Advances, Systems and Applications
Table 7 Evaluation of Statistical Methods for Cloud Data Center Resource Management
From: A systematic review on effective energy utilization management strategies in cloud data centers
Author/ Year | Algorithm/Method | Data set/ Workload | Tools/ Experiment Environment | Objective | Performance Metrics/ Pros | Limitations |
|---|---|---|---|---|---|---|
Z. Cao et al. [82], 2012 | EV_MCE | PlanetLab | CloudSim | Host Overload Detection, VM Selection, DVMC, IaaS | Energy Consumption, QoS, SLAV, VM Migrations, SLATAH | Little worse than previous work for EC, simulation, required real infrastructure |
F. Farahnakian et al. [83], 2013 | LiRCUP | PlanetLab | CloudSim | Detection of Overloaded and Underloaded PM, Utilization Prediction | Energy Consumption, SLA Violation, power cost, CPU usage prediction | Simulation, Prediction utilization is approximated as a function |
A. Nadjar et al. [84], 2015 | ARIMA MSV_ML | PlanetLab | CloudSim | Dynamic Consolidation of VM, IaaS | SLA, Energy Savings, Migration count, performance, predicting resource usage, number of active hosts | Required real infrastructure, less accuracy in predicting VM and host resource requirements in the near future |
X. Ruan et al. [85], 2015 | PPRGear | PlanetLab | CloudSim | VM Allocation, VM Migration | Host Utilization, Energy Consumption, SLA, Shutdown times, migration times | Performance degradation, required primitive characteristics of host computers, heavy workload |
A. Abdelsamea et al. [86], 2017 | MRHOD HLRHOD | PlanetLab | CloudSim | PM Overload and Underload Detection, VMC, Better predictions of host overloading | Migration of VM, Energy Consumption, SLA Violation | Complex, required real cloud, less host utilization |
M A Khoshkholghi et al. [87], 2017 | PCM (IWLR,V-VMS,BRB,MRUHD) | PlanetLab | CloudSim | Utilization Prediction, VM Consolidation, workload-independent | Energy Utilization, VM Relocation, SLAV, heterogeneous physical servers, IaaS environment | N/W topology required, performance aware strategy, only CPU usage considered. |
Hemavathy et al. [88] (2019) | PTASC model, Extended Multiple Linear Regression (EMLR) | Own data | CloudSim | workload prediction, load balancing, server consolidation, VM scheduling, VMP, resource provisioning, | energy efficiency, cost reduction, Response time | Required numeric and local architecture into consideration, statistical learning method, intrusion detection, and prevention systems needed. |
Lianpeng et al. [89] (2019) | RobustSLR | PlanetLab and random workload | CloudSim simulator | VM consolidation, host overload, underload detection | Energy consumption, SLAV, SLATAH, PDM, Average SLAV, VM migration, host shutdown | Required RAM and N/W usage to improve energy efficiency and SLAV |
X. Liu et al. [50], 2020 | DCMMT | PlanetLab | CloudSim | VM Migration Thrashing, Dynamic Consolidation | VM Migration, SLA Violation, Thrashing Index, SLATAH, PDM | Real-world cloud platform needed, required workload statistical properties |
Maryam C.-Samani et al. [90] (2020) | PCVM.ARIMA | PlanetLab | CloudSim | DCVM, Resource utilization, VM Placement | Energy Consumption, Migration of VM, SLA Violation, hosts shutdown | Memory and disk utilization are required for future prediction, required reduction in CO2 emission, failure tolerance, and security. |