Media content in its digital form has scaling up every day, so cloud computing is gating more popularity. Additional features like, omnipresent access, further service creation, discovery of services and resource management also play an important role in cloud computing. Matured and standardized, inter-cloud computing is supposed to provide services which would be more scalable, better managed, and efficient. Such tasks are provided through a middleware unit called as cloud broker. A broker is responsible for reserving resources, managing them, discovering services according to customer’s needs, Service Level Agreement (SLA) cooperation, and match-making between the involved service provider and the customer. Exceptional feature of this study is that consideration of dynamic management of customer’s characteristics and historical record in evaluating the economics related factors. Additionally, a mechanism of encouragement and penalties is provided, which helps in trust build-up for the service providers and customers, prevention of resource underutilization, and profit gain for the involved things.
The framework is modeled on Amazon Elastic Compute Cloud (EC2) On-Demand and Reserved Instances service pricing. For some features required in the framework, data was collected from Google Cluster trace. The work is the initial step towards studying the behaviors and strategies of cloud service providers, brokers, and end customers when offering or facing a pricing model with volume discounts. This paper studies how a broker can schedule the jobs of users to weight the pricing model with volume discounts so that the maximum cost saving can be achieved for its customers by using. Randomized online stack-centric Algorithm (ROSA).
1.1 State Of Art
With the help of proper resource allocation the service cost can be reduced to proper amount without affecting the performance. The customers are generally charged on hourly basis in current pay as you go billing mechanisms which is subject to unfairness. If pricing is unfair, it becomes a reason of disappointment for the cus tomers and as a result, service providers would fail to gain the faithfulness of its customers. In addition to that, service underutilization also depends upon the loyalty of the customers. Underutilization can be opposed by having a dynamic and customers historical record based resource management and pricing mechanism, which ensures resource allocation according to the type and behaviour of the customers.
Cloud computing faces some other helpless challenges as well. Cloud federation or inter-cloud computing has been proposed to offer better stability, availability, cost efficiency and Quality of service (QOS). Though the research on intercloud computing is still not advanced enough, its efficiency cannot be denied in any way. CSPs have their consumers spread all over the world. CSPs have to establish multiple data canters at different places to provide services to their consumers. Usual systems are not accomplished enough to manage the load distribu-tion among data centers to define optimal location for presenting services to achieve expected performance. Consumers geographic location cannot be projected many times; hence load man-agement and service broadcasting have to be completed automatically. Inter-cloud computing offers measurable provisioning of services with steady performance, under variable load and rapidly varying requirements. Inter-cloud computing helps vibrant growth and reduction of re-sources and dealing with erratic demands. A brokers duty is to sort suitable CSP, according to the requirements of its consumers, through cloud exchange. A broker negotiates with gateway to allocate resources such that users and service requirements are fulfilling properly. There is no complete existing brokerage model which could handle all the significant jobs, including resource calculation, arrangement, billing, and refunding, particularly on the basis of changing features. customers, collapse in service quality and different conditions faced during service provision. There is a cloud brokerage model for dynamically managing resource pricing and refunding. The resource pricing and scheduling can be done by taking the historical record of customer in consideration, as the demands are vibrant and may change any time it is essential to the broker to choose what quantity of resources has to booked on demand and perform their dynamic pricing.
By resource allocation the amount of cost can be reduced by right amount, without having any effect on performance. If this system takes costumers viewpoint in consideration, honesty is a major worry in resource allocation and pricing. The customers are usually charged on hourly basis in current pay as you go billing mechanisms which is subject to unfairness. There are two types of pricing fairness, Personal fairness and social fairness in the terminology of economics. If pricing is unfair, it becomes a reason of dissatisfaction for the customers and as a result, service providers would fail to gain the loyalty of its customers. In addition to that, service underutilization also depends upon the loyalty of the customers. It is predictable that many datacenters experience 5 to 20 percent of utilization of their total resources, which shows that the datacenters are significantly underutilized. Underutilization can be countered by having a dynamic and customers historical record based resource management and pricing mechanism, which ensures resource allocation according to the type and behavior of the customers.
