Analysis Of Load BalancingAlgorithms with QOS-AwareWorkLoads in MicroservicesBasedContainerized Applications

Abstract

ABSTRACT QualityofService(QoS)representsasystem’sabilitytoconsistentlydeliverreliableper- formance undervaryingworkloadconditions,ensuringthatcriticalservicesmeetspecific metrics suchaslatency,throughput,availability,andfaulttolerance.Intoday’sdynamic digital landscape,cloudcomputing,containerization,andmicroservicesserveastheback- boneofmodernapplicationdevelopmentanddeployment.Thisresearch,titled“Analysis Of LoadBalancingAlgorithmswithQoS-AwareWorkLoadsinMicroservicesBasedCon- tainerized Applications,”conductsacomparativeanalysisofloadbalancingalgorithmsand QoS mechanismswithincontainerizedmicroservicesarchitectures,withaparticularfocuson cloud servicesprovidedbyAmazonWebServices(AWS).Thestudyevaluatesperformance, load distribution,andothercriticalaspectstoofferinsightsintooptimizingdistributedap- plications forenhancedefficiencyandreliability.Thisinterestingstudyinvestigatesthe impact ofloadbalancingalgorithmsandQualityofService(QoS)featuresinAWSElastic KubernetesService(EKS)onperformanceoptimizationandresourceutilization.Withcloud service customersincreasinglyfocusedonhighperformanceandefficientresourceallocation, particularly forcriticaltraffic,theresearchexamineshowQoScanmeetthesedemands. The studyexplorestheeffectivenessofQoSinprioritizingcriticalloadsandoptimizingper- formance, usingrealloadconditionswithAWSEKS-deployedSpringBootmicroservices. Unlikesimulation-basedresearch,thisstudyprovidesacomparativeanalysisbasedonactual data. ProcessingtimewithQoSawarehasbeenimproved2timescomparedtowithoutQoS awareness.QosawareMemoryusageis66%reduced,andCPUusageis5.5%reducedwith QoS awarenesswithveryhighloadsmonitoringinalongerperiodoftime.Thefindingsshow that applyingQoSprioritizationtopodclassesresultsinconsiderableimprovedperformance, alongside someenhancedresourceutilizationefficiencyinhighloadconditions.