A Critique of the Proposed National Education Policy Reform
Eomas cloud erp query flow control simulation
1. +
CLOUD ERP QUERY FLOW CONTROL
SIMULATION WITH QUALITY
RESTRICTIONS AND PROFIT GAINING
CRITERIA
Victor Romanov, Alexandra Varfolomeeva
EOMAS 2013
17 June 2013, Valencia, Spain
In conjunction with CAiSE 2013
2. +
Introduction
Cloud ERP solutions, providing business processes
automation and improving visibility across the entire
organization, are the highest growth segment of the ERP
software industry.
One of the important challenges faced by cloud ERP
providers is the effective management of cloud services
performance.
Cloud ERP clients and cloud service providers pursue
different goals:
Users want to minimize their expenses while meeting the
cloud ERP performance requirements.
The ultimate goal of a cloud service provider is the
maximization of its profit through reducing the number of
quality of service violations and decreasing service costs.
3. +
Worldwide SaaS Market Size
Source: IDC
Content Management and Data Capture in the Cloud Computing 3
25.3% CAGR Through 2014
$ 0.00
$ 5.00
$ 10.00
$ 15.00
$ 20.00
$ 25.00
$ 30.00
$ 35.00
$ 40.00
$ 45.00
2009 2014
$ 13.10
$ 40.50
billions
4. +
4
2. Rapid Elasticity
1. On Demand
Self-service
3. Resource
Pooling
4. Measured
Service
Service Models
Cloud Essential Characteristics
What Does the Cloud Mean?
Deployment Methods
SaaS
Paas
IaaS
Private cloud
Community cloud
Public cloud
Hybrid cloud
5. Broad Network
Access
6. + Cloud Service Provision Model
Clients
SaaS
Provider
IaaS Provider
Request for Cloud
ERP Service
Cloud ERP
Response
Process/Reject
Cloud CRM
Other
Software
Application
Admission
Control
Scheduling
Platform Layer
Application Layer
Request/Response on VMs Schedule on VMs
VM
VM
ServersDatacenter
7. +
Cloud Services Provision Model
Clients
SaaS Provider
On the clients’ side, a request for cloud ERP service is sent to a
SaaS provider’s application layer.
The utility function for each service user, U(B,D), is a function of
budget, B, that client is going to spend on the service and
deadline, maximum time user would like to wait for response, D.
Cloud service providers are trying to ensure the quality of service
provided and to reduce the service response time.
where ωi – is the number of requests, τi – is the service time.
The cost of processing the query should not exceed v0:
8. +
Cloud Services Provision Model
The profit gaining criteria are the following:
Service Initiation Time: How long it takes to deploy a VM.
Price: How much a SaaS provider has to pay per hour for using a
VM from a resource provider?
Data Transfer Price: How much a SaaS provider has to pay for
data transfer from local machine (their own machine) to resource
provider’s VM or for transfer from resource provider’s VM to local
machine?
Processing Speed: How fast the VM can process?
IaaS Provider
9. +
Cloud ERP Queries Classification
All incoming queries should be divided into several services classes
Different kinds of requests require different types of resources
Business_Queries Level
Business_Queries
Processed Outflow
Business_Queries
Rejected Outflow
Business_Queries
Incoming Flow
Business_Queries Processed
Business_Queries Rejected
Standard_Queries Level
Standard_Queries
Incoming Flow
Standard_Queries
Processed Outflow
Standard_Queries
Rejected Outflow
Standard_Queries Processed
Standard_Queries Rejected
10. + Service Queries Processing
Service requests processing can be determined using the Cobb-Douglas
function, often used by economists to model production.
Let N be the total number of queries that can be processed by Cloud ERP
provider, A – the number business service queries, and B – the number of
standard queries.
The function for business service is expressed as:
where A – the number of business queries, K, capital and L, labour are factors
of business queries production, α-productivity factor, β1+β2=1.
Business class services generate more revenue. The number of business
class requests is about 20%, their service is more expensive and in case of
idle capacity the losses are unacceptable.
The function for standard service is expressed as:
The number of standard queries is about 80% and the service price is much
lower.
11. +
Cloud ERP Provider’s Profit
Let a new user submit a business service request at
submission time to the SaaS provider.
The client offers a specific amount of money (Budget) to SaaS
provider.
Let Business Service Price be the total cost incurred to the
SaaS provider by processing the user request.