Cloud computing faces some other helpless challenges as well. Cloud federation or inter-cloud computing has been proposed to offer better consistency, accessibility, cost efficiency, and Quality of Service (QoS). Though the research on inter-cloud computing is still not advanced enough, its efficiency cannot be denied in any way. CSPs have their consumers spread all over the world. CSPs have to establish multiple datacenters at different places to provide services to their consumers. Usual systems are not accomplished enough to manage the load distribution among datacenters to define optimal location for presenting services to achieve expected perfor-mance. Likewise, consumers geographic distribution cannot be projected as well. Hence, load management and service broadcasting have to be completed automatically. Inter-cloud comput-ing is intended to resolve these issues. Inter-cloud computing offers measurable provisioning of services with steady performance, under variable load and rapidly varying requirements. Inter-cloud computing helps vibrant growth and reduction of resources and dealing with erratic service demands. A brokers duty is to categorize suitable CSP, according to the requirements of its consumers, through cloud exchange. To meet user and service requirements, a broker negotiates with the gateway to allot resources. It is not possible that a single cloud may always achieve the required computing and storage resources demand. Cloud customers are searching for better service at a reasonable price. All such circumstances inspire the beginning of cloud federation and cloud broker. There is no complete existing brokerage model which could handle all the significant jobs, including resource calculation, arrangement, billing, and refunding, par-ticularly on the basis of changing features of customers, collapse in service quality and different conditions faced during service provision. As the demands are vibrant and may vary any time, it is a challenge for the broker to choose what quantity of resources has to be booked on Demand and perform their dynamic pricing. There are several services which charge their customers on hourly clock. One well-known example is Amazon EC2. Even if a resource is utilized for a few minutes, the cost would be for the full hour, which is unfair to the users. Cloud brokerage model for dynamically managing resource pricing and refunding, taking further, in which the focus was on resource estimation. Taking into consideration the historical record of cloud cus-tomers, the model presents a dynamic and fair way of estimating service prices and managing refunds. For customers historical record, its relinquished or service give-up probability, overall resign probability, earned profit, and base value of service is considered. Service can be quit at any stage.
Design a framework for dynamic resource management and resource pricing for public as well as private cloud. The framework will consists of different novel like maintaining quality of service, if degradation in quality then refunding the price, SLA based decided quality, the level of utilization and profit earned from it. Resource Allocation Problem in Cloud
M physical machines are available and their resource capacities given with memory, CPU and Network bandwidth dimensions.
There are N virtual machines to be placed. The requirements of these virtual machines are given with the dimensions of memory, CPU and network bandwidth.
We have to find a mapping between VMs and PMs that satisfies the VMs resource require-ments while minimizing the number of physical machines used.
1.4 Choice of topic with reasoning
Existing work does not focus on overall resource management, pricing, refunding or sim-ilar important features of cloud broker but its only focus is on the reliability of transactions made in purchasing and consuming resources. Dynamic pricing strategy is still a challenge and static pricing strategy is dominating the cloud computing market. Customer behavior and de-mand changes randomly, Dynamic allocation and pricing of resources are required for meeting customer demands and to fairly distribute resources and their pricing.
Design and implement a dynamic resource management randomized online stack-centric scheduling algorithm (ROSA) and resource pricing algorithm for public as well as private cloud
Implement Resource Economics Model by Broker in a dynamic way considering the pre-vious usage record and traits of the customers.
Minimize the problem of resource underutilization for cloud datacenters.
Implement trust based utility for the service provider and the customer. customer is pro-vided with the resources according to its usage behavior , in such a way that more profit could be earned and resource does not go underutilized.
Design refund management model.
