optimization of resource provisioning cost in cloud computing
DESCRIPTION
In cloud computing, cloud providers can offer cloud consumers two provisioning plans for computing resources, namely reservation and on‐demand plans. In general, cost of utilizing computing resources provisioned by reservation plan is cheaper than that provisioned by on‐demand plan, since cloud consumer has to pay to provider in advance. With the reservation plan, the consumer can reduce the total resource provisioning cost. However, the best advance reservation of resources is difficult to be achieved due to uncertainty of consumer's future demand and providers' resource prices. To address this problem, an optimal cloud resource provisioning (OCRP) algorithm is proposed by formulating a stochastic programming model. The OCRP algorithm can provision computing resources for being used in multiple provisioning stages as well as a long‐term plan, e.g., four stages in a quarter plan and twelve stages in a yearly plan. The demand and price uncertainty is considered in OCRP. In this paper, different approaches to obtain the solution of the OCRP algorithm are considered including deterministic equivalent formulation, sample‐average approximation, and Benders decomposition. Numerical studies are extensively performed in which the results clearly show that with the OCRP algorithm, cloud consumer can successfully minimize total cost of resource provisioning in cloud computing environments.TRANSCRIPT
OPTIMIZATION OF RESOURCE PROVISIONING
COST IN CLOUD COMPUTING
Aswin K K
S8 CSE-A
MKAJECS009
Guide:
Mr. Sunil Kumar P V
Contents• Overview of Cloud Computing
• Challenge of Resource Provision in the Cloud
• Optimal Cloud Resource Provisioning
• OCRP Model
• Provisioning Phases
• Provisioning Stages
• Reservation Contracts
• Uncertainty
• Benders Decomposition
• Sample-Average Approximation
• Numerical Results: Provisioning Cost
Overview of Cloud Computing
• Large distributed system
• Large pool of resources
• Multiple provider
• Multiple data-centers
• Virtualization
• Internet access
• Pay-per-use basis
• Provisioning options/plans
• On-demand
• Reservation
• Example: Amazon
Overview of Cloud Computing: Provisioning
Plans• Reservation can reduce the total provisioning cost
• On-demand (Small Instance): 0.085 x 365 x 24 = $744.60 for 1yr contract
• Reservation: 227.50+(0.03x365x24) = $490.30 for 1yr contract or 34% cheaper but 49% cheaper for 3yr contract
Source: http://aws.amazon.com/ec2
Challenge of Resource Provision in the Cloud
• Resource provision = activity to provide / supply resource (to accommodate users/systems)
• Goal: How many VMs (i.e., how much resource) do we need to provision in advance (i.e., provision with reservation plan) ?
• Challenge
• Multiple cloud providers and Quality of Service(QoS) & Service Level Agreement(SLA)
• Multivariate uncertainty e.g., demand, price, availability
• Optimal solution under uncertainty
• Computational complexity
VM = Virtual Machine
Challenge of Resource Provision in the Cloud:
Uncertainty• Uncertainty of price
• On-demand price might be fluctuated
• Uncertainty of availability
• Free / cheap resources offered by a cloud provider might be provided based on weak SLAs
• Internet bandwidth is not reliable until cloud resources might not be accessible
Challenge of Resource Provision in the Cloud:
Uncertainty
• Uncertainty of demand
Optimal Cloud Resource Provisioning
• OCRP algorithm is proposed
• To minimize the expected resource provisioning cost in multiple provisioning stages e.g., 4 stages in quarter plan, 12 stages in 1-Y plan, 36 stages in 3-Y plan, etc.
• To consider multivariate uncertainty
• Optimal solution is obtained by formulating and solving stochastic integer programming with multi-stage recourse
• Techniques to solve OCRP: deterministic equivalence, benders decomposition, sample-average approximation
• Several experiments show that OCRP can minimize the cost under uncertainty
OCRP Model
• System model of cloud computing
Provisioning Phases
• Provisioning phase: time interval when resources need to be provisioned or utilized1. Reservation phase: reserve resources
2. Expending phase: utilized the reserved resources
3. On-demand phase: provision more resources on-demand
Provisioning Stages
• Provisioning stage: time epoch when cloud broker makes a decision
• Examples
• Two stages: current and future (e.g., now and next month)
• Twelve stages: Yearly plan = January to December
Reservation Contracts
• Reservation contract: signed contract stating the time duration of availability of reserved resource
• During the contract period, price is discounted
• Examples: 3-month (K1) and 6-month (K2) contracts
Uncertainty• Stochastic programming requires uncertainty
parameters, namely scenarios given by set Ω
• Scenarios can be described by a probability distribution
• Set Ω has finite support with probabilities p(ω) Є [0,1] where ω=(ω1,…, ω|T|) Є Ω
Ω = ∏ Ωt = Ω1 x Ω2 x…x Ω|T|tЄT
Benders Decomposition• Benders decomposition
breaks down an optimization problem into smaller problems which can be solved independently (parallel)
• Given the results obtained from master & sub-problems, the lower & upper bounds of solution can be calculated• Convergence bounds checked by zv
(ub) - zv(lb) < Є
where zv(ub) - zv
*(e) and zv(ub) = zv
*(e) – αv + zv*(r) + Σ
zv*(o) (ω)
Sample-Average Approximation• If the number of scenarios (Ω) is numerous, it
may not be efficient to obtain the solution of OCRP
• SAA addresses the problem by sampling N scenarios, then SAA-based OCRP formulation is solved given the N samples
• We modelled OCRP based on SAA approach to choose N that yields the acceptable approximated solution
• SAA can be parallelized as well
Numerical Results: Provisioning Cost
Conclusion• Resource provisioning algorithms based on
stochastic programming and robust optimization have been proposed
• The algorithms can be applied in real world to minimize provisioning costs
• Resource management framework for cloud computing will be composed
Reference• Paper on “Optimization of Resource
Provisioning Cost in Cloud Computing” presented by Sivadon Chaisiri in PDCC Seminar, Parallel & Distributed Computing Centre, Friday 21st, 2011
• Paper on “Cloud Computing for on-Demand Resource Provisioning” presented by Ignacio M Llorente in 7th NRENs and Grids Workshop at Trinity College, Dublin, September 2, 2008
Questions
THANK YOU