European Cooperation in the field of Scientific and Technical Research - COST - Brussels, 24 May 2013 COST 024/13 MEMORANDUM OF UNDERSTANDING Subject : Memorandum of Understanding for the implementation of a European Concerted Research Action designated as COST Action IC1304: Autonomous Control for a Reliable Internet of Services (ACROSS) Delegations will find attached the Memorandum of Understanding for COST Action IC1304 as approved by the COST Committee of Senior Officials (CSO) at its 187th meeting on 15-16 May 2013. COST 024/13 1
MEMORANDUM OF UNDERSTANDING For the implementation of a European Concerted Research Action designated as COST Action IC1304 AUTONOMOUS CONTROL FOR A RELIABLE INTERNET OF SERVICES (ACROSS) The Parties to this Memorandum of Understanding, declaring their common intention to participate in the concerted Action referred to above and described in the technical Annex to the Memorandum, have reached the following understanding: 1. The Action will be carried out in accordance with the provisions of document COST 4154/11 Rules and Procedures for Implementing COST Actions, or in any new document amending or replacing it, the contents of which the Parties are fully aware of. 2. The main objective of the Action is to create a European network of experts, from both academia and industry, aiming at the development of autonomous control methods for a reliable and quality-aware Internet of Services. 3. The economic dimension of the activities carried out under the Action has been estimated, on the basis of information available during the planning of the Action, at EUR 60 million in 2013 prices. 4. The Memorandum of Understanding will take effect on being accepted by at least five Parties. 5. The Memorandum of Understanding will remain in force for a period of 4 years, calculated from the date of the first meeting of the Management Committee, unless the duration of the Action is modified according to the provisions of Chapter IV of the document referred to in Point 1 above. COST 024/13 2
A. ABSTRACT AND KEYWORDS Currently, we are witnessing a paradigm shift from the traditional information-oriented Internet into an Internet of Services (IoS). This transition opens up virtually unbounded possibilities for creating and deploying new services. Eventually, the ICT landscape will migrate into a global system where new services are essentially large-scale service chains, combining and integrating the functionality of (possibly huge) numbers of other services offered by third parties, including cloud services. At the same time, as our modern society is becoming more and more dependent on ICT, these developments raise the need for effective means to ensure quality and reliability of the services running in such a complex environment. Motivated by this, the aim of this Action is to create a European network of experts, from both academia and industry, aiming at the development of autonomous control methods and algorithms for a reliable and quality-aware IoS. Keywords: Service oriented Internet, cloud services, autonomous control, reliability, pricing B. BACKGROUND B.1 General background Over the past few decades, the use of the Internet has grown explosively and has fundamentally changed the global society. The emergence of concepts like service-oriented architecture (SOA), Software as a Service (SaaS), Platform as a Service (PaaS), Infrastructure as a Service (IaaS) and Cloud Computing has catalyzed the migration from the information-oriented Internet into an Internet of Services (IoS). This has opened up virtually unbounded possibilities for the creation of new and innovative services that facilitate business processes and improve the quality of life. As a consequence, modern societies and economies have become heavily dependent on Information and Communication Technologies (ICT). Failures and outages of ICT-based services (e.g., financial transactions, booking services, Web-shopping, governmental services) may cause economic damage and affect people s trust in ICT. Therefore, providing reliable and robust ICT services (resistant against system failures, cyber-attacks, high-load and overload situations, flash crowds, etc.) is crucial for our economy at large. Moreover, in the competitive markets of ICT service offerings, it is of great importance for service providers to be able to realize short time-to-market and to deliver services at sharp price-quality ratios. These observations make the societal and economic importance of reliable Internet services evident. A fundamental characteristic of the IoS is that services combine and integrate functionalities of COST 024/13 3
other services. This has led to complex service chains with hundreds - and often thousands - of services offered by different third parties, each with their own business incentives. In current practice, service quality for composite services is usually controlled on an ad-hoc basis, while the consequences of failures in service chains are not well understood. The problem is that, although such an approach might work for small service chains, this will become useless for future complex global-scale service chains. B.2 Current state of knowledge Over the past few years, significant research has been devoted to controlling Quality of Service (QoS) and Quality of Experience (QoE) for IoS. To this end, much progress has been made in the functional layer of QoS-architectures and frameworks, and system development for the IoS. However, relatively little attention has been paid