Multipath Protocol for Delay-Sensitive Traffic

1
Multipath Protocol for Delay-Sensitive Traffic

• Jennifer Rexford
• Princeton University

Joint work with Umar Javed, Martin Suchara, and
Jiayue He
http//www.cs.princeton.edu/jrex/papers/comsnets0
9.pdf
2
Clean-Slate Network Architecture

• Network architecture
• More than designing a single protocol
• Definition and placement of function
• Clean-slate design
• Without the constraints of todays artifacts
• To have a stronger intellectual foundation
• And move beyond the incremental fixes
• But, how do we do clean-slate design?

3
Protocols as Distributed Optimizers

• Example TCP congestion control
• Additive increase, multiplicative decrease
• Implicitly maximizes aggregate utility
• TCP variants have different utility functions
• Optimization for forward engineering
• Start with a central optimization problem
• Decompose to divide the computation
• among the sources and the links

Research by Frank Kelly, Steven Low, Mung Chiang,
and others
4
Our Focus Delay-Sensitive Traffic

• Interactive applications
• Voice over IP (VoIP)
• Online gaming
• IP television
• Path-selection goals
• Paths with low propagation delay
• as long as paths are not overloaded

For now, assume the network carries only
delay-sensitive traffic
5
Strawman Min Propagation Delay
Operator Sets weights to propagation delay
3
1
2
4
2
3
Routers Link-state routing
2
2
But links may become congested, causing packet
loss and delay
6
Our Goal Adaptive Load Balancing

• Division of functionality
• Links feedback on network conditions
• Edge routers balance load over paths

Distributed protocol that automatically minimizes
delay
7
Multiple Paths With Flexible Splitting

• Multiple paths between edge nodes
• Paths with low propagation delay
• Flexible traffic-splitting ratio
• Traffic rate xi for src-dest pair i
• Traffic rate zij over path j

x1 z11 z12 z13
z11
z21
z31
8
Objective Minimize Average Delay

• Minimize average delay
• End-to-end delay on each path
• Weighted by the traffic on the path
• Delay for link l
• Propagation delay pl
• Congestion penalty f(load on link l)

Delay for link l pl f()
Summed ?i ?j ?l zij Rilj (pl f())
Weighted zij Rilj (pl f())
9
Constraints

• Carry the offered load for each source
• ?j zij xi
• Avoid overloading each link
• ?i ?j zij Rilj cl
• Carry non-negative traffic on each path
• 0 zij

10
Optimization Decomposition

• Deriving source and link algorithms
• Prices penalties for violating a constraint
• Path rates updates driven by prices
• 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

11
Example Decomposition Link Capacity

• Capacity constraint
• Subgradient feedback price update
• Stepsize controls the granularity of reaction
• Link computes price l as feedback to sources

link load cl
l l(t1) ll(t) stepsize(link load cl )
Source does similar update for carry all offered
load constraint.
12
Path Rate Updates

• Each source i does a local optimization
• To update the path rates zij
• Based on
• The prices of violating constraints
• and the objective function
• Closed-form expression
• With piecewise-linear queuing function f()
• See the paper for the exact equation

Derived by taking the Lagrangian and applying KKT
conditions.
13
Distributed Multipath Protocol
Operator Select function f
Tune step sizes
Routers Set up multiple paths Measure link
load Update link prices
Edge node Update path rates z Split traffic
over paths
14
Theoretical Results

• Optimality and stability
• Provably optimal
• Provably converges for diminishing step sizes
• Practical limitations
• Must have well-chosen step sizes
• to achieve fast convergence
• Matlab experiments to sweep parameters
• Good heuristics for setting (constant) step sizes

15
Converting to Packet-Level Protocol

• Packets rather than fluid
• Links compute load over a time interval
• Counting the sizes of the packets
• Feedback delay of round-trip time
• Multiple paths have different RTTs
• Path rate updates once per max of RTTs
• Implemented in ns-2 simulator
• For more realistic evaluation

16
Comparison With Shortest-Path Routing

• Shortest-path routing
• Link weights equal propagation delay
• Under low load
• The two protocols behave the same way
• Under higher load
• Our protocol gradually shifts traffic
• to longer paths to avoid overload
• while keeping end-to-end delay small

17
Convergence Under Dynamic Traffic
18
Multiple Classes of Traffic

• Satisfying multiple traffic classes
• Delay-sensitive VoIP and gaming
• Throughput-sensitive file transfers
• Running separate virtual networks
• Customized protocol for each traffic class
• Dynamic update to bandwidth shares
• Provably maximizes aggregate performance
• Derived using optimization theory

http//www.cs.princeton.edu/jrex/papers/davinci.p
df
19
Conclusions

• Delay-sensitive applications
• VoIP, online gaming, IPTV
• Customized routing protocol
• Load balancing over multiple paths
• Minimizing end-to-end delay
• Optimization decomposition
• Rigorous way to design new protocols
• With provable optimality and stability
• Ongoing work network virtualization

20
Backup Slides
21
Protocol Dynamics

• Good heuristics for setting step size
• Converges quickly under range of settings
• Relatively fast convergence
• Small tens of seconds in worst case
• Better under more realistic settings
• Quick response to changes in load
• Fast adaptation to new traffic demands