Use Case Requirements and Functional Requirements (WP6.3: Cloud Bursting)

Transcription

1 Project Acronym: OPTIMIS Project Title: Project Number: Optimized Infrastructure Services Instrument: Thematic Priority: Integrated Project ICT Internet of Services, Software and Virtualisation Use Case Requirements and Functional Requirements (WP6.3: Cloud Bursting) Activity 6: WP 6.3: Integration and Experimentation Cloud Bursting Due Date: M12 Submission Date: 09/06/2011 Start Date of Project: 01/06/2010 Duration of Project: 36 months Organisation Responsible for the Deliverable: FLEX Version: 1.0 Status Final Author(s): Craig Sheridan Tabassum Sharif Francisco Javier Nieto FLEX FLEX ATOS

2 Reviewer(s) Ana Juan Ferrer Srijith Nair ATOS BT

3 Project co-funded by the European Commission within the Seventh Framework Programme Dissemination Level PU Public X PP Restricted to other programme participants (including the Commission) RE Restricted to a group specified by the consortium (including the Commission) CO Confidential, only for members of the consortium (including the Commission). OPTIMIS Consortium Page 2 of 29

4 This is a public deliverable that is provided to the community under the license Attribution- NoDerivs 2.5 defined by creative commons Full licensing information is contained in Annex A. OPTIMIS Consortium Page 3 of 29

5 Table of Contents 1 EXECUTIVE SUMMARY INTRODUCTION PURPOSE GLOSSARY OF ACRONYMS USE CASE SCENARIO BUSINESS CASE ANALYSIS OF THE MARKET EXPECTED MARKET GROWTH EXPLOITATION PLAN STATE OF PRACTICE EXPECTED INNOVATION USE CASE SCENARIO MORE DETAILED DESCRIPTION ACTORS INVOLVED AND THEIR ROLES REQUIREMENTS IDENTIFICATION Identified functional requirements Gaps analysis HIGH LEVEL DESIGN OF THE USE CASE OPTIMIS COMPONENTS INVOLVED BURSTING POSSIBILITIES CONCLUSION REFERENCES ANNEX A. LICENSE CONDITIONS OPTIMIS Consortium Page 4 of 29

6 Index of Figures Figure 1. Gartner 2010 Hype Report...11 Figure 2. E-commerce Transaction...17 Figure 3. High Level Design Workflow...20 Figure 4. Cloud Bursting Interaction...21 Figure 5.OPTIMIS Architecture...22 OPTIMIS Consortium Page 5 of 29

7 1 Executive Summary Deliverable is representative of the work carried out in WP6.3 relating to a Cloud Bursting use case scenario from month The purpose of the work package itself is to manage the provision of an environment capable of supporting the Cloud Bursting use case. The use case involves an autonomous company with their own private cloud who are willing when circumstances are apt, to use resources provided from an external Cloud provider to supplement their own resources. This effectively allows the autonomous company to utilise on-demand elasticity and scalability of resources in an optimal and managed fashion. In the use case, Atos Origin will act as the autonomous company, with Flexiant acting as the external provider who are willing to rent out their available platform resource. More specifically, this deliverable defines the use case requirements and functional description required to build the use case scenario to ensure interoperability between the internal and external systems. A business analysis of the current market is identified in regards to revenue and future projected growth in the parent and similar sectors detailed as well as being contextualized with risk factors. The ways in which the use case scenario can be exploited by relevant players is explored in terms of the areas in which adoption will prove to be advantageous. The innovative areas created by the project relating to the particular use case are discussed in areas such as the monitoring based decision engine which will ensure the external resource currently being used does not breach any agreed SLA, therefore giving a quality of service guarantee. The high level architecture is documented with an emphasis on optimising service construction, deployment and execution for the service lifecycle. OPTIMIS components required for the use case are acknowledged such as the base toolkit TREC components for the evaluation of IPs, self-* management components to conform to agreed SLAs and Provisioning Components to manage resources and licensing. OPTIMIS Consortium Page 6 of 29

