Advances, Systems and Applications
Table 5 Cost-aware schedulers and their properties
From: MapReduce scheduling algorithms in Hadoop: a systematic study
Reference | Year | Key Ideas | Advantages | Disadvantages | Comparison Algorithms | Evaluation Techniques |
|---|---|---|---|---|---|---|
Seethalakshmi et al. [59] | 2022 | Real-coded multi-objective genetic algorithm with effective queuing model for efficient job scheduling in heterogeneous Hadoop environment | Reducing the cost. Minimizing the execution time. Improving the resource utilization | No particular | GA, FIFO, fruit fly, firefly | Simulation: MRSIM simulator |
Tang et al. [60] | 2021 | Cost-efficient workflow scheduling algorithm for applications with deadline constraint on heterogeneous Clouds | Reducing the total workflow execution cost | No ensuring the deadline of all jobs was met | LHCM, PCP | Simulation: CloudSim |
Vinutha et al. [61] | 2021 | Budget Constraint Scheduler for Big Data Using Hadoop MapReduce | Reducing the budget. Reducing the completion time | Not considering node failures. Not considering data locality | FIFO, Fair scheduler | Implementation: Hadoop cluster with 12 data nodes |
Javanmardi et al. [62] | 2020 | A unit‑based, cost‑efficient scheduler for heterogeneous Hadoop systems | Reducing the cost of executing jobs. Minimizing the jobs execution times. Improving data locality | No pipelining between Map and reduce phases | FIFO, Fair | Implementation: Hadoop cluster with 6 nodes Simulation: MRSIM simulator |
Rashmi et al. [63] | 2016 | Deadline constrained cost effective workflow scheduler for Hadoop clusters in cloud datacenter | Reducing the cost. Meeting the deadline | They have considered only VM cost, but there are other costs in terms of bandwidth usage, data store usage | SciCumulus adaptive approach | Implementation: cloudsim environment |
Zacheilas et al. [64] | 2016 | ChEsS cost-effective scheduling across multiple heterogeneous Mapreduce clusters | Minimizing the cost. Optimizing the execution time | Not considering data locality | NSGA-II, GDE3, Weighted Sum, Starfish | Simulation |
Palanisamy et al. [65] | 2015 | Cost-effective resource provisioning for MapReduce in a cloud | Reducing the cloud compute infrastructure cost. Reducing job response times | No ensuring the deadline of all jobs was met. Not considering the complex workloads | Dedicated cluster, Per-job Cluster | Simulation |
Chen et al. [66] | 2014 | CRESP: towards optimal resource provisioning for MapReduce computing in public clouds | Minimizing the monetary or time cost | Not considering heterogeneity | No comparison, different input datasets in different possible scenarios | Implementation: 16-node Hadoop cluster and Amazon EC2 |