1.6 Organization of Thesis
The thesis of project work Entitled ”Resource Pricing and Scheduling of Cloud using Brokers economic model & ROSA” is organized into nine Chapters
In Chapter 1: ”Introduction” is given which comprises of overview Resource Pricing and Scheduling of Cloud with different technique. It defines problems of exiting system. After that it defines why we choose this topic and then we give objective of proposed system.
In Chapter 2: is all about the ”Literature review” where literature review about each area re-lated to proposed system is described. As well as it shows different methodology and technique used by various researches.
In Chapter 3: ”Basic system architecture” is given in a where overview of the proposed system and block diagram of system architecture is described. Here we get outline of proposed system and software, hardware resources required.
In Chapter 4: ”Implementation” of Proposed System is described.
In Chapter 6: ”Conclusion and Future Scope” gives conclusion for system and future work perform on system.
In Chapter 8: ”Paper presented and published” in various conferences and journal.
In Chapter 9:”References and Bibliography” used for development of proposed system.
Cloud Customers Historical Record Based Resource Pricing in this paper they proposed a model which enables customers to shift their workloads across clouds in an easy way. Also, resources can be handled effectively through cloud brokerage which is an advantageous aspect of inter-cloud computing. A unique feature of this study is that they have considered dynamic management of customers characteristics and historical record in evaluating the economics re-lated factors. Additionally a mechanism of incentive and penalties is provided, which helps in trust build-up for the customers and service providers, prevention of resource underutilization, and profit gain for the involved entities. For practical implications, the framework is modelled on Amazon Elastic Compute Cloud (EC2) On-Demand and Reserved Instances service pricing. For certain features required in the model, data was gathered from Google Cluster trace.
Dynamic cloud resource reservation via cloud brokerage  proposed a brokerage service for instance reservation. The authors proposed a brokerage service which is for on-demand reservation or resources for laaS clouds. And their work has limitation to only on-demand jobs while they do not present anything beyond that. Jrad et al. has given generic broker architec-ture. They presented brokers handling of SLA management and interoperability of resources
- Foundation as-a-Service mists offer various estimating alternatives, including on-request and held examples with different refunds to draw in various cloud clients. A practical issue opposing cloud clients is the manner by which to minimize their expenses by picking among various valuing alternatives in view of their own requests. In this paper, they propose another cloud sponsor benefit that saves an expansive pool of occasions from cloud suppliers and serves clients with value rebates. The merchant ideally misuses both valuing advantages of long haul occurrence reservations and multiplexing picks up. They propose dynamic methodologies for the intermediary to reserve occurrence spot with the target of minimizing its administration cost. These procedures influence dynamic programming and surmised calculations to quickly handle huge volumes of interest. Our broad reenactments driven by huge scale Google bunch use follows have demonstrated that huge value rebates can be acknowledged by means of the agent.
According to Dynamic cloud pricing for revenue maximization , dynamic pricing strat-egy is still a challenge and static pricing strategy is dominating the cloud computing market. Customer behavior and demand changes randomly, Dynamic allocation and pricing of resources are required for meeting customer demands and to fairly distribute resources and their pricing. Author is discussing the Amazon, its spot price history drawbacks, and motivation for dynamic pricing development. In distributed computing, a supplier rents its registering assets as virtual machines to clients, and a cost is charged for the period they are utilized. Despite the fact that static estimating is the predominant valuing technique in today’s market, naturally value should be powerfully redesigned to enhance income. The principal test is to outline an ideal element estimating approach, with the nearness of stochastic request and perishable assets, so that the normal long haul income is boosted. In this paper, they make three commitments in tending to this question. To begin with, They direct an experimental investigation of the spot value history of Amazon, and find that shockingly, the spot cost is probably not going to be set by request. This has essential ramifications on comprehension the present market, and rouses them to create and dissect showcase driven element evaluating instruments. Second, they receive an income administration structure from financial matters, and figure the income expansion issue with el-ement evaluating as a stochastic element program. They describe its optimality conditions, and demonstrate critical basic outcomes. At long last, we reach out to consider a nonhomogeneous request display.
