智能与预测性故障转移迈向弹性的错误对抗之路 conv.docxVIP

IntelligentandPredictiveFailover

-TheraceagainsterrorsbymarchingtowardsResiliency..!!

TEJASWAROOPMYLAVARAPU(LeadSoftwareEngineer-CapitalOne)

AGENDA

1.TerminologyDefinitions2.WhatisResiliency?

3.RTO,RPOandMTD

4.InfrastructurePremisewithAlarmsandnotifications5.FailoverMechanism-Active-ActiveFailover

6.DistributedSystemtightlyintertwinedwithDownstreamSystems7.FailoverMechanism-ReactiveFailover

8.ConceptofPredictiveFailover

9.PredictiveFailover-IntelligentandDeepHealthChecks10.IntelligentMonitoringwithML

11.Synthetictrafficanddisastersimulation12.ChaosEngineering

AWSInfrastructure-ClustervsEC2vsDockerContainers

Definitions

Container

EC2Instance

Database

AStandalone,executablepackagethatincludesalltheconfigsneededtorunanapplicationinanyenvironment/OS

AvirtualserverinAmazonWebServices(AWS).SimilartoVirtualMachinesinAzure

Alogicalcollectionofstructuredinformationactingasastoragemechanism.

AWSInfrastructure-Lambdas,LoadBalancerandRoute53

Definitions

Lambda

LoadBalancer

Route53

AWSLambdaisanevent-driven,serverlesscomputingplatformmanagedbyAWSwhichcanbeusedforevent-drivenorwebbasedapplications

DistributesincomingapplicationtrafficacrossmultipletargetgroupssuchasEC2InstancesorLambdas

ScalableDomainNameSystem(DNS)webservicewhichconnectsuserrequeststoAWSComponents

WhatisResiliency?

TextbookDefinition:TheabilityofanapplicationtoresistorrecoverfromanydisastersrelatedtoInfrastructure,downstreamsystems,trafficspikes,networkissues.

Itisnotonlytheabilitytorecoverfromdowntimes,butalsotheabilitytoavoidthedowntimesandstillbeconsistentwithsecurity,consistencyandperformance.

Resilience=BrandValue

OveralllowCustomerImpact,lowfailuresandlowerwaittime=Bettertrust+increaseinCustomerconfidenceingeneral.

Companiesprioritizetheirinvestmentso