: improving internet telephony quality using predictive...

Post on 21-Jun-2020

10 Views

Category:

Documents

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

VIA:ImprovingInternetTelephonyQualityUsingPredictiveRelaySelection

Junchen Jiang,Rajdeep Das,GaneshAnanthanarayanan,PhilipA.Chou,Venkata N.Padmanabhan,VyasSekar,Esbjorn Dominique,Marcin

Goliszewski,Dalibor Kukoleca,Renat Vafin,HuiZhang

1

Keytakeawaysinoneminute

Onefifth ofcallsusepathswithpoornetworkperformance

“ManagedOverlay”couldalleviateoverhalfofcallsonthesepaths

VIA:Data-drivenrelayingcanrealizemostManagedOverlay’spotential

Westudied430millionSkypecallsandfound…

2

0

20

40

60

80

2004200520062007200820092010201120122013

Million

Peak#ofusersconcurrentonlineonSkype

Internettelephonyiseverywhere!

MoreappsfocusingonInternettelephony

https://blogs.skype.com/2013/04/03/thanks-for-making-skype-a-part-of-your-daily-lives-2-billion-minutes-a-day/

Rapidgrowthoverthelastdecade

https://www.statista.com/chart/1417/skype-usage/3

Callqualityissensitivetobadnetworkperformance

4

0

0.2

0.4

0.6

0.8

1

0 200 400 600Normalize

dPo

orCallRate

RTT(ms)

0

0.2

0.4

0.6

0.8

1

0 1 2 3 4Lossrate(%)

0

0.2

0.4

0.6

0.8

1

0 5 10 15 20 25Jitter(ms)

Thresholdsofpoornetworkperformance

Datasets:Avg RTT,lossrate,jitterforeachcallof430millionSkypecallsSmallfractionhasuser-providedqualityscores

320ms 1.2% 12ms

Manycallshavepoornetworkperformance

5

17%17%12%Onefifth ofcallshavepoornetworkperformance

Ourgoal:AlleviatepoornetworkperformanceforSkype

Outline

• Problem: NetworkperformanceofSkypeisbad

è Opportunity: ManagedOverlayhashugepotential

• Solution: VIAforoptimalrelayselection

• Evaluation: VIAisclose-to-optimal

6

RevisitingOverlayNetworksbyManagedOverlay

ManagedOverlayhasnewbenefits• World-wide distributedDCs asrelays• Wellconnected• Deployedbymanyproviders• Single administrativeentity

HowmuchcanInternettelephonybenefitfromit?

7

Selectingthebestrelayoption

Keyistoselectthebestrelayoption(direct,one-relay,ormulti-relay)

Q1:Doespickingbestrelayoptionhavesignificantimpact?Q2:Ifso,howtopickthebestrelayoption?

8

Directpath

One-relayhop

Multi-relayhops

ManagedOverlayhashugepotentialbenefit

Substantialfractionofbad-performancecallscould bealleviated

Consideranoraclethatpicksthebestrelayoptionforeachsrc-dst ASpairin24hours

9

0102030405060

RTT Loss Jitter Atleastonebad

Poorperformance:RTT>320ms;Lossrate>1.2%;Jitter>12ms

%calls“Bad”è “Good”

Outline

• Problem: NetworkperformanceofSkypeisbad

• Opportunity: ManagedOverlayhashugepotential

è Solution: VIAforoptimalrelayselection

• Evaluation: VIAisclose-to-optimal

10

VIA:RealizingthebenefitofManagedOverlayusingCentralizedPredictiveControl

Qualityofexistingcalls

Relayselection

VIAControlAlgorithmPredictthebestrelayoptionbased

onothercalls’performance

11

Strawman 1:Pureprediction-based

12

Uselong-termhistorytopredictperformance

Relayoptions

Anewcall

Problem:CallperformancehasgreatinherentvarianceE.g.,predicting“nextday”using“lastweek”leadstoover30%erroronlatency.

CallhistoryQualityprediction

Strawman 2:Pureexploration-based

13

Inashorttimewindow,explorerelayoptions,thenexploitthebestone

Relayoptions

CallsofoneASpairperday

Problem:Calldistributionishighlyskewed“#ofactualcalls>>#ofrelayoptions”NOTtrueformostASpairs

Keyidea:Guidedexploration

Prediction-based Exploration-basedXORStrawmenVIA

14

00.20.40.60.81

0 1 2 3 4 5 6 7 8 9Prob

abilityofthe

best

relaybe

inginto

pk

k

Topk canbemoreeasilypredictedandcanbemore efficientlyexplored!

VIA’sidea:GuidedExplorationRoughpredictioncanstillidentifytopk candidates,whichcanbeexploredefficiently.

Predictingtop1istoohard

Step1:Prediction-basedpruning

Latency

Confidenceintervalofqualityprediction

Top-kcandidates

Focusonrelayoptionswhoseconfidenceintervalsarebetterthanthoseofothers.

15

Callhistory

Upperboundof“blue”and“yellow”isbetterthanthelowerboundof“green”and“red”

Step2:Exploringtop-kcandidates

16

Multi-armedbanditprocessUpperConfidenceBounds(UCB1)

Reward

OurproblemlookslikeMABUCB1withdomain-specifictwists

Reward

Reward

Howtomaximizerewards?

Howtominimizelatency?

UCB1:AlwayspicktheonewiththehighestUCB

Puttingthemtogether:GuidedExplorationinaction

Callhistory

Performancemeasurements

Tomography-basedcoverageexpansion

Predictiveselectionoftopk candidates

ModifiedUCB1onthetopk candidates

Toexploremorerelays Real-timerelayselection

Runspercall

Prediction-basedPruningUpdateseveryT hours

Top-k candidatesperASpair

17

Allrelayoptions

Moreinourpaper

• Budgetedrelaying

• Networktomography

• Granularityofprediction

• Internationalvs.domesticcalls

18

Outline

• Problem: NetworkperformanceofSkypeisbad

• Opportunity: ManagedOverlayhashugepotential

• Solution: VIAforoptimalrelayselection

è Evaluation: VIAisclose-to-optimal

19

VIAachievesclose-to-optimalperformance

20

Poorperformance:RTT>320ms;Lossrate>1.2%;Jitter>12ms

%calls“Bad”è “Good”

0

10

20

30

40

50

60

RTT Loss Jitter Atleastonebad

Optimal VIAPrediction-based Exploration-based

BenefitvariesacrossASes

0

20

40

60

80

100

SourceAS1 SourceAS2 SourceAS3

Oracle VIA

Substantialimprovement

Limitedroomforimprovement

Roomforfurtherimprovement

21

%calls“Bad”è “Good”

Conclusion

• Internettelephony isheretostay• Problem:Onefifthofcallshavebadnetworkperformance.

• Opportunity:ManagedOverlaycould significantlyreducebad-performancecalls.

• Challenges:OptimalrelayselectionPurepredictionandpureexplorationwon’twork!

• Solution:VIAcanrealizemostofthebenefitofManagedOverlay.Keyidea:Guidedexploration:

Predictivepruningofrelayoptions+Explorationovertopkcandidates22

top related