8 1 Introduction The purpose of this document is to detail the main aspects of the Cloud Bursting use case referenced in Work Package 6.3 (WP6.3) of the DoW. The document will cover the reasons why the use case is required from a business needs perspective and detailing the justification for such a use case. This will allow contributors to describe the functional requirements of the use case from the Infrastructure Provider (IP) perspective for the Cloud bursting scenario described in the DoW. This will help ensure the interoperability between the internal and external Clouds as well as showing the benefits of renting the extra resources externally and using parallel computation. For doing so, it will necessary to determine what kind of functionalities are necessary in such a scenario like finding required resources, selecting external clouds, performing internal rearrangement of resources, etc. The objective of the Cloud Bursting use case will be to incorporate as many of the OPTIMIS innovations as possible, in order to show that scaling resources is feasible in a way that any users and providers constraint will be respected and that the system performance and the overall quality will be improved. It is important though that the tail does not wag the dog and requirements from the Service Providers and Infrastructure Providers are fed back to feed the development of the toolkit. To showcase the use case, an application will be developed which provides standard e- commerce type transactions. The Business Level Objectives (BLOs) of a Cloud scenario will be clearly laid out. Consideration will also be given to the fact that developed components will have an effect on the work of brokers, independent software vendors (ISVs), and service consumers (end-users). 1.1 Purpose The purpose of this document is to detail the main aspects of the Cloud Bursting use case referenced in WP6.3 of the DoW. The document will cover the reasons why the use case is required from a business needs perspective and detailing the justification for such a use case. The main purpose of this deliverable is to introduce the Cloud Bursting Use Case: Definition of use case scenario Identification of Use Case requirements and functional capabilities Define Business and exploitation aspects for the use case Define a High level architecture for the solution 1.2 Glossary of Acronyms Acronym D DoW WP CB IaaS IP Definition Deliverable Description of Work Work Package Cloud Broker Infrastructure as a Service Infrastructure Provider OPTIMIS Consortium Page 7 of 29

9 SaaS SP CAE Software as a service Service Provider Cloud Aggregation Ecosystem OPTIMIS Consortium Page 8 of 29

10 2 Use Case Scenario This will allow contributors to describe the functional requirements of the use case from the IP perspective for the Cloud bursting scenario described in the DoW. This will help ensure the interoperability between the internal and external Clouds as well as showing the benefits of renting the extra resources externally and using parallel computation. The aim is to demonstrate the Cloud bursting deployment scenarios as detailed in D1.2.1 Architecture Design document. Specifically demonstrating the IP level operations, dynamic resource deployment and SP level operations. For doing so, it will necessary to determine what kind of functionalities are necessary in such a scenario like finding required resources, selecting external clouds, performing internal rearrangement of resources, etc. The objective of the Cloud Bursting use case will be to incorporate as many of the OPTIMIS innovations as possible, in order to show that scaling resources is feasible in a way that any users and providers constraint will be respected and that the system performance and the overall quality will be improved. It is important though that the tail does not wag the dog and requirements from the Service Providers and Infrastructure Providers are fed back to feed the development of the toolkit. To showcase the use case, an application will be developed which provides standard e- commerce type transactions. The Business Level Objectives (BLOs) of a Cloud scenario will be clearly laid out. Consideration will also be given to the fact that developed components will have an effect on the work of brokers, independent software vendors (ISVs), and service consumers (end-users). OPTIMIS Consortium Page 9 of 29

11 3 Business Case There is a clear value to IaaS providers to enhance the possible usage of Cloud platforms by utilising the capabilities of the toolkit. It is envisaged that by maintaining a historical performance of IPs and providing an indication of real-time resource availability and license management that consumers will have the confidence to run their applications on a bursted Cloud environment. For merchants, there is clear value in not having to invest in further hardware to cope with seasonal and event-based traffic which would be idle during normal levels of operation. For an e-commerce website to run their application on an elastic platform for example, resolves the issues of traditional server provisioning, time, contracts, resource prediction, tieins, resource scalability. It will be possible to guarantee good QoS in any situation, as the infrastructure will be adapted in an automatic way to the requirements. Transparency is a key aspect and the look and feel of the system operation should be as close to traditional methods as possible while providing the added functionality in the background. The business case is related to the optimization of the resources available. Starting from a limited set of resources, it will be possible to provide a good Quality of Service (QoS) to several service providers by not needing to increase or improve the available hardware, so the initial investment will be reduced to a level in which the normal usage of services does not overload or push the hardware to its limit, providing resources to as many service providers as possible. The bursting would be only done in those moments where there is a peak of resources need or when there are unexpected problems with some internal resources and the agreed QoS cannot be maintained. This scenario provides an essential solution to e-business by integrating IT solutions to the core of a companies business in an unobtrusive way, provides ways of a company adding value for it's customers, provides ease of use and better experience for customers and of course allowing merchants and service providers inexpensive access to expandable versatile resources without the traditional CAPEX spend associated with this. Acting this way will allow Infrastructure Providers or autonomous companies to maintain a minimum set of hardware resources and reduce other associated costs (refrigeration, electricity, needed room, etc) while benefits are maximized. 3.1 Analysis of the Market Parties can be identified that have a stake in the Cloud bursting scenario, such as the IaaS providers, the on-ramp sector and PaaS providers. On-ramps are the second-largest cloud management segment, representing the 24% of the total revenue generated by cloud management providers (2009). Revenue generated by on-ramps totalled $14m in 2009 and is expected to grow at a CAGR of 66% to reach $108m in Overall Cloud market growth has been widely analysed and has been shown to have a steep upward curve in terms of market size. The Cloud bursting market has not received analysis on potential market size specifically in terms of the size of share that this sector may receive in the projected growth. This is simply due to the OPTIMIS project addressing state of the art areas that market analysts have yet to breakdown. It can be stated though, with some confidence, that the market is OPTIMIS Consortium Page 10 of 29