Online Resource Scheduling Under Concave Pricing for Cloud Computing  in this pa-per they defined a optimal scheduling algorithm, whose focus is on how a broker can help a group of customers to fully utilize the volume discount pricing strategy offered by cloud ser-vice providers through cost-efficient online resource scheduling. By strategically scheduling multiple customers resource requests, a cloud broker can fully take advantage of the discounts offered by cloud service providers. They present a randomized online stack-centric schedul-ing algorithm (ROSA) and theoretically prove the lower bound of its competitive ratio. Three special cases of the offline concave cost scheduling problem and the corresponding optimal al-gorithms are introduced. simulation shows that ROSA achieves a competitive ratio close to the theoretical lower bound under the special cases. Trace-driven simulation using Google cluster data demonstrates that ROSA is superior to the conventional online scheduling algorithms in terms of cost saving.
A service-oriented broker for bulk data transfer in cloud computing  in this paper they said on the clouds, most of the data-intensive applications are now installed. These applica-tions, storage, and data resource are located in so many different manners that they have to reach even cross-continental networks. Due to this issue, the performance of cloud systems and user requests is affected by performance degradation in networks. The need to ensure service quality, especially for bulk-data transfer, makes resource reservation and utilization a serious issue. Distributed computing develops as new registering standards in which virtualized assets give dependable and ensure administration to clients request. Really, cloud is an administra-tion situated stage since all sort of virtual assets are dealt with as administration to clients. These days, the greater part of dataintensive applications have been produced on cloud frame-work. These applications achieves geologically isolated capacity or information asset with even cross-mainland systems. At that point, the execution corruption of systems will without a doubt influence the cloud application execution and client ask. Keeping in mind the end goal to guar-antee ensure administration of mass information move in distributed computing, the reservation and consolidated assets usage get to be distinctly basic issues which incorporate information and system assets. This issue includes hold and relegate consolidated assets to meet client’s QoS prerequisite. As indicated by this issue, a cloud foundation benefit system (CISF) is proposed to accomplish ensure benefit for information escalated applications in this paper. What’s more, an administration arranged asset specialist (SRB) in light of this system which is proposed to disclosure, select, hold and dole out best consolidated assets. At long last, under client’s QoS requirement dynamic asset choice calculation has been actualized for advancement of consoli-dated assets allotment.
Seamless support of multimedia distributed applications through a cloud  in this paper pair of proxy was introduced , at the users side a client proxy and at the cloud side a server proxy, in order to incorporate the cloud to the wireless unit unlined.
Media cloud: Sharing contents in the large And Media cloud: An open cloud comput-ing middleware for content management ,  these both papers also introduce proxy as a bridge between home cloud to other home clouds and to the outside public media clouds for the purpose of sharing items. This proxy performs multiple tasks of arranging the multimedia items, allowing public cloud to form search database and the categorization of content. The users get opportunity to look for the contents of their own choice through discovery service. However, these studies do not take into account quality degradation or SLA fulfillment.
- in this article the proxy for transcoding and transfer of media is given. While it also suggest usage of peer to peer (P2P) way for transferring media stream outside media cloud. It estab-lishes an intercrossed architecture for P2P as well as media cloud. A large number of resources are needed for transcoding and compression of media. Pereira et al. proposed an architecture in the usage of Map-Reduce model for this purpose, in private and public clouds.
Toward secure and dependable storage services in cloud computing,  in this paper As-sessment for security and privacy in cloud storage was presented. Introduce their work based on access control in cloud federation environment. Also concentrates on data integrity in clouds.