to the development, evaluation and optimization of algorithms for autonomous control (see reference [1] in Part II-D below for a recent and extensive overview of the state-of-the-art on QoS-architectures, and the challenges in the years to come). In this context, the main goal of this COST Action is to bring the state-of-the-art on autonomous control to the next level by developing quantitative methods and algorithms for autonomous control for a reliable IoS. Traditionally, in the area of quantitative control methods the main focus has been on traditional controls for QoS provisioning at the network and lower layers. In this context, it is important to note that control methods for the IoS operate at the higher protocol layers and typically involve a multitude of administrative domains. As such, these control methods and their effectiveness are fundamentally different from the traditional network-level control methods, posing fundamentally new challenges. For example, for composite service chains the main challenges are self-star methods for dynamic re-composition, to prevent or mitigate the propagation of failures through the service chains, and methods for overload control at the services layer. Another challenging factor in quality provisioning in the IoS is its highly dynamic nature, imposing a high degree of uncertainty in many respects (e.g., in terms of number and diversity of the service offerings, the system load of services suddenly jumping to temporary overload, demand for cloud resources, etc.). This raises the urgent need for online control methods with self-learning capabilities that quickly adapt to or even anticipate to changing circumstances. This COST Action will bring the state-of-the-art in the area of autonomous quality-based control in the IoS to the next level by ground-breaking research on developing efficient methods and algorithms that enable service providers to fully exploit the enormous possibilities for realizing a COST 024/13 4
reliable and quality-aware service-oriented Internet. Research on quantitative methods for control in the IoS faces tremendous scientific challenges for which at best little is known today. This COST Actions envisions the realistic situation where a huge number of composite service providers operate in an environment that is dynamic in different respects (e.g., availability of services, performance of services, user context, pricing schemes), where services may be composed of a large number of sub-services (including cloud services) offered by third parties. This requires research on the following important areas. 1. Autonomous control mechanisms: To deliver reliable services in the IoS, (composite) service providers will need to implement control mechanisms, ranging from simplistic to highly advanced. Typical questions are: How can we realize the efficient use of control methods by properly setting parameter values and decision thresholds? How can we effectively use these mechanisms depending on the specific context of a user (e.g., in terms of user location, the user role, operational settings or experienced quality)? How do control methods implemented by multiple providers interact? How does the interaction between multiple control methods affect their effectiveness? What about stability? How to resolve conflicts? Ideally, control mechanisms would be fully distributed and based on (experienced) quality. However, some level of centralized coordination among different autonomous control mechanisms may be needed. In this context, a major challenge is to achieve a proper trade-off between fully distributed control (having higher flexibility and robustness/resilience) and more centralized control (leading to better performance under normal conditions). This will lead to hybrid approaches, aiming to combine the best of two worlds. 2. QoS-monitoring and service prediction: A crucial element for autonomous control in the IoS is monitoring and service prediction. For autonomous real-time quality control in large, dynamic, complex many-domain environments like the IoS, there is a great need for scalable, non-intrusive monitoring and measurement of service demands, service performance and resource usage. Additional constraints regarding e.g. privacy and integrity will further complicate the challenges for monitoring and measurement. In addition, proactive service adaptation capabilities are rapidly becoming increasingly important for serviceoriented systems like IoS. In this context, there is a need for online quality prediction methods in combination with self-adaptation capabilities (e.g., service re-composition). 3. Pricing and competition in many-domain systems: Each service provider in the IoS will implement its own pricing mechanism, which may range from simple static to advanced dynamic pricing policies. The involvement of third-party and cloud COST 024/13 5
services in making up a composite service in these dynamic and competitive environments (all of the involved parties striving for maximization of their own profit) raises challenging questions that are new, and moreover, one can learn from the past. For example, in the traditional Internet, flat-fee charging schemes tend to be replaced by volume-based charging schemes. In this context, typical questions are: What are the implications of implementing different pricing mechanisms in a manydomain setting? How do quality levels and pricing mechanisms relate? How can one develop smart pricing mechanisms that provide proper incentives for the involved parties (regarding brokering, SLA (Service-Level-Agreement) negotiation strategies, federation etc.) that lead to a stable ecosystem? What governing rules are needed to achieve this? B.3 Reasons for the Action There are several important reasons for this Action: Scientific reasons for the Action: As outlined above, over the past few years significant progress has been made in the design and development of QoS-architectures and mechanisms for the IoS, but only little is known about how to effectively exploit these mechanisms (cf. [1]). In this context, this COST Action is aimed at bringing the state-of-the-art on autonomous control for the IoS to the next level by developing methods and algorithms to fully exploit the possibilities for controlling QoS and QoE for the IoS. This Action is aimed at filling the gap between the potential possibilities for autonomous QoS- and QoE-based control mechanisms in the IoS and the actual exploitation of these mechanisms (as sketched above in section B.2). Building up a European Community of Practice: Research on reliability and quality aspects of the IoS is dominated by the US, catalyzed by highly successful expert communities such as Computer Measurement Group (CMG), Association for Computing Machinery (ACM) and Institute of Electrical and Electronics Engineers (IEEE) (with influential conferences like ICSOC, ICWS, WWW, CMG, Sigmetrics). In comparison, Europe - although having a long-standing tradition in research on quality aspects of communication networks - is lagging behind. Moreover, European research in this field is scattered, and research groups typically focus on specific sub-topics, so that a huge potential for collaboration remains largely under-utilized. This Action is aimed at building up a European Community of Practice on autonomous QoS control for the IoS by bringing together and cross-fertilizing the knowledge and experience of leading industrial and academic partners in the field. COST 024/13 6
B.4 Complementarity with other research programmes Over the past few years, significant research efforts have been devoted to the design of QoSarchitectures and system development for the IoS and the cloud. One can refer to recently finished EU-projects like S-CUBE [1], SOA4All [2], SLA@SOI [3], SeQual [4], VISION Cloud [7], OPTIMIS [8], RESERVOIR [9], and to current EU-projects like Broker@Cloud [5], MODAClouds [6] and CELAR [40]. However, these projects have a different focus: they are primarily oriented towards the functional layers of quality-aware architectures and mechanisms In contrast, the goal of the present Action is to make an important step forward by developing quantitative (autonomous) control methods to effectively exploit the developed systems, and to design - and evaluate the potential of - alternative architectures and mechanisms for realizing reliable and robust composite service chains. In this way, the Action is largely complementary to other research programs, but will certainly make use of the valuable results obtained in these projects, bringing the state-of-the-art to the next level by developing methods to fully exploit the potential of quality-aware architectures and mechanisms. C. OBJECTIVES AND BEFITS C.1 Aim The overall goal of this Action is to create a European network of experts, from both academia and industry, aiming at the development of autonomous control methods for a reliable and quality-aware Internet of Services. C.2 Objectives The overall goal of this Action, stated above, is translated into the following more detailed objectives: 1. Establishment of a platform to fuel and coordinate research collaborations throughout Europe in the field of reliability, robustness and quality-aware management and control of Internet-based service infrastructures to overcome the current knowledge segmentation. COST 024/13 7
2. Development of algorithms for autonomous decision and control regarding service composition, resource allocation and user demand in the IoS, in order to ensure service quality and reliability. 3. Development of scalable methods and tools for system monitoring & analysis to support autonomous decision and control. 4. Development of rules and smart pricing schemes in multi-domain, highly competitive service environments (possibly based on game theory and self-organization principles) providing proper incentives to establish an economically viable ecosystem. The methods and algorithms to be developed in the Action will be evaluated and validated through extensive system simulations and (where possible) experiments with real equipment. The resulting implementations may be further developed towards proof of concepts and used for demonstration purposes. After four years, (part of) the activities of the ACROSS Action will be consolidated through establishment of a European Community of Practice, coupled with a yearly workshop or conference where scientists and practitioners meet each other and exchange ideas and experiences. C.3 How networking within the Action will yield the objectives? This Action brings together and will catalyze cooperation among researchers and research groups on quantitative modeling and control for the IoS, which are currently scattered over Europe leaving a huge potential largely under-utilized. In addition, the Action also aims at realizing cross fertilization by involving researchers working on related complementary topics, like IoS architectures, dynamic control, statistics and data analysis. In order to support the collaