21
A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University www.cs.bu.edu/~byers Joint work with Jeffrey Considine (BU) and Ketan Mayer-Patel (UNC)

A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

  • View
    214

  • Download
    1

Embed Size (px)

Citation preview

Page 1: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

A Constraint Satisfaction Approach to Testbed Embedding

Services

John ByersDept. of Computer Science, Boston

Universitywww.cs.bu.edu/~byers

Joint work with Jeffrey Considine (BU)and Ketan Mayer-Patel (UNC)

Page 2: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• Simulation “Blank slate” for crafting experiments Fine-grained control, specifying all details No external surprises, not especially realistic

• Emulation All the benefits of simulation, plus: running real protocols on real systems

• Internet experimentation None of the benefits of simulation, minus: unpredictability, unrepeatability, etc. But realistic!

Experimental Methodologies

Page 3: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• All three approaches have their place.• Improving aspects of all three is essential.

Focus of recent workshops like MOME Tools Internet experimentation is the most primitive by

far.

• Our question: Can we bridge over some of the attractive features

of simulation and emulation into wide-area testbed experimentation?

• Towards an answer: Which services would be useful? Outline design of a set of interesting services.

Our Position

Page 4: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• Canonical target testbed: PlanetLab• What services would we like to bridge

over? Abstract: repeatability, representativeness Concrete:

• specify parameters of an experiment just like in ns • locate one or more sub-topologies matching

specification• run experiment• monitor it while running (“measurement

blackboard”)• put it all in a cron job

Useful Services

Page 5: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

1. Topology specification2. Testbed characterization

• Relevant parameters unknown, but measurable

3. Embedding discovery• Automatically find one or more embeddings of

specified topology

Synergistic relationships between above services.

Existing measurements guide discovery. Discovery feeds back into measurement

process.

Embedding Services

Page 6: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• In the emulation world, Emulab and Netbed researchers have worked extensively on related problems [OSDI ’02, HotNets-I, CCR ’03]

• Rich experimental specification language.• Optimization-based solver to map desired

topology onto Netbed to: balance load across Netbed processors minimize inter-switch bandwidth minimize interference between experiments incorporate wide-area constraints

Emulab/Netbed

Page 7: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• Conditions change continuously on wide-area testbeds - “Measure twice, embed once”.

• The space of possible embeddings is very large; finding feasible ones is the challenge.

• We argue for a constraint satisfaction approach rather than optimization-based. Pros and cons upcoming.

Wide-area challenges

Page 8: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Specifying Topologies

• N nodes in testbed, k nodes in specification

• k x k constraint matrix C = {ci,j}

• Entry ci,j constrains the end-to-end path between embedding of virtual nodes i and j.

• For example, place bounds on RTTs:ci,j = [li,j, hi,j] represents lower and upper bounds on target RTT.

• Constraints can be multi-dimensional.• Constraints can also be placed on nodes.• More complex specifications possible...

Page 9: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Feasible Embeddings

• Def’n: A feasible embedding is a mapping f such that for all i, j where f(i) = x and f(j) = y:

li,j ≤ d (x, y) ≤ hi,j

• Do not need to know d (x, y) exactly, only thatli,j ≤ l’(x, y) ≤ d (x, y) ≤ h’ (x, y) ≤ hi,j

• Key point: Testbed need not be exhaustively characterized, only sufficiently well to embed.

Page 10: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Why Constraint-Based?

• Simplicity: binary yes-no answer.• Allows sampling from feasible embeddings.• Admits a parsimonious set of

measurements to locate a feasible embedding.

• For infeasible set of constraints, hints for relaxing constraints can be provided.

• Optimization approaches depend crucially on user’s setting of weights.

Page 11: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Hardness

• Finding an embedding is as hard as subgraph isomorphism (NP-complete)

• Counting or sampling from set of feasible embeddings is #P-Complete.

• Approximation algorithms are not much better.

• Uh-oh...

Page 12: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Our Approach

• Brute force search.• We’re not kidding. • Situation is not as dire as it sounds.

Several methods for pruning the search tree.

Adaptive measurements. Many problem instances not near boundary of

solubility and insolubility.

• Off-line searches up to thousands of nodes.• On-line searches up to hundreds of nodes.

Page 13: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Adaptive Measurements

• Must we make O(N2) measurements? No. Delay: coordinate-based [Cox et al (today)] Loss: tomography-based [Chen et al ’03] Underlays may make them for you [Nakao et al

’03]

• In our setting: We don’t always need exact values. Pair-wise measurements are expensive.

• How do we avoid measurements? Interactions with search. Inferences of unmeasured paths.

Page 14: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Suppose constraints are on delays, and the triangle inequality holds.

Triangle Inequality Inferences

i

j

k

[ 90, 100 ] [10, 15 ]

[ , ]115i

j

k

[ 90, 100 ] [10, 15 ]

[ , ]11575

Using APSP algorithms, can compute all upper & lower bounds.

Page 15: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Experimental Setup

• Starting from PlanetLab production node list, we removed any hosts… not responding to pings with full file systems with CPU load over 2.0 (measured with uptime)

• 118 hosts remaining• Used snapshot of pings between them

Page 16: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Finding Cliques

• Biggest clique of nodes within 10 ms Unique 11 node clique covering 6

institutions• If 1ms lower bound added,

Twenty 6 node cliques 5 institutions always present, only 2 others

111098765432Size

000002014260

38784

50161

32535

1-10ms Cliques

1143

863

3158

7718

13278

164517

14759

936 5

4038

0-10ms Cliques

Page 17: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Finding Groups of Cliques

0000004

00000129503

0001004623868982

0201423875013251

765432

Clique Size# of Cliques

1-10ms within same clique, 20-50ms otherwise

Page 18: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Triangle Inequality in PlanetLab

• In our PlanetLab snapshot, 4.4% of all triples i,j,k violate the triangle inequality

• Consider a looser version of TI, e.g.di,j ≤ α ( di,k + dk,j ) + β

• There are fewer than 1% violations if α = 1.15, β = 1 ms α = 1.09, β = 5 ms

Page 19: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

• Metric: mean range of inference matrix

• Compare measurement orderings Random Smallest Lower Bound First (greedy) Largest Range First (greedy)

Inference Experiments

i ij

jiji

n

lhM

2

,,

Page 20: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Inference Results

Random order performs poorly

Largest range first performs best

Page 21: A Constraint Satisfaction Approach to Testbed Embedding Services John Byers Dept. of Computer Science, Boston University byers Joint work

Future Work

• Build a full system More interesting constraints Better search and pruning Synergistic search and measurement Integration with simulation/emulation

tools• Other questions

What do search findings say about PlanetLab?

Can we plan deployment?