Project - Evolutionary Algorithms for Multicloud Deployment of Data-intensive Software
Cloud computing allows organisations to save costs by running software on-demand in cloud data centres. Multiple cloud platforms are available with different computing resources in different locations, different pricing schemes and different quality-of-service promises (e.g., response times). Multiclouds are an emerging technology where organisations can dynamically use and share distributed resources from multiple clouds simultaneously to deploy software in a cost-effective manner. Unfortunately, current approaches for selecting resources are not mature. We will investigate the optimisation problems of multicloud deployment, and develop innovative single- and multi-objective evolutionary algorithms for finding the best combination of cloud resources as per their demands
Datasets
Price data:
https://azure.microsoft.com/en-us/pricing/calculator/
Azure2017, Azure 2019:
- collected from Microsoft Azure public cloud platform and representative workload traces across thirty consecutive days
- used in paper Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms https://dl.acm.org/doi/10.1145/3132747.3132772
Bitbrain:
- consists of traces of resource utilization metrics form 1750 VMs running on BitBrain distributed datacenter
- used in paper Statistical Characterization of Business-Critical Workloads Hosted in Cloud Datacenters https://ieeexplore.ieee.org/document/7152512
Code
from ZhengHan:
https://gitlab.com/hz741713526/GPHH_MOOCFD.git 