vitus masters defense

24
PlanetenWachHundNetz Instrumentation Infrastructure for PlanetLab Vitus Lorenz-Meyer

Post on 13-Sep-2014

1.038 views

Category:

Technology


0 download

DESCRIPTION

 Final defense of my masters thesis at UTEP, 2006

TRANSCRIPT

Page 1: Vitus Masters Defense

PlanetenWachHundNetz

Instrumentation Infrastructure for PlanetLab

Vitus Lorenz-Meyer

Page 2: Vitus Masters Defense

Peer-to-Peer

• Distributed on open internet• All participants both receive & provide

services to/from others• Not centrally administered• Membership changes over time (churn)• Example: file sharing (napster, gnutella,…)

Any node can publish a named file Any node can obtain file from another node who

has it. Range of strategies to find nodes containing

desired contentVitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso2

Problem

Page 3: Vitus Masters Defense

Rel. Work

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso3

The Problem

• P2P systems hard to tune, requires understanding of complex behavior Requires instrumentation & analysis

• Many P2P systems constructed without scalable instrumentation infrastructure Frequently done in ad-hoc manner

Data transmitted to single collection & analysis node Inadequate for understanding behavior of large systems

of many (hundreds to MILLIONS) of nodes• My work: development of a flexible tool to

enable scalable instrumentation algorithms, data structures

P2P

Page 4: Vitus Masters Defense

Related work (1 of 2: cousins)

• Distributed Database Mngmt. Systems Select data at sources Optimize joins (run near to sources…) Commercially used in non-p2p configurations P2P (research): PIER, Sophia

• Sensor Networks Unmanaged radio-connected nodes

provide “network” of surveilance SQL; Compiled into a 3-step process Software communicates through same mechanism IrisNet, TAG

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso4

Page 5: Vitus Masters Defense

Related work (2of2: Siblings)

• Aggregation Overlays Information collection subsystem Nodes provide information tuples

Internal aggregation language Computed using parallel prefix of pre-defined

assoc/comm ops Astrolabe, SDIMS, SOMO

• Google’s MapReduce Data selection & aggregation in distributed

system User provides “map” and “reduce” program

Not fully p2p (resource mgmt. overlay)

Page 6: Vitus Masters Defense

High-level Approach

• User specifiable programs like MapReduce • Split data collection into 3 ‘phases’

1. Generate values on all nodes2. Pairwise aggregation throughout system3. Evaluate results

• emit measured vals (val,num=1)• Aggregate: (val1+val2,num1+num2)• Evaluate (avg) (val / num)

Easy to use: user provides 3 programs (scripts)

Impl.Rel. Work

Page 7: Vitus Masters Defense

Illustration of Binary Aggregation Impl.Rel. Work

Page 8: Vitus Masters Defense

Why is this hard in P2P?

• Problem: membership churn• Nodes continuously enter & leave

system• Nobody in charge (p2p)

• Nobody knows membership list!

Exposes following challenges1. Finding all participating nodes2. Constructing an (appx) balanced tree

Impl.Rel. Work

Page 9: Vitus Masters Defense

Building Structure Upon Anarchy: Key Based

Routing

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso9

GoalAppr.

02160

2158

2159

2159 + 2158

Page 10: Vitus Masters Defense

“Chord” Routing

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso10

GoalAppr.

Page 11: Vitus Masters Defense

Chord lookup

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso11

GoalAppr.

Page 12: Vitus Masters Defense

Building a tree upon KBR

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso12

GoalAppr.

a aa b

b

d ea b d

e

g if hd ea b

f

g

h

i

Page 13: Vitus Masters Defense

Building a tree: FTT & KBT

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso13

GoalAppr.

FTT: finger-based tree• Operation associated with a “target” node• Systems send data to finger closest to target• Ambiguous

Depends on all nodes’ fingertables• Tree useful only for aggregation

• KBT: Maps tree on key-space Operation associated w/ target node System/tree-node mapping:

Node assigned to node w/ nearest key

Non-ambiguous Tree useful for both dissemination &

aggregation Single, global tree

Page 14: Vitus Masters Defense

Our Structure

101…

111…101…

111…101… 110…100…

001…

011…001…

011…001… 010…000…

101…

011…

001…

110…

As 001..!

GoalAppr.

• KMR: Subset of KBT, rooted at specific node• One tree / root

Better load-balancing• Tree fully determined by set of active nodes and root

Page 15: Vitus Masters Defense

Implementation details

• PWHN-Server layered on FreePastry• PWHN-Client connects to PWHN-Server and

makes query• Callee builds tree making itself root

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso15

Goal

SS S

SS

S

S

S

C

Appr.

Page 16: Vitus Masters Defense

Our Goal

• Develop toolkit for data collection/aggregation in P2P networks Useful for PlanetLab-community

• Extend MR’s model to P2P K.I.S.S.

Users provide programs for gen/agg/eval• Use techniques from P2P

Construct aggregation tree upon key-based-routing

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso16

ExampleImpl. details

Page 17: Vitus Masters Defense

Example (1)

• First implementation: Script version, flat, to test approach

• Example 1: Overall average system load Gen emits (1,<1load>,<5load>,<15load>) for each server Agg adds all numbers Eval divides last 3 numbers by first to get average

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso17

EvaluationGoal

Page 18: Vitus Masters Defense

Example (2)

• PWHN client (Java) Can start and stop

server Used for specifying

all programs and parameters (Servers, username for flat, method)

Front-end for connecting to servers and making query

Allows saving and graphically representing result

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso18

EvaluationGoal

Page 19: Vitus Masters Defense

Example (3)

• Graphing of queried results

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso19

EvaluationGoal

Bar Chart Color bubbles on world map

Page 20: Vitus Masters Defense

Example (4)

Graphing of tree

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso20

EvaluationGoal

Graphing of paths of the query

Page 21: Vitus Masters Defense

Evaluation

• Minimize disruption Minimize incoming

bytes to client

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso21

SynopsisExamples

•More efficientLower average fan-in of aggregation tree

Page 22: Vitus Masters Defense

Evaluation: Fern

• Global Update latency histogram

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso22

SynopsisExamples

10 clients 701 clients

Page 23: Vitus Masters Defense

Summary

• PWHN - Instrumentation toolkit Extends MR’s model to P2P Uses P2P techniques (DHTs) Combines FTT and KBT to be more efficient Conclusion: Useful tool that is more efficient

than to build infrastructure into software What did I do?

Survey of systems that provide aggregation in dynamic networks

Classification and naming of aggregation trees upon DHTs

Design and implementation of my own tool (KMR/PWHN)

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso23

Examples

Page 24: Vitus Masters Defense

Questions

Vitus Lorenz-Meyer: Thesis defense University of Texas @ El Paso24

Synopsis