CC and TE
  • Use congestion price as link weights
    (WangLiLowDoyle05, HeChiangRexford06)
  • Contributions
  • Modeled the interaction between CC/TE
  • Studied the interaction
  • Proposed backward compatible design

18
PART TWO
  • TRUMP TRaffic-management Using Multipath
    Protocol

Joint work with Maayan Bresler and Martin Suchara
19
Motivation for Redesign
  • Shortcomings of todays traffic management
  • Congestion control assumes routing is fixed
    traffic engineering assumes traffic is inelastic
  • Traffic engineering occurs at the timescale of
    hours, slower than traffic shifts
  • Not taking full advantage of path diversity
  • Goal redesign traffic management
  • from scratch using optimization tools

20
Top-down Redesign
Problem Formulation
Optimization decomposition
Distributed Solutions
Compare using simulations
TRUMP algorithm
Translate into packet version
TRUMP Protocol
21
A Balanced Objective
max. ?iUi(xi) - w?lf(ul)
Penalty weight
Congestion Control Maximize throughput Generate
bottlenecks
Traffic Engineering Minimize congestion Avoid
bottlenecks
22
Topologies with Different Pattern of Bottleneck
Links
Access-Core
Abilene Internet2
Multihome
23
Effect of Penalty Weight w
(U-wf)/U
Depends on of flows on each bottleneck link
User utility w Operator
penalty
Can achieve high aggregate utility for a range of
w
24
Top-down Redesign
Problem Formulation
Optimization decomposition
Distributed Solutions
Compare using simulations
TRUMP algorithm
Translate into packet version
TRUMP Protocol
25
Multipath Formulation
  • Path rate z captures source rate and routing

max. ?i Ui(?j zji) w?l f(ul) s.t. link load
cl var. path rates z
i source-destination pair, j path number
z11
z21
z31
26
Overview of Distributed Solutions
Operator Tune w, U, f Parameters tuned very
rarely
s
s
s
Routers Set up multiple paths Measure link
load Update link prices s
Edge node Update path rates z Rate limit
incoming traffic
Distributed algorithm runs on the timescale of
RTTs
27
Evaluating Four Decompositions
  • Four decompositions differ in number of tunable
    parameters
  • Theoretical results and limitations
  • All proven to converge to global optimum for
    well-chosen parameters
  • Little guidance for choosing parameters
  • Only loose bounds for rate of convergence
  • Sweep large parameter space in MATLAB
  • Compare rate of convergence
  • Compare sensitivity of tunable parameters

28
Convergence Properties
Iterations to convergence
o average value x actual values
Parameter sensitivity
Best rate
Tunable parameter
Tunable parameters impact convergence time
29
Convergence Properties (MATLAB)
  • For all algorithms
  • Parameter sensitivity correlated to rate of
    convergence
  • Trade-off between convergence and utility
  • Comparing between algorithms
  • Extra parameters do not improve convergence
  • Allowing packet loss improves convergence
  • Direct update converges faster than iterative
    update (with constant tunable parameter)

30
Top-down Redesign
Problem Formulation
Optimization decomposition
Distributed Solutions
Compare using simulations
TRUMP algorithm
Translate into packet version
TRUMP Protocol
Construct TRUMP with different parts of previous
algorithms
31
TRUMP Algorithm
Link l pl(t1) pl(t) (ßp)(cl link
load) ql(t1) wf(ul)
Price for path j ? l on path j (plql)
Source i Path rate zji(t1) max. Ui(?kzki)
(zji )(path price)
32
TRUMP Properties
  • Theorem TRUMP converges if
  • w is sufficiently large such that p0
  • nl lt af '(ul) (1/ a 1)/f ''(ul) , nl number of
    flows
  • Proof technique contraction mapping
  • TRUMP trumps previous distributed algorithms
    (MATLAB)
  • Observe convergence to optimum
  • Faster convergence
  • Converges in many scenarios if ßp 0.05/cl2

33
Top-down Redesign
Problem Formulation
Optimization decomposition
Distributed Solutions
Compare using simulations
TRUMP algorithm
Translate into packet version
TRUMP Protocol
So far, assume fluid model and constant feedback
delay
34
TRUMP Packet-Based Version
Link l link load (bytes in period T) / T
Update link prices every T
Arrival and departure of flows are implicitly
conveyed through price changes
Source i Update path rates at maxj RTTji
35
Packet-level Experiments (NS-2)
  • Set-up
  • Topologies and delays of large ISPs (Rocketfuel
    data)
  • Selected flows and paths
  • Link failures and recoveries
  • ON-OFF traffic model
  • Questions
  • Does TRUMP react quickly to dynamics?
  • How many paths does TRUMP need?

