Skip to main content

Advances, Systems and Applications

Table 2 Reactive approaches of cloud load balancing in existing literature

From: Load balancing in cloud computing – A hierarchical taxonomical classification

ReferenceAlgorithm UsedState of AlgorithmTrait UsedType of Load BalancingTechnique
Algorithm ComplexityAdvantagesDisadvantages
[48]Conventional non classical AlgorithmDynamicTask SchedulingTask LBNon Classical, DeterministicNot SpecifiedBetter makespanTask deadline not considered
Better resource utilizationSLV not considered
Less waiting timeLess fault tolerant
Less execution timeLess energy efficient
[39]Classical and Linear ProgrammingDynamicTask SchedulingTask LBOptimization (Linear programming Based)Not SpecifiedBetter makespanReduced quality of service
Better resource utilization
[49]GA and Min-MinHybridTask SchedulingTask LBHeuristic (Evolutionary)O(m) and O (mn)Better scalabilityLess resource utilization
Less response timeHigh SLV
Small execution costLess degree of balance
[50]BFO+ Lamarack EvolutionaryHybridResource SchedulingCPU LBOptimizationNot SpecifiedLow VM downtime, execution timeLow scalability and throughput
Less transfer timeLow resource utilization
[51]PSODynamicTask SchedulingTask LBOptimizationNot SpecifiedLow energy consumptionLow scalability
High resource utilizationLow fault tolerance
Small degree of balance
Less makespan
High SLV
[52]GADynamicVM SchedulingVM LBMetaheuristicNot SpecifiedLess response timeLow throughput
Less makespanLow scalability
Less task rejection ratioSmall degree of balance
Small resource utilization
[53]GADynamicTask SchedulingTask/VM LBOptimizationG = O {n1 + (c × k) + (n2 + 1) (m + m + m)}High degree of balanceLow scalability
Less makespanlow energy efficiency
Less execution timelow fault tolerance
Less task rejection ratio
[54]ACO and PSODynamicVM SchedulingVM LBMetaheuristicO(n2MAI)low response timelow throughput
low execution timelow degree of balance
high SLV
low resource utilization
[55]GA and GELHybridTask SchedulingVM LBOptimizationNot Specifiedhigh scalabilitylow degree of balance
high fault tolerancehigh SLV
low overheadlow resource utilization
low migration time and power consumptionhigh TRR
[56]Honey Bee AlgorithmDynamicTask SchedulingTask LBOptimizationNot Specifiedlow response timelow throughput and scalability
low execution timelow degree of balance
low execution costlow resource utilization
[57]Non ClassicalDynamicResource SchedulingResource LBHeuristicNot SpecifiedHigh throughputLow SLV
High scalabilityLow resource utilization
Low response timeHigh task rejection ratio
Low execution timeLow degree of balance
High migration time
[58]Non ClassicalDynamicVM SchedulingVM LBOptimizationNot SpecifiedLow migration timeLow throughput
High degree of balanceLow makespan
Low response timeHigh SLV
Low resource utilization
Low scalability
[59]BAT AlgorithmDynamicResource/Task SchedulingResource/ Task LBOptimizationNot SpecifiedLess execution timeHigh makespan
Low execution costLow throughput
Energy inefficient
Low resource utilization
[60]Non ClassicalDynamicVM SchedulingVM LBOptimizationNot SpecifiedLess response timeLow scalability
Low execution costHigh SLV
Low degree of balance
[61]Simulated AnnealingDynamicTask SchedulingTask LBOptimizationNot SpecifiedHigh throughputLow fault tolerance
High scalabilityEnergy inefficient
Low overheadHigh SLV
Less makespan
High resource utilization
[62]Round RobinStaticVM SchedulingVM LBHeuristicNot SpecifiedHigh Fault toleranceLess scalability
Small overheadHigh SLV
Less migration timeLow task rejection ratio.
Good resource utilization
[63]Round RobinDynamicResource SchedulingResource LBHeuristicNot SpecifiedHigh Fault toleranceLess scalability
Less migration timeHigh SLV
Good resource utilizationLow task rejection ratio
[64]Non ClassicalStaticResource SchedulingResource LBOptimizationNot SpecifiedLess Response timeLow throughput and scalability
Low execution costLow resource utilization
Low degree of balance
[65]Active MonitoringDynamicVM SchedulingVM LBHeuristicNot SpecifiedLess response timeLow throughput
Less execution timeLow scalability
Less execution costLow degree of balance
Low resource utilization
[66]Active MonitoringDynamicVM/Task SchedulingVM/Task LBHeuristicNot SpecifiedHigh scalabilityLow throughput
Less response timeLow fault tolerance
High resource utilizationHigh makespan
High SLV
[67]Active MonitoringDynamicResource SchedulingResource LBHeuristicNot SpecifiedLow overheadLow throughput
Less makespanPower inefficient
High resource utilizationHigh SLV
[68]Joint use of min-min and max-minStaticTask SchedulingTask LBOptimizationNot SpecifiedHigh degree of balanceLow scalability
Low makespanLow fault tolerance
Low execution timeHigh SLV
High resource utilization
[69]Min-minStaticTask/Resource SchedulingTask/Resource LBOptimizationNot SpecifiedLow makespanLow throughput and scalability
Low response timeHigh SLV and task rejection ratio
High resource utilizationPower inefficient
[70]Max-MinStaticTask SchedulingTask LBOptimizationO (mn)2High throughput and scalabilityLow resource utilization
Low fault toleranceLow degree of balance
Low overheadHigh makespan
[71]Round RobinDynamicTask SchedulingTask LBHeuristicNot SpecifiedLow makespanLow fault tolerance
Low power consumptionLow degree of balance
Low SLVLow resource utilization