On Cost and Energy Efficient Resource Allocation and Provisioning in Cloud Computing Environment

PhD Thesis Proposal Defence


Title: "On Cost and Energy Efficient Resource Allocation and Provisioning in 
Cloud Computing Environment"

by

Mr. ABADHAN SAUMYA SABYASACHI


Abstract:

Cloud computing offers data center resources on-demand for hosting applications 
as a utility, which is a significant shift in the IT service delivery model. 
That helps businesses and organizations to access complex and expensive IT 
facilities offered by the cloud service provider (CSP) without large up-front 
investment to establish their own IT in-infrastructure. CSPs such as Amazon Web 
Services (AWS) offer excess compute capacity as spot instances at a much lower 
cost compared to regular On-demand models. The spot instance prices change 
dynamically as per the long-term trend in supply and demand. The cloud user can 
submit their request for a spot instance with a maximum price, and if it 
exceeds the current spot price and also the spot instance is available then the 
user will be assigned the spot instance to run its jobs. While executing a job, 
the spot instance can be revoked abruptly at any time when the current spot 
price rises above the user’s maximum price or there is an increase in the 
demand for fixed-price On-demand instances. In such a scenario, the cloud 
provider does not assure any SLA guarantee for the user’s job execution. 
Therefore, how to complete job execution reliably and cost-effectively in such 
an uncertain and dynamic cloud computing environment is challenging.

Our thesis proposal report focuses on the application of cost-effective 
resource allocation and provisioning in the cloud computing environment, 
particularly on the Infrastructure-as-a-Service (IaaS) cloud with the prices 
offered are important factors. Reliable completion of the computing jobs 
through Amazon spot instances (SIs) with proper bargaining is challenging. 
Therefore, an SI bidding system is developed for deadline-constrained jobs 
considering both the conditions of the market and the condition of the user. 
The system tries to bargain with the provider by bidding low when the task is 
not urgent. After that, the system increases the price or the price 
distribution gradually when the progress is lower than required. To calculate 
the bid distribution, we compute the probability density of the price after 
five minutes. Then, we apply our developed equations to compute bid prices from 
the probability density function. Equations are easily interpreted by both 
humans and machines. We also consider long-term probability distributions of 
the prices for the reliable completion of the job. Tasks with several days’ 
deadlines are prescribed to bid considering the daily price curve. According to 
the evaluation of Amazon SI price, the proposed system effectively saves 
79%-87% for jobs with several hours of deadlines. It saves 82%-100% for jobs 
with several days' deadlines compared to the on-demand instances. Moreover, our 
algorithm helps all bidders by keeping the price low. We are proposing methods 
from both the perspectives of the cloud provider and the user. Therefore, the 
users are suggested to choose different maximum prices based on the nature and 
urgency of their job so that they can both negotiate and finish their job on 
time. To evaluate our dynamic pricing and spot instance acquisition strategies, 
we have analyzed real-world cloud traces derived from Amazon EC2 spot price 
history. We also discussed the advantages from both the users' and providers' 
points of view for achieving job resiliency in the dynamic cloud computing 
environment.

Cloud computing supports the fast expansion of data centers; therefore, energy 
and load balancing are key concerns. The growing popularity of cloud computing 
has raised power usage and network costs. Frequent calls for computational 
resources might cause system instability. Load balancing in the host requires 
the migration of virtual machines (VM) from overloaded to underloaded hosts, 
which affects energy usage. The proposed cost-efficient whale optimization 
algorithm for virtual machine (CEWOAVM) the technique places migrating virtual 
machines. CEWOAVM optimizes system resources like CPU, storage, and memory. To 
solve this problem, the study proposes energy-aware virtual machine migration 
with the WOA Algorithm for dynamic cost-effective cloud data centers. The 
experimental results showed that the proposed algorithm saved 18.6%, 27.08%, 
and 36.3% of energy compared with the PSOCM, RAPSO- VMP, and DTH-MF algorithms 
respectively. It also showed 12.68%, 18.7%, and 27.9% improvements for the 
number of virtual machine migrations, and 14.4%, 17.8%, and 23,8% reduction in 
SLA violation.


Date:			Tuesday, 8 November 2022

Time:                  	3:00pm - 5:00pm

Venue: 			Room 1410
 			Lifts 25/26

Committee Members:	Dr. Jogesh Muppala (Supervisor)
 			Prof. Dimitris Papadias (Chairperson)
 			Dr. Tristan Braud
 			Prof. Andrew Horner


**** ALL are Welcome ****