Security and privacy in cloud computing  highlights security and privacy concerns in cloud computing. Rogers et al. has presented resource allocation mechanism but did not con-sider estimation of resources, their valuation, and refund management. they presented a secure mutually verifiable billing system to resolve different future disputes. Their work does not focus on overall resource management, pricing, refunding or similar important features of cloud bro-ker but their only focus is on the reliability of transactions made in purchasing and consuming resources.
|5||Agreed Contract Between Customer and provider|
Resource Allocation in cloud is the process of assigning available resources to the needed user/requester over the internet.
Figure 3.1: The system model is comprised of two or more data owners and a cloud.
Cloud federation or inter-cloud computing is a term referred to a situation when two or more clouds communicate with each other or another intermediary comes into play and federates the resources of two or more clouds. This is done to cater the increasing users demands of media content. The intermediary between two clouds is called a cloud broker or simply broker. Br ker is the entity which introduces the cloud customer to the CSP and vice versa. Inter-cloud resource management is also known as Cloud Broker Aggregation. A cloud broker helps in managing, controlling, and monitoring multiple clouds and share resources. It helps the cus-tomer in finding out the best provider and service according to its need and specified SLA. Thus, a broker benefits both the parties. Broker manages commercial services by using a cost management system which includes Application Programming Interfaces (APIs) and a standard abstract API. Different modules as shown in fig 1 perform a specified task, the detail explana-tion is as follows.
- Registration of new services is handled by Service Registration Manager.
- Deployment Manager deploys services and makes them available. Similarly, each module has its own specific utility.
3.3.2 Online Resource Scheduling(ROSA)
An efficient online scheduling algorithm with a positive, non-decreasing and concave cost func-tion F (.) is introduced. The basic idea of this online algorithm is to stack the processing times of multiple jobs whenever possible and run the jobs with the maximum possible resource in order to reduce the total cost. We will prove the lower bound for the competitive ratio of the proposed online algorithm against the optimal schedule. The online resource scheduling prob-lem assumes that, at any time instant t, the scheduler only knows the tasks which arrive upon or before t. The scheduler does not trust on any knowledge of future information. Online task scheduling is required in many cases, because the cloud service provider or service broker may not have information of all tasks in advance and has to make decision with information available so far.
- while while Q is not empty do
- t = Q.poll();
- bot = getBoT(t);
- if no provisioning plan RP exists for bot then
- r = remaining budget ;
- Tr = remaining tasks ;
- distributeBudget(r,Tr) ;
- update bot budget;
- update t.level spare budget ;
- create provisioning plan RPbot;
- Tr = remaining tasks ;
- if there is an idle VM vmidle bot VMidle bot then
- schedule(t,vmfree bot ) ;
- else if there is a suitable general purpose idle VM vmidle gp VMidle gp then
- schedule(t,vmidle gp ) ;
- if bot BoThom and bot.hasRemainingVMQuota() then
- vmType = RPbot.nextVMTypeToLease() ;
- vmnew = provisionVM(vmType) ;
- else if bot BoThet then
- vm = RPbot.getVmForTask(t)
- if vm is not leased then then
- vm = provisionVM(vm.vmType)
- schedule(t,vm) ;
- else if bot BoTsin and t.level spare budget is enough to lease RPbot.VMType then
- vmnew = provisionVM(RPbot.VMType) ;
|31||schedule(t,vmnew);||33 / 66|
|Department of Technology Shivaji University, Kolhapur|
Figure 3.4: Algorithmic steps
3.3.3 Inter Cloud Gateway
- A composite and flexible system of services provides assistance to inter-cloud. Inter-cloud Gateways are responsible for interoperability and transcoding related tasks.
- Inter-Cloud Exchanges (ICX) are responsible for introducing attributes of cloud environ-ment for inter-cloud computing. These systems are responsible for aggregating infrastruc-ture demands from the broker and match them. Inter-cloud
Root contains services like, Naming Authority, Directory Services, Trust Authority, etc. Cloud Service Customer (CSC) can directly access CSP(s) as well but in that case, transcod-ing related tasks, SLA cooperation, and match-making are done by the CSC itself. Since services are not yet standardized, it makes it difficult to compare the services provided by the CSPs.
