a middleware for data-centric and dynamic distributed complex … · 2016-07-12 · 4 data-centric...

21
Laboratory for Advanced Collaboration LAC A Middleware for Data-centric and Dynamic Distributed Complex Event Processing for IoT Real-time Analytics in the Cloud Gustavo B. Baptista, Felipe Carvalho, Sergio Colcher and Markus Endler Department of Informatics – Pontifical Catholic University of Rio de Janeiro (PUC-Rio) Rio de Janeiro, Brazil. {gbaptista,fcarvalho,colcher,endler}@inf.puc-rio.br

Upload: others

Post on 17-Apr-2020

4 views

Category:

Documents


0 download

TRANSCRIPT

Laboratory for Advanced Collaboration

LAC

AMiddlewareforData-centricandDynamicDistributedComplexEventProcessingfor

IoTReal-timeAnalyticsintheCloud

Gustavo B. Baptista, Felipe Carvalho, Sergio Colcher and Markus Endler

Department of Informatics – Pontifical Catholic University of Rio de Janeiro (PUC-Rio)

Rio de Janeiro, Brazil.{gbaptista,fcarvalho,colcher,endler}@inf.puc-rio.br

2

Introduction

§ IoTbigdatareal-timeanalyticssystems• Massiveamountsofdata• Streamsproducedbydistributeddatasources

§ ReactiveParadigm• DistributedCEPSystemsareverysuitable

• Challenges indeployingandmanagingprocessinglogicatexecutiontime

• 24x7availability

3

Data-centricParadigm

§ Data-centricParadigm• Themeanofinteractionisdata.(vs.message-centric,isthemessage)

• Themiddleware§ Hasthedefinitionofstructureanddata§ Awareofcontents (i.e.instances)ofstructures§ Imposesrulesonstructures, changesandaccess§ Managesdistributedstate

• Data-centricPublish-Subscribe (DCPS)§ GlobalSharedDataSpace§ Logicaldecentralized spacemaintainedbyallpeers§ Containsthestructureandinstancesofdata§ Nodesread/writedata§ Infrastructure ensuresallparticipantsaconsistentandup-to-dateview

4

Data-CentricParadigm

§ DataDistributionServiceforReal-TimeSystems(OMG-DDS)• Fullydistributedpeer-to-peer (i.e.broker-less)• Real-timedata-centricpublish/subscribe• Highperformancecommunication,scalabilityandavailability• SpecificationofQualityofService(QoS)contracts• Mechanismsfordealingwithreal-timeaspects• PriorityandotherspecificQoSpolicies• Interoperabilityacross

§ DDSimplementations§ Programminglanguages§ Platforms

• Automaticdiscovery

5

Data-CentricParadigm

§ DataDistribution ServiceforReal-TimeSystems(OMG-DDS)

6

D3CEPMiddleware

§ WepresentamiddlewareforDistributedCEP• BenefitsofData-centricanddynamicdesignapproach

§ Dynamicdefinition anddeploymentofCEPrules§ Peer-to-peerroutingofeventsamongCEPrules§ Reducedcouplingofproducers,consumersandCEPrules§ Availabilityprovidedbypeer-to-peermodel§ Highthroughputandlowlatencyincommunication&detection§ EsperasaCEPengineateachnode

• Architectureandtestsregardingperformance andscalability

7

D3CEPMiddleware

§ GlobalSharedReactiveDataSpace• Datadisseminationandreactivebehaviormodeledtogether

§ Descriptionofeventsatprocessingandcommunicationlayers• Consumers,producersandCEPrules

§ Dynamicallydefinedanddeployed seamlessly• Additionalmechanisms:

§ GlobalCatalogofeventtypes§ GlobaldefinitionofDCEPentitiesanddeployment§ EPAs,EPNs,CEPRules§ CEPservices

8

D3CEPMiddleware

9

Architecture

10

Architecture

§ ProcessingNodeDaemon

11

Architecture

§ MetadataSharingService

12

Architecture

§ D3CEPAdministration

13

Architecture

§ DynamicDDSTopicsService

14

UseCase

§ TelemetryApplication

15

Evaluation

§ DetectionTime

16

Evaluation

§ DetectionTime

17

Evaluation

§ Throughput

18

Evaluation

§ Throughput

19

RelatedWork

20

ConclusionsandFutureWork

§ MainContributions• Data-centricdesignapproachtoDCEP

§ UseofDDSforpeer-to-peer routingofevents§ DynamicDeploymentandAutomaticDiscovery

§ FacilitatesdeploymentofCEPrules§ DistributedStateManagement

§ CEPrulesseamlesslyread/writeGlobalSharedDataSpace

§ Futurework• QoScontractsatthedetection level[Appeletal.2010]§ IoTandmissioncriticalapplications

21

Questions

Thank you