36
TRUMP Link Dynamics (NS-2)
Link failure or recovery
TRUMP reacts quickly to link dynamics Same
observation for ON-OFF flows
Throughput (Mbps)
Time (s)
37
TRUMP A Few Paths Suffice
Throughput (Mbps)
Time (s)
Sources benefit the most with a few alternative
paths
38
Summary of TRUMP Properties
39
Related Work
  • Multiple decompositions (PalomarChiang06)
  • Design traffic-management protocols
  • Congestion control (FAST TCP)
  • Dynamic traffic engineering (REPLEX, TeXCP)
  • Traffic management (KeyMassoulieTowsley07,
    LinShroff06, Shakkottai et al 06, Voice07)

40
Contributions
  • Design process
  • Formulated new objective for traffic management
  • Compared four distributed algorithms (from
    decomposition)
  • Constructed TRUMP based on insights
  • TRUMP
  • Universal parameter setting
  • Packet-level protocol and simulations

41
PART THREE
  • DaVinci Dynamically Adaptive Virtual Networks
    for a Customized Internet

Joint work with Rui Zhang-Shen, Ying Li, Mike
Lee, Martin Suchara, and Umar Javed
42
Internet Has Many Applications
  • Different application requirements
  • Throughput-sensitive file transfer, web
  • Delay-sensitive VoIP, IPTV, online gaming

43
Support Multiple Traffic Classes
  • Key research areas
  • QoS provides separate resources to support
    multiple traffic classes in parallel
  • Overlays provide customized protocols for each
    traffic class
  • Network virtualization is emerging
  • Current applications router consolidation,
    experimental test beds, VPNs
  • Router virtualization separate resources
  • Programmable routers customized protocols

44
Virtual Networks
Each virtual node/link has isolated resources
45
Motivation for Virtualization
  • Two traffic classes
  • Delay-sensitive traffic (DST) fixed demand
  • Throughput-sensitive traffic (TST) elastic

  • Single queue
  • TST can fill up both links
  • DST may not be satisfied
  • Shared routing
  • DST chooses shorter path
  • Capacity wasted

5ms, 100 Mbps
2
1
10ms, 1000 Mbps
46
Adaptive Network Virtualization
  • How to partition resources?
  • Static partitioning
  • Simple, but can be inefficient
  • One virtual network could be congested while
    another is idle
  • Dynamically allocate bandwidth shares!

47
Dynamically Adaptive Virtual Networks for a
Customized Internet
  • DaVinci is an architecture to realize adaptive
    network virtualization
  • Virtual networks indexed by (k)
  • One per traffic class
  • Run customized traffic-management protocols
  • Substrate network
  • Provides separate queues
  • Computes per link bandwidth shares
  • Enforce bandwidth shares with traffic shapers

48
DaVinci Substrate Link
s l(1)
Bandwidth shares computation
Congestion price computation
link load
yl(1)
yl(2)
yl(N)
Use optimization to determine the computations
49
ISP Maximize Aggregate Performance
weighted aggregate performance objective
max. ?k w(k)U(k)(z(k), y(k)) s.t. ?k H(k)z(k)
c var. z(k) , y(k)
bandwidth shares
path rates
? users efficiently using resources
50
Primal Decomposition
  • ISP problem decomposes into multiple subproblems
    (per traffic class)
  • Master problem update y(k) using
  • Indication of congestion s(k)
  • Indication of performance d/dy(k) U(k)(z(k), y(k))

max. U(k)(z(k), y(k)) s.t. H(k)z(k)
y(k) var. z(k)
51
Bandwidth Allocation for Link l
Adjust bandwidth in two steps
?(k)l s(k)l d/dy(k) U(k)(z(k), y(k))
v(k)l(t1) y(k)l(t) (ßy)(w(k)?(k)l)
Projection onto feasible region
v
?k y(k)l cl
52
Theorem
  • Theorem the bandwidth share computation together
    with per traffic class problem maximizes
    aggregate performance if
  • The objective function and constraints are convex
  • The stepsize ßy is diminishing
  • The bandwidth shares are updated when the
    congestion prices have converged
  • Proof technique primal decomposition