12 there for Cloud bursting and there is money to be made due to the clear advantages detailed throughout the project for the relevant stakeholders. To give some kind of context to Cloud Computing analysis, Gartner has released its 2010 Hype Cycle Report, identifying those technologies it thinks have reached the "Peak of Inflated Expectations" as well as those languishing in the "Trough of Disillusionment." Interestingly there exist analysis 1 that shows Cloud Computing at around the peak of hype so poses the risk of over enthusiasm and the possibility unrealistic projections, but taken in context these figures should promote the view that the risk of inaccurate projections is at it's height. Over the coming period a more enlightened perspective on Cloud computing predictions should take place where the winners and losers of this sector will start to be exposed and the true market growth becomes clearer. Figure 1. Gartner 2010 Hype Report 1 _the_pea.php OPTIMIS Consortium Page 11 of 29

13 3.2 Expected Market Growth Market growth can be broken down to various levels where a basic conclusion of overall growth can be insinuated but should be treated in the context described in section 2.1. Below are some figures representing expected growth in IaaS, SaaS, onramps, geographical locales and some examples of overall Cloud market growth expectations. This predictive information gives validity to the chosen use case scenario and a confidence that there will be a sizeable potential marketplace. Experton Group 2 is forecasting that the German cloud computing market is forecast to grow from EUR 1.14 billion in 2010 to EUR 8.2 billion in This is equal to average annual growth of 48 percent. In 2015, cloud computing will account for around 10 percent of total IT expenditure in Germany. Around half of revenue in 2015 will be generated from cloud services, with a third coming from investment in cloud infrastructure, mainly data centres. The use of so-called private clouds by businesses will account for EUR 2.6 billion in revenues by 2015, up from EUR 400 million in Source: Gartner 3 predicts worldwide software as a service (SaaS) revenue within the enterprise application software market is forecast to reach $9.2 billion in 2010, up 15.7 percent from 2009 revenue of $7.9 billion. The market is projected for stronger growth in 2011 with worldwide SaaS revenue totaling $10.7 billion, a 16.2 percent increase from 2010 revenue. These market forecasts are included in the report Forecast Analysis: Software as a Service, Worldwide, , Update. Gartner 4 analysts write in the report Predicts 2011: New Relationships Will Change BI and Analytics, that by 2013, 33% of business intelligence functionality will be consumed via handheld devices, and 15% of BI deployments will combine BI, collaboration and social software into decision-making environments. By 2014, 30% of analytic applications will use inmemory functions to add scale and computational speed. In addition, 30% of analytic applications will use proactive, predictive and forecasting capabilities and 40% of spending on business analytics will go to system integrators, not software vendors. All of this is predicated on the security and scalability of cloud-based analytics. Gartner 5 predicts in the report, Top Predictions for IT Organizations and Users, 2011 and Beyond: IT s Growing Transparency that by 2015, 80% of enterprises using external cloud services will demand independent certification that providers can restore operations and data. Also by 2015, 20% of non-it Global 500 companies will be cloud service providers. Companies will generate 50% of Web sales via their social presence and mobile applications according to Gartner in this same report. Infonetics Research 6 is forecasting spending on security-related SaaS applications will experience a compound annual growth rate of 31% through plummer.pdf 6 Highlights.asp OPTIMIS Consortium Page 12 of 29

14 International Data Corporation (IDC) 7 expects the automated software quality (ASQ) and emerging testing as a service (TaaS) segments of the market to generate a 35.9% CAGR from and $954 million in projected revenue in TechMarketView predicts the value of the UK cloud computing market will more than double between now and 2014 from 2.4bn to 6.1bn according to the study UK Software and IT Services Market Forecast published in December by the firm. MarketsandMarkets.com 8 in their report, Cloud Computing Market Global Forecast ( ) predicts that the global cloud computing market is expected to grow from $37.8 billion in 2010 to $121.1 billion in 2015 at a CAGR of 26.2% from 2010 to SaaS is the largest contributor in the cloud computing services market, accounting for 73% of the market s revenues Market 9 Intel Group has published the report, The Future of Virtualization, Cloud Computing and Green IT Global Technologies & Markets Outlook In this report, they predict that the North American and European Virtualization Market Forecast are both going to grow by a healthy 8.5% CAGR over period, creating aggregate markets of $123.8 Billion in North America and $94.7 Billion in Europe over the same period. Pike Research 10 has released the report, Cloud Computing Energy Efficiency which is one of the most ambitious to date in quantifying the sustainability advantages of cloud computing. The analysis states that the adoption of the cloud computing services will lead to a 38% reduction of the worldwide data center energy expenditures by Data center energy cost reductions will lower total data center energy costs from $23.3bn in 2010 to $16.0bn in 2020, and cause a 28% reduction in GHG emissions from 2010 levels is another key finding of the report. It seems clear that there is a general prediction of growth in the Cloud computing sector. What is just as clear is that there is disagreement about the rate of this growth. The commonality among market analysts is for significant growth under the general umbrella of Cloud computing but there is a lack of corelation between forecasts when it comes to specific sectors and services within the area. This can be explained by the Gartner Hype Report which places the Cloud market at optimal levels of possible unrealistic predictions. 3.3 Exploitation Plan Work package 6.3 provides a proof of concept to showcase the use of the OPTIMIS concept of Cloud bursting from one OPTIMIS enabled Cloud to another when extra resources are required with various toolkit related enhancements. It is envisaged that IaaS providers, service providers and subsequently end users will benefit heavily from this scenario as they can offer enhanced value added services as they can offer service guarantees to end users. IaaS providers will benefit directly due to the increased uptake of their infrastructure and the fact that SMEs can compete more easily with enterprise level players and are no longer 7 Id=null&pageType=SYNOPSIS OPTIMIS Consortium Page 13 of 29

