Benefits of coordination in multipath flow control Laurent Massoulié & Peter Key Microsoft Research Cambridge
Multipath data transfers Already a large fraction of current Internet traffic (P2P file sharing); A necessary feature for efficient mesh and ad hoc networking.
Question: What coordination between flow control on component paths needed? Extreme scenario: no coordination; e.g., individual TCP connections on each path, transferring disjoint data items. Is it good enough?
Methodology: Focus on flow-level system models; Assess performance from: Schedulable region; Equilibrium costs.
Network model: Flows of types s 2 S; Each type s has an associated set of routes, r 2 R(s) Total rate sent along each route r: r Network cost is where : convex, increasing cost function Ex:
Coordinated (fair) flow control: N s : number of type s-flows; each sends at rate r /N s over route r, where r solves: maximise where: (alpha-fairness: [Mo-Walrand]; multipath version: [Kelly-Maulloo-Tan]; [Mo-Walrand]; [HSHST])
Uncoordinated (fair) flow control: N s : number of type s-flows; each sends at rate r /N s over route r, where r solves: maximise where: Suitable for modelling uncoordinated TCP flows on each path
“Fluid” dynamics: Arrival rate of type s transfers: s ; Mean volume of type s transfers: s. Consider dynamics: “drift” of stochastic process where flow arrivals at instants of Poisson process (intensity s ) and volumes exponentially distributed (parameter s ) Interpretation: describes behaviour of stochastic system after joint rescaling of arrival rates and service capacities
Performance metrics: Schedulable region: Set of demand vectors ( s = s / s ) s S for which fluid dynamics asymptotically stable. Equilibrium cost: For demand vector ( s ) s S in schedulable region, network cost ({ r (N * )} r R ) at equilibrium point N *.
Performance under coordination: 1)Schedulable region contains any vector ( s ) s S such that: there exists a vector of route loads ( r ) r R int(dom( )) verifying (eg, for sharp capacity constraints: ) 2) Given ( s ) s S, equilibrium cost achieves minimum of (( r ) r R ) over all such ( r ) r R irrespective of alpha-fairness criterion used.
Bad performance without coordination: Example network: sharp link capacity constraints Schedulable region with coordination: C a bc 2C a b c b + c < 2C, a + c < 2C, b + a < 2C.
Bad performance (ctd) Schedulable region without coordination: Assume alpha-fair sharing with identical weights w. Symmetric load vector ( a = b = c = schedulable iff: < C[1+2 -1/ ]/[ / ] With coordination: iff < C. e.g. for =2, a loss of 29% efficiency. C a bc
Beyond the triangle network grids cliques a d c b
Network: links l, capacity C l. Routes: single link. Then schedulable region (with or without coordination): However uncoordinated multipath produces higher equilibrium cost, sensitive to fairness criterion. e.g., for network: At equilibrium load split into The case of 1-hop routes abc 1324 a 2 1
Flow-level models can help select fairness objective of congestion control. Previously proposed coordination optimal in terms of both schedulable region and equilibrium cost. Open problems: route selection? Concluding remarks