53
System Properties from Theorem
  • Resources are efficiently utilized to maximize
    aggregate performance
  • Bandwidth shares converge to a stable value and
    the computation is
  • Based only on local link information
  • Each virtual network runs its own protocols
    independently
  • Bandwidth shares updated more slowly than
    congestion prices

54
DST on High Capacity, High Delay Link
5ms, 100 Mbps DST 50Mbps
2
1
Mbps
10ms, 1000 Mbps DST 500Mbps
Number of iterations
DST does not use all the allocated bandwidth
55
Related Work and Contributions
  • Related Work
  • QoS, overlays, and network virtualization
  • Primal decomposition
  • Contributions
  • Introduced adaptive network virtualization
  • Introduced DaVinci
  • Proved stability and optimality of DaVinci

56
Conclusions
  • Traffic management today is
  • An organic evolution
  • Complex for operators
  • Redesign of traffic management to support
    multiple traffic classes
  • TRUMP design of an individual traffic class
  • DaVinci design of resource allocation between
    traffic classes

57
Future Research Directions
  • Extending DaVinci
  • Tailoring to application-specific requirements,
    e.g. R-factors for voice traffic
  • Running sub-optimal but simpler protocols
  • Interdomain traffic management requires
  • Economic incentives
  • Protection against malicious users

58
Publications Related to Thesis
  • Part One Globecom, JSAC
  • Part Two CoNext, submitted to ToN
  • Part Three under preparation
  • Related publications
  • Multipath survey IEEE Network Magazine
  • Design Optimizable Protocols CCR Editorial,
    invited book chapter

59
The End
  • Thank you!

60
Abilene Topology f e(yl/cl)
Aggregate utility gap
Gap exists
Standard deviation of capacity
61
Abilene Continued f n(yl/cl)n
Aggregate utility gap
n
Gap shrinks with larger n
62
Optimization Decomposition
  • Deriving prices and path rates
  • Prices penalties for violating a constraint
  • Path rates updates driven by penalties
  • Example TCP congestion control
  • Link prices packet loss or delay
  • Source rates AIMD based on prices
  • Our problem is more complicated
  • More complex objective, multiple paths

63
Effective Capacity (Links)
  • Rewrite capacity constraint
  • Subgradient feedback price update
  • Stepsize controls the granularity of reaction
  • Stepsize is a tunable parameter
  • Effective capacity keeps system robust

link load yl effective capacity yl cl
link load cl
sl(t1) sl(t) stepsize(yl link load)
64
Key Architectural Principles
  • Effective capacity
  • Advance warning of impending congestion
  • Simulates the link running at lower capacity and
    give feedback on that
  • Dynamically updated
  • Consistency price
  • Allowing some packet loss
  • Allowing some overshooting in exchange for faster
    convergence

65
Four Decompositions - Differences
Differ in how link source variables are updated
Iterative updates contain stepsizes They affect
the dynamics of the distributed algorithms
66
TRUMP versus File Size
TRUMPs is better for large files
Achieved aggregate rates ()
Average File Size (Mbps)
TRUMPs performance is independent of variance
67
Delay-sensitive Traffic Minimizes Delay
Links are indexed by l
Propagation delay
Cost f(ul)
min. ?l Hljizji(plf(ul)) s.t. ul ?i
Rlixi/cl ?i zji xDi var. z
ul 1
Link Utilization ul
Traffic demand
Cost function represents penalty for long queues
68
Voice Traffic R-factor
End-to-end delay
Packet loss
R Ra-a1d a2(d a3)H ß1 ß2log(1ß3f)
constants
R-factor 50-60, 60-70, 70-80, 80-90,
90-100 Voice quality poor, low, medium, high,
best
About PowerShow.com