15 restricted to the size of their own underlying architecture. As the scope and type of offerings that can be implemented on a typical SME Cloud infrastructure has increased, the size of the market share that is available to the SME increases exponentially. The fact that QoS offerings can be improved and reliability enhanced while optimising the offerings provided to end users within these fields and also with the added benefits of trust, reliability and cost. This has the consequence of bringing a consequence to the market in OPTIMIS enabled infrastructure although the identification of the target markets also enhances the need to promote the value of the enhanced offering. OPTIMIS Consortium Page 14 of 29

16 4 State of Practice Currently, when an end user implements their service in a Cloud environment, a detailed discussion must take place with the Infrastructure provider about the nature of their service, the predicted resource usage over the coming year generally and predicted levels during peak periods. This scenario has many inherent problems such as changing business requirements that are inextricably linked to the hardware requirements. The predicted resource usage can therefore miss the mark by some margin. This can have serious consequences in terms of either paying too much for no longer required hardware while tied into contracts or even worse, the infrastructure provider can no longer cope with the demands of the user resulting in an outage of services while either the infrastructure provider arranges further resources or the end user moves services to an alternative infrastructure provider. As the infrastructure provider is only able to manage its internal resources, it is not possible to scale them in order to solve sporadic problems because of failures or wrong predictions. Even if it is possible to plug new hardware and resources, there will be some time during which quality levels cannot be maintained, as the reaction of the infrastructure provider is very limited and may require manual activities to be carried out. In this situation, the infrastructure provider would violate signed SLAs (if any), loosing clients or even having legal problems because of existing contracts. The process even could require analyzing what happened (in order to determine if it was the client s fault because of wrong predictions) and compensating clients. In academia, projects such as RESEVOIR hace proven the concept of Cloud Bursting but in a pre-configured way without involving the dynamic negotiation proposed in OPTIMIS. The next IEEE International Conference on Cloud Computing (CLOUD ) lists Cloud-bursting as a topic of interest for the submission of papers. Current industry practice involves companies such as Amazon with their EC2 platform and Flexiant with their Flexiscale platform, to offer use of resources via their APIs but with no understanding of consistency, QoS or automatic redirection of requests to another provider. [1] 11 OPTIMIS Consortium Page 15 of 29

17 5 Expected Innovation With the usage of the OPTIMIS platform, it is expected to solve the mentioned problems, increasing the elasticity and improving the system reaction when any issue rises up. Monitoring systems can determine when SLAs are not going to be fulfilled, launching an automatic process for optimizing the available resources without human intervention. If necessary, this system will select an external Cloud platform in order to use external resources when the internal resources are not enough. In the Cloud bursting scenarios as well as the other multi-cloud scenarios, the relationship between the IP and SP goes beyond the state of the art. Previously, as demonstrated in projects such as Resevoir, a bursting scenario was possible as long as all required functions were pre-configured in advance so this merely proved the possibility to rent resource externally but failed to do this in any intelligent dynamic way. In OPTIMIS, a negotiation process takes place where user requirements are noted and an OPTIMAL platform chosen for their needs along with a QoS within agreed SLAs. In short, the external resource is chosen dynamically depending on the needs on the internal cloud and will look to offer an optimal choice depending on those needs. As everything will happen in an automatic way, it will be possible to reduce time to solve any problem, as any decision is taken in an automatic way (unless human intervention is required explicitly). Moreover, whenever possible, potential issues will be predicted, anticipating countermeasures and reducing the risk of problems. As using external resources implies to assume some risks, the systems provided in OPTIMIS include tools which evaluate and reduces this risk. Thanks to the TREC tools, it is possible to guarantee quality levels, as they are used for determining and evaluating the criteria used to select external Clouds. It is expected that resources selected will fulfill minimum quality requirements and that services will be executed in trustworthy platforms, avoiding malicious behaviors in selected infrastructure providers. Finally, this set of tools allows obtaining as many benefits as possible from the available resources. Without OPTIMIS, peaks in resources need to be taken into account, having some resources unused but ready to be exploited. Using the OPTIMIS optimization tools, it is possible to offer resources to more service providers by adjusting the amount of available resources, according to the calculated profit. Then, when peaks are expected or problems appear, it will be possible for OPTIMIS to reassign services to external Clouds without decreasing the quality of service the end user will experience. OPTIMIS Consortium Page 16 of 29