3.3.4 Brokers API
Brokers resource economics model: The billing of broker is very helpful. Pay-as-you-go billing model is one of its attributes. It helps the customers to scale their requirements and then pay accordingly. CSC contacts cloud broker to acquire the required service(s) at best price. Broker performs negotiation and SLA tasks with the CSP. After the contract is finalized, the broker not only provides its services on ad hoc basis, but also predicts and allocates the consumption of resources in advance. Prediction and pre-allocation also depends upon user behavior and its probability of using those resources in future . For this purpose, broker performs pricing and billing.
In other words, cloud services are not really commodity services yet. Though it is possible to identify a common set of core services that CSPs are expected to provide. Even if the services provided by different CSPs have same set of functionalities, there can be substantial difference in terms of their convenience, accessibility, and ease of use. Reputation also plays an important role here. Well-known service providers like Amazon are more likely to be chosen over some less known but cheaper service provider. Brokers responsibility here is to do match-making and provide the pros and cons of the service in detail to the customer. With the dynamic and historical record based resource management provided the concerns discussed above can be somehow addressed, since cloud customers are also treated according to their loyalty and the acquired quality, not the expected quality, of the service.
4.1.1 Methods of data collection
Different dataset collected from the following data source for data analysis
- Data sets
For assessing practical implications of this methodology, the modeling can be performed on Amazon EC21 On-Demand and Reserved Instances services. For certain historical records of cloud customers, Google Cluster2 trace of 41 GB, 12,000 machines, for a month-long period used in this model. The required data sets can be downloaded directly from the site http://aws.amazon.com/ec2 all the history of customers is available on this site.
4.1.2 Probable methods of data analysis
The set of experiments aims at evaluating the performance of ROSA under a generic con-cave cost function in comparison to other conventional online scheduling algorithms using the Google cluster trace data . The data required for analyzing this system can be downloaded from the site https://code.google.com/p/googleclusterdata/ the analysis will be done on the data available from the history of cloud customer.
4.2.1 Software Requirement Details
- Operating system: windows XP/7/8.
- Programming language :JAVA Script,PHP.
- Processor: PENTIUM IV-2.7 GHZ and above.
- RAM: 1 GB DDR min.
- Hard Disk: 20 GB and above
4.2.3 Software Tools
- Macromedia DreamViewer.
4.3 Data Flow Diagram
The data flow diagram is a graphical representation of flow of data between different pro-cesses of project. A process is a business activity where manipulation and transformation of data takes place. A data store represents storage of persistent data required. Data flow repre-sents flow of information with its direction represented by an arrow head that shows end of flow. DFD Level-0
Figure 4.1: DFD Level -0
Figure 4.2: DFD Level -1
Activity diagram represents flow from one activity to another activity. The activity can be described as operation of system. Activity is a particular operation of the system. The activity diagram is used to draw the activity flow of a system. It is used to describe the sequence from one activity to another and it describes the parallel, branched and concurrent flow of the system. The main element of an activity diagram is the activity itself. An activity is a function performed by the system. After identifying the activities we need to understand how they are associated with constraints and conditions. The figure below shows different activities and their corresponding relationship between them.
Figure 4.3: Activity diagram
A use case diagram is a graphic depiction of the interactions among the elements of a system. A use case is a methodology used in system analysis to identify, clarify, and organize system requirements.
Figure 4.4: Use Case Diagram
|Table 5.1: Test Cases-Booking|
|Sr. No||Function to be||Input||Expected Output||Actual Output||Remark|
|Table 5.2: Test Cases-Login|
|Sr. No||Function to be||Input||Expected Output||Actual Output||Remark|
|Directly enter is||Please enter userid||Message Please enter|
|001||Blank textboxes||userid and passwordis||Pass|
|Directly enter is||Please enter userid||MessagePlease enter|
|002||Blank textboxes||userid and password||Pass|
|Incorrect User Type|
|Valid username||Enter Invalid||or User Name or|
|003||username||Login Fail||Password Please||Pass|
|and password||Enter Correct|
|Valid username||Enter Invalid||Home|
|004||username||Login Successfully||Page For Particular||Pass|
The screen displays the available Resources for the cloud.user can request resources only those
Figure 5.2: Resources Availability
available for that cloud.