‘Business Service Profit’=’Business Clients
Budget’–‘Business Service Price’
Data Transfer Business
Cost
VM Business Cost
Business Request
Processing CostBusiness Penalty Cost
Business Service Price
12. +
Cloud ERP Query Flow Control Simulation
Business Queries Level
Business Queries
Processed Outflow
Business Queries
Rejected Outflow
Total Business Service
Cost
Data Transfer Business
Cost
VM Business Cost
Business Clients
Budget
Business Request
Processing CostBusiness Penalty Cost
Business Service Price
Business Service Profit
Business Queries
Incoming Flow
Business Queries Processed
Business Queries Rejected
Standard Queries Level
Standard Queries
Processes Outflow
Standard Queries
Rejected Outflow
Total Standard Service
Cost
Standard Data Transfer
Cost
VM Standard Cost
Standard Clients
Budget
Standard Request
Processing Cost
Standard Penalty Cost
Standard Service Price
Standard Service Profit
Standard Queries
Incoming Flow
Standard Queries Processed
Standard Queries Rejected
Total Profit
Service Class Control
Rate
13. +
1 фев 3 фев 5 фев 7 фев
2
3
4
Queries/min
BusinessQueries
IncomingFlow
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
0
5
10
15
20
25
Queries/min
StandardQueries
IncomingFlow
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
(USD60 000)
(USD30 000)
USD0
Business Service Profit Standard Service Profit
For evaluation purposes only!
Service Class Rate
0 20 40 60 80 100
%
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
(USD150 000)
(USD100 000)
(USD50 000)
USD0
TotalProfit
For evaluation purposes only!
Example 1
14. +
1 фев 3 фев 5 фев 7 фев
2
3
4
5
6
Queries/min
BusinessQueries
IncomingFlow
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
0
5
10
15
20
25
30
Queries/min
StandardQueries
IncomingFlow
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
(USD300 000)
(USD200 000)
(USD100 000)
USD0
Business Service Profit Standard Service Profit
For evaluation purposes only!
Service Class Rate
0 20 40 60 80 100
%
For evaluation purposes only!
1 фев 3 фев 5 фев 7 фев
(USD300 000)
(USD200 000)
(USD100 000)
USD0
TotalProfit
For evaluation purposes only!
Example 2
15. +
Conclusion
Although cloud ERPs are widely deployed, the dynamic cost-effective
service requests processing with the reduced number of SLA
violations is still one of the most intractable problems of cloud service
providers.
Profit gaining is the key driving force for service providers, and the
quality of service provided is the focus of clients’ attention.
Previous approaches do not take into consideration the personalized
features of the client’s request and the elasticity of SLA properties in
cloud computing scenario, which can be used for increase the profit
of cloud service providers.
In order to improve this situation, we first establish a cloud ERP query
flow control model, where incoming queries are divided into several
classes, taking in account SLA constraints and profit gaining criteria.
In our research we have built a cloud ERP simulation model that is
based on macroeconomic sub-model, expressing the production
function, and stochastic sub-model, presenting stochastic the flow of
queries.
17. +
CLOUD ERP QUERY FLOW CONTROL
SIMULATION WITH QUALITY
RESTRICTIONS AND PROFIT GAINING
CRITERIA
Victor Romanov, Alexandra Varfolomeeva
EOMAS 2013
17 June 2013, Valencia, Spain
In conjunction with CAiSE 2013
Editor's Notes
Compound annual growth rateSaaS and cloud ERP solutions are becoming an extremely popular alternative to traditional ERPs. Companies such as Salesforce[7], Plex Systems[5], and Netsuite[3] all continue to show robust increases in revenue while traditional, on-premise ERP systems are showing much slower rates of growth. As outlined in Panorama’s 2012 ERP Report, the market share of cloud-based ERP systems has grown from 6% to 16% in a single year. Even the traditional ERP vendors are redirecting R&D dollars to cloud and SaaS ERP solutions. By 2015, nearly $1 of every $6 spent on packaged software, and $1 of every $5 spent on applications, will be consumed via the SaaS model. And by 2015, about 24% of all new business software purchases will be of service-enabled software, and SaaS delivery will constitute about 13.1% of worldwide software spending across all primary markets and 14.4% of applications spending.Although cloud ERPs are widely deployed, the dynamic cost-effective service requests processing with the reduced number of SLA violations is still one of the most intractable problems of cloud service providers. Cloud service providers and cloud ERP clients pursue absolutely different goals; providers want to maximize revenue by achieving high resource utilization, while users want to minimize expenses while meeting their performance requirements. However, it is difficult to allocate resources in a mutually optimal way due to the lack of information sharing between them.This paper represents cloud ERP query flow control simulation model. The model is designed in Powersim and is based on the incoming query flow control mechanism and cloud services demand forecast. The main goal of the current research is to support cloud provider’s decision-making process of resource allocation in order to provide SLA-aware profit optimization. The rest of this paper is organized as follows: Section 2 presents literature review; Section 3 describes simulation methods overview, in Section 4 the basic principles of cloud computing are provided; Section 5 contains cloud services provision model; Section 6 provides cloud ERP flow control model with SLA constraints and the revenue function of cloud service provider; Section 7 presents the simulation experiment results and comparative analysis; Section 8 concludes the paper. Our simulation experiments show that proposed model improves virtual resource utilization and increases profit of cloud service providers.Panorama’s 2012 ERP Report: http://panorama-consulting.com/resource-center/2012-erp-report/Robert P. Mahowald, Amy Konary, Connor G Sullivan. Market Analysis Perspective: Worldwide SaaS & Cloud Services, 2011: New Models for Delivering Software http://www.idc.com/getdoc.jsp?containerId=232239
Дорого high quality level – increased service cost