18 6 Use case Scenario 6.1 More detailed description The use test application that will run within a Cloud Bursting scenario will showcase the use of a common e-commerce type transaction from selection of goods through to payment. Figure 2. E-commerce Transaction The scenario is focused on the activities to be performed for buying some products to providers, when our company is running out of them, as part of one of the company s business processes. It implies simple activities such as performing a search about the offers in the market (look for providers and their offers), selecting the products to be bought and the provider which sell them, and proceed with payments. This will use the following steps: Step 1: The retailer defines products required including quantity, price, availability etc. Step 2: The retailer requests quotes for products A, B and C from a number of providers(create Customer Quote) Step 3: The retailer selects the best set of vendors based on requirements defined in step 1. Step 4: Retailer creates a sales order with each vendor selected in step 3. (Create Sales Order_V2) Step 5: Product vendor automatically creates an invoice and sends it to the retailer. (Notify of Customer Invoice) Step 6: Retailer process receivables and payments Step 7: Value added process such as the example below. An example would be a shop which sells a lot of products such as Amazon's e-commerce website, in certain dates (such as Christmas) may sell so many products that they are running out of some of them, so they need to order more. If there are many products to order and many users accessing to their systems, some concrete tasks might be migrated as a mean to guarantee that the process is finished. Another example would be a large manufacturer which is running a lot of processes in parallel and needs to execute all of them in proper conditions. In this case, it would be possible even to perform bursting for migrating the whole process or other services and applications for releasing local resources. OPTIMIS Consortium Page 17 of 29

19 During the Cloud Bursting, some steps will be carried out internally in the platform: Step 1: Gather monitoring information about local resources Step 2: Determine resources needed, identifying whether more resources are needed or not Step 3: Re-arrange local resources optimizing their usage Step 4: If more resources are needed, gather information about external Clouds which can be used Step 5: Evaluate other platforms using inputs from the TREC tools and generate a ranking Step 6: Negotiate with the selected external Cloud the terms for using its resources Step 7: Use the resources from the external Cloud, according to the agreed terms These steps are expected to happen automatically, so the scenario design will indicate the components involved, but the main effort will be put on the workflow definition and the usage of the services involved, which should be deployed in the OPTIMIS platform. 6.2 Actors involved and their roles Within a trust dependable ecosystem of providers, there are service providers utilising cloud bursting between Infrastructure providers (possibly provided by a broker) to complement local capacity for peak loads to cater for end user needs. These roles are detailed as follows. Service Providers (SPs) offer economically efficient services with assessed and guaranteed environmental impact. The services are directly accessed by end-users or orchestrated by other SPs Infrastructure Providers (IPs) offer cloud infrastructure resources required for hosting services. Their goal is to maximize their profit from tenants (Service Providers) by making efficient use of the infrastructures, and possible by outsourcing partial workloads to partner data centres. Broker, is a trusted party who acts as an intermediary conduit to provide IPs for the use of federations and bursting scenarios. There are other definitions which should be taken into account, as they imply different levels of QoS and even scope of the resources to be provided from the IP to a SP. These definitions depend on the kind of cloud to be provided and can be categorized as follows: Dedicated Cloud: All processing and storage hardware is logically and physically separated from the one being used by other customers; Private Cloud: Some of the resources are always reserved in order to be used by the same customer, although there is no physical independence of other resources; Shared Cloud :The resources are shared by several customers, so they are assigned on demand. Depending on the clouds to be provided and which are coexisting in an IP, this may imply to perform additional steps (such as resources rearrangement) when performing bursting actions. OPTIMIS Consortium Page 18 of 29

20 6.3 Requirements Identification Identified functional requirements If the scenario will involve bursting out to external computational resources for searches that are part of the same overall transaction then data consistency must be ensured. The bursting scenario should have the option to be brokered or not as the internal provider may or may not have a fixed partner in mind and by not using a broker the assumption is a cheaper burstable service. Bursting should be done automatically with no need to quiz the end user on requirements and should be part of a transparent service. The decision of when to burst (enact elasticity rules) due to lack of resources should be launched based on a perceived lack of the following resources: CPU (Load) Disk (Space Left) RAM (Space Left) Latency (Speed of operations) Mechanisms are required for constant monitoring of service status and for triggering actions to increase and decrease capacity, i.e., enact elasticity rules, in order to meet the BLOs specified during service construction. They are also important to decide when to stop the cloud bursting, going on with the execution in local resources. TREC considerations should be given where trust cost and energy efficiency are independently verified by querying a trusted database containing historical information pertaining to these factors. The resources optimization system should take them into account for selecting which platform use for externalizing resources. Dynamic decisions also must be made based on real-time data to cater for situations where for example, there is a loss of power to an IP. This means that the optimizer will go on analyzing the situation while external resources are under use Gaps analysis There are no known gaps between the requirements identified in this deliverable and the scope of the OPTIMIS toolkit components in order to fulfill the deliverable. OPTIMIS Consortium Page 19 of 29