After login with credential users are able to reserve resources for specific time.the screen show
Figure 5.3: Schedular
different types of slots ie,reserved ,pending,restricted etc.user can select and reserve available resource for particular time .
This screen shows graphically how many user allocated which resources and its time of allo-
Figure 5.4: Scheduling in the form of chart
user can check Reservation Details For current day by using this screen.
Figure 5.5: Reservation Details For current day
user can check Reservation Details For weekly by using this screen.
Figure 5.6: Total weeks reservations
here this screen used to check Reservation Details For 10 users Historical record .
Figure 5.7: Historical record of top 10 users
As per the requested resources cost of used resources will get calculate and displays to the
Figure 5.8: Cost utilization of resources
user.cost refunding also considered to improve cloud service and to increase benefits to user.
This page is used for reserve resources for any registered user.all requests are stored and process by ROSA to allocate maximum resources and process all request simultaneously.
Email will send to all users who have reserved resources and also those users having system enable to process due to unavailability of resources,time conflict etc.
|Resources||Service Price||Utilized Resources||Value Utilized||Refunding Cost|
|(Per Day)||(Per Hours)||Resources|
The analysis of system is done on the basis of Resource Pricing and Scheduling of Cloud,and system implemented as economic model using ROSA.the refunding cost has been calculated on the basis of resource utilization.above table shows RAM used only 2/hr in day so refunding amount will be 916,and actual utilised amount will be 83.33.
Figure 6.1: Demand service refunding of cost
N physical machines are available and their resource capacities given along memory, CPU and Network bandwidth dimensions. There are M virtual machines to be placed. The require-ments of these virtual machines are given along the dimensions of memory,CPU and bandwidth.
|Resource||Resource Availability||Resource Demand||Resource||Accuracy(%)|
Figure 6.2: Scheduling of the cloud
The graph shows number of Resources Availability,scheduled and demand of resources.suppose demand are greater than available resources system not able to process the request,it generate alert and send by email.in table 6.2 bandwidth and speed demand is high so system not able to process these two requests.
Figure 6.3: Accuracy of the scheduled resources
Table 6.3 shows accuracy of the system to reserve Resources.here system gives 98% accu-racy.if resources and demand are equal.
|No. of Resource Demand||Firstfit||ROSA|
Figure 6.4: Execution Time Firstfit and ROSA
Graph 6.4 used to compare two technique for resources scheduling Firstfit and ROSA.Rosa having more filtering steps and also included Firstfit to check possibilities of requests so ROSA used more time as compare to simple Firstfit.
|Request from||Processed||Unprocessed||Unprocessed Request||Scheduled|
|Request due to||Accuracy %|
|25 users||Request||due to,less Volume set||requests|
Figure 6.5: Analysis N Request with n Users using ROSA
Graph 6.5 used to analyze N Request with n Users using ROSA.the accuracy of the algo-rithm is 100% if requests and available time period (available resources is equal).Graph 6.5 schedule all resources on the basis of volume and Firstfit basis using ROSA.
This framework gives Resource management, pricing, similar important features of cloud bro-ker.It focuses on problems faced by customer while purchasing different cloud services. Here dynamic resource allocation system is proposed, which will help to meet customer demand and fair distribution of resources with competitive price. The system implements Randomized online stack-centric algorithm (ROSA) and brokers resource economics model to fulfill cus-tomers demand. The system will be more useful and benefit cloud customer than current static scheduling system.
7.2 Future Scope
Finding a trade-off between the gain from volume discounts and the induced security risks is also an interesting research problem.