21 7 High level design of the use case In the mentioned context, it is important to put especial attention in service construction, deployment and execution for optimising the service lifecycle. According to the OPTIMIS architecture, TREC components from the base toolkit are basic tools for evaluating IPs and perform bursting operations. Self-* management components will be used to conform to agreed SLAs and, finally, Provisioning Components will be used for accessing to extra resources and license management features. According to the scenario defined, the following workflow can be used as the high level design of the operations to be performed: Figure 3. High Level Design Workflow Activities to be performed are those explained in section 5.1, which represent the process for buying products. These activities, sometimes, are represented by software assets which will be invoked or executed during the workflow execution. Moreover the execution of the workflow itself may be managed by a large ERP of the company which takes care of all the processes defined. The idea for the deployment is to have a platform with the full OPTIMIS Toolkit installed and some VMs up and running where all the services necessary for executing the workflow are deployed, including payment services (which belong to a different provider). OPTIMIS Consortium Page 20 of 29

22 Figure 4. Cloud Bursting Interaction In the case it is necessary to perform could bursting, OPTIMIS platform would negotiate with a Flexiant Cloud which will be able to provide available resources. This cloud will manage its own VMs and assign required resources with the aim at fulfilling agreed SLAs. 7.1 OPTIMIS Components Involved In order to implement the described scenario, it would be necessary to use many of the components included in the OPTIMIS high level architecture (from D ). OPTIMIS Consortium Page 21 of 29

23 Figure 5.OPTIMIS Architecture The previous figure represents the OPTIMIS architecture, where some components have been highlighted (with an *) because of their importance in this concrete scenario: TREC: This is the group of tools which gather information about what is going on in the system in terms of Trust, Risk, Ecology and Cost. They will provide crucial inputs when deciding which resources to use; Elasticity Engine: It is the responsible of managing VMs to be set up and down, so it will control the provision of more resources; Cloud Optimizer: This component is very important, as it integrates functionalities of several components (TREC and Elasticity Engine, between others) and it is able to take decisions about when and how to provide resources for fulfilling SLAs; License Manager: As bursting may require migrating whole applications, it is necessary to be sure that licenses will not represent a problem. It will take care of licenses management in order to avoid any legal problem. 7.2 Bursting Possibilities Taking into account the workflow and the kind of services and systems which may be involved, there are two possibilities for cloud bursting: Perform bursting of the whole workflow management, if there are so many processes under execution that more resources are needed; Perform bursting of concrete activities, by migrating only concrete services which represent one or more of the activities in the workflow. In the first case, it would be necessary to use a workflow management system which allows migrating all the systems involved or most of them if there are some systems which are OPTIMIS Consortium Page 22 of 29

24 independent enough. Unless this is not exposed as a service or an application deployed in a VM, in principle, OPTIMIS is not designed to migrate these kinds of applications. In the second case, concrete services (already deployed in a VM) will be migrated to other resources where they can be executed fulfilling any required QoS level. In OPTIMIS, this is done by setting up new VMs in another IP, so the new VMs will host a service and/or its database in order to operate in the expected conditions (similar as in a Federated Cloud environment). According to a preliminary analysis performed, we consider the following activities as candidates for performing bursting of the services which represent them: A3: This activity is expected to gather a lot of information coming from different sources. Moreover, if the process is automatic, it may require a lot of CPU and memory resources for executing those algorithms which will decide between the quotes received; A5: In the case many invoices have to be received, it would be possible to expose a service which receives this information in other VMs, transferring later the information to a centralized database, if required. Although not specified in the workflow, some of the activities require the usage of third party services, such as in A6. In this case, as payment services are offered by another company and they may get saturated because of multiple transactions, the service in charge of the payments could take advantage of the cloud for guaranteeing minimum QoS levels. OPTIMIS Consortium Page 23 of 29

25 8 Conclusion This document details the status of the use case detailed in WP6.3 and represents the work carried out towards the use case requirements, functional description and develops the template for the prototype architecture that will integrate with the OPTIMIS model. A business case is detailed along with an accompanying market analysis and exploitation plan. Future work in WP6.3 will involve prototype implementation, enabling interoperability between providers. This will implement an application that dynamically scales out over external resources. Following the implementation, there will be a period of testing to evaluate the interaction between platforms and the encapsulated cloud service. This will ensure interoperability between the internal and external cloud providers, ensure provision of sufficient capacity for cloud service needs and be demonstrable of a criteria based selection process. OPTIMIS Consortium Page 24 of 29

26 9 References [1] The 451 Group: 451 Market Monitor, Cloud Computing, Segment Focus: Cloud On- Ramps, Report IV, March 2010 [2] Requirements Analysis, Deliverable D of OPTIMIS project, ULEEDS and other partners, September [3] Architecture Design document, OPTIMIS project Deliverable D , UMEA and other partners, September [4] Verification and Validation Cases, Deliverable D of the OPTIMIS project, ULEEDS and other partners, October [5] the_pea.php [6] [7] [8] r_2011_dplummer.pdf [9] [10] elementid=null&pagetype=synopsis [11] [12] [13] [14] OPTIMIS Consortium Page 25 of 29

27 Annex A. License conditions. This is a public deliverable that is provided to the community under the license Attribution-NoDerivs 2.5 defined by creative commons This license allows you to to copy, distribute, display, and perform the work to make commercial use of the work Under the following conditions: Attribution. You must attribute the work by indicating that this work originated from the IST- OPTIMIS project and has been partially funded by the European Commission under contract number IST No Derivative Works. You may not alter, transform, or build upon this work without explicit permission of the consortium For any reuse or distribution, you must make clear to others the license terms of this work. Any of these conditions can be waived if you get permission from the copyright holder. This is a human-readable summary of the Legal Code below: License THE WORK (AS DEFINED BELOW) IS PROVIDED UNDER THE TERMS OF THIS CREATIVE COMMONS PUBLIC LICENSE ("CCPL" OR "LICENSE"). THE WORK IS PROTECTED BY COPYRIGHT AND/OR OTHER APPLICABLE LAW. ANY USE OF THE WORK OTHER THAN AS AUTHORIZED UNDER THIS LICENSE OR COPYRIGHT LAW IS PROHIBITED. BY EXERCISING ANY RIGHTS TO THE WORK PROVIDED HERE, YOU ACCEPT AND AGREE TO BE BOUND BY THE TERMS OF THIS LICENSE. THE LICENSOR GRANTS YOU THE RIGHTS CONTAINED HERE IN CONSIDERATION OF YOUR ACCEPTANCE OF SUCH TERMS AND CONDITIONS. 1. Definitions "Collective Work" means a work, such as a periodical issue, anthology or encyclopedia, in which the Work in its entirety in unmodified form, along with a number of other contributions, constituting separate and independent works in themselves, are assembled into a collective whole. A work that constitutes a Collective Work will not be considered a Derivative Work (as defined below) for the purposes of this License. "Derivative Work" means a work based upon the Work or upon the Work and other pre-existing works, such as a translation, musical arrangement, dramatization, fictionalization, motion picture version, sound recording, art reproduction, abridgment, condensation, or any other form in which the Work may be recast, transformed, or adapted, except that a work that constitutes a Collective Work will not be considered a Derivative Work for the purpose of this License. For the avoidance of doubt, where the Work is a musical composition or sound recording, the synchronization of the Work in timed-relation with a moving image ("synching") will be considered a Derivative Work for the purpose of this License. "Licensor" means all partners of the OPTIMIS consortium that have participated in the production of this text "Original Author" means the individual or entity who created the Work. "Work" means the copyrightable work of authorship offered under the terms of this License. "You" means an individual or entity exercising rights under this License who has not previously violated the terms of this License with respect to the Work, or who has received express permission from the Licensor to exercise rights under this License despite a previous violation. 2. Fair Use Rights. Nothing in this license is intended to reduce, limit, or restrict any rights arising from fair use, first sale or other limitations on the exclusive rights of the copyright owner under copyright law or other applicable laws. 3. License Grant. Subject to the terms and conditions of this License, Licensor hereby grants You a worldwide, royalty-free, nonexclusive, perpetual (for the duration of the applicable copyright) license to exercise the rights in the Work as stated below: to reproduce the Work, to incorporate the Work into one or more Collective Works, and to reproduce the Work as incorporated in the Collective Works; to distribute copies or phonorecords of, display publicly, perform publicly, and perform publicly by means of a digital audio transmission the Work including as incorporated in Collective Works. For the avoidance of doubt, where the work is a musical composition: Performance Royalties Under Blanket Licenses. Licensor waives the exclusive right to collect, whether individually or via a performance rights society (e.g. ASCAP, BMI, SESAC), royalties for the public performance or public digital performance (e.g. webcast) of the Work. OPTIMIS Consortium Page 26 of 29

28 Mechanical Rights and Statutory Royalties. Licensor waives the exclusive right to collect, whether individually or via a music rights society or designated agent (e.g. Harry Fox Agency), royalties for any phonorecord You create from the Work ("cover version") and distribute, subject to the compulsory license created by 17 USC Section 115 of the US Copyright Act (or the equivalent in other jurisdictions). Webcasting Rights and Statutory Royalties. For the avoidance of doubt, where the Work is a sound recording, Licensor waives the exclusive right to collect, whether individually or via a performance-rights society (e.g. SoundExchange), royalties for the public digital performance (e.g. webcast) of the Work, subject to the compulsory license created by 17 USC Section 114 of the US Copyright Act (or the equivalent in other jurisdictions). The above rights may be exercised in all media and formats whether now known or hereafter devised. The above rights include the right to make such modifications as are technically necessary to exercise the rights in other media and formats, but otherwise you have no rights to make Derivative Works. All rights not expressly granted by Licensor are hereby reserved. 4. Restrictions. The license granted in Section 3 above is expressly made subject to and limited by the following restrictions: You may distribute, publicly display, publicly perform, or publicly digitally perform the Work only under the terms of this License, and You must include a copy of, or the Uniform Resource Identifier for, this License with every copy or phonorecord of the Work You distribute, publicly display, publicly perform, or publicly digitally perform. You may not offer or impose any terms on the Work that alter or restrict the terms of this License or the recipients' exercise of the rights granted hereunder. You may not sublicense the Work. You must keep intact all notices that refer to this License and to the disclaimer of warranties. You may not distribute, publicly display, publicly perform, or publicly digitally perform the Work with any technological measures that control access or use of the Work in a manner inconsistent with the terms of this License Agreement. The above applies to the Work as incorporated in a Collective Work, but this does not require the Collective Work apart from the Work itself to be made subject to the terms of this License. If You create a Collective Work, upon notice from any Licensor You must, to the extent practicable, remove from the Collective Work any credit as required by clause 4(b), as requested. If you distribute, publicly display, publicly perform, or publicly digitally perform the Work or Collective Works, You must keep intact all copyright notices for the Work and provide, reasonable to the medium or means You are utilizing: (i) the name of the Original Author (or pseudonym, if applicable) if supplied, and/or (ii) if the Original Author and/or Licensor designate another party or parties (e.g. a sponsor institute, publishing entity, journal) for attribution in Licensor's copyright notice, terms of service or by other reasonable means, the name of such party or parties; the title of the Work if supplied; and to the extent reasonably practicable, the Uniform Resource Identifier, if any, that Licensor specifies to be associated with the Work, unless such URI does not refer to the copyright notice or licensing information for the Work. Such credit may be implemented in any reasonable manner; provided, however, that in the case of a Collective Work, at a minimum such credit will appear where any other comparable authorship credit appears and in a manner at least as prominent as such other comparable authorship credit. 5. Representations, Warranties and Disclaimer. UNLESS OTHERWISE MUTUALLY AGREED TO BY THE PARTIES IN WRITING, LICENSOR OFFERS THE WORK AS-IS AND MAKES NO REPRESENTATIONS OR WARRANTIES OF ANY KIND CONCERNING THE MATERIALS, EXPRESS, IMPLIED, STATUTORY OR OTHERWISE, INCLUDING, WITHOUT LIMITATION, WARRANTIES OF TITLE, MERCHANTIBILITY, FITNESS FOR A PARTICULAR PURPOSE, NONINFRINGEMENT, OR THE ABSENCE OF LATENT OR OTHER DEFECTS, ACCURACY, OR THE PRESENCE OF ABSENCE OF ERRORS, WHETHER OR NOT DISCOVERABLE. SOME JURISDICTIONS DO NOT ALLOW THE EXCLUSION OF IMPLIED WARRANTIES, SO SUCH EXCLUSION MAY NOT APPLY TO YOU. 6. Limitation on Liability. EXCEPT TO THE EXTENT REQUIRED BY APPLICABLE LAW, IN NO EVENT WILL LICENSOR BE LIABLE TO YOU ON ANY LEGAL THEORY FOR ANY SPECIAL, INCIDENTAL, CONSEQUENTIAL, PUNITIVE OR EXEMPLARY DAMAGES ARISING OUT OF THIS LICENSE OR THE USE OF THE WORK, EVEN IF LICENSOR HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. 7. Termination This License and the rights granted hereunder will terminate automatically upon any breach by You of the terms of this License. Individuals or entities who have received Collective Works from You under this License, however, will not have their licenses terminated provided such individuals or entities remain in full compliance with those licenses. Sections 1, 2, 5, 6, 7, and 8 will survive any termination of this License. Subject to the above terms and conditions, the license granted here is perpetual (for the duration of the applicable copyright in the Work). Notwithstanding the above, Licensor reserves the right to release the Work under different license terms or to stop distributing the Work at any time; provided, however that any such election will not serve to withdraw this License (or any other license that has been, or is required to be, granted under the terms of this License), and this License will continue in full force and effect unless terminated as stated above. 8. Miscellaneous Each time You distribute or publicly digitally perform the Work, the Licensor offers to the recipient a license to the Work on the same terms and conditions as the license granted to You under this License. If any provision of this License is invalid or unenforceable under applicable law, it shall not affect the validity or enforceability of the remainder of the terms of this License, and without further action by the parties to this agreement, such provision shall be reformed to the minimum extent necessary to make such provision valid and enforceable. No term or provision of this License shall be deemed waived and no breach consented to unless such waiver or consent shall be in writing and signed by the party to be charged with such waiver or consent. This License constitutes the entire agreement between the parties with respect to the Work licensed here. There are no understandings, agreements or representations with respect to the Work not specified here. Licensor shall not be bound by any OPTIMIS Consortium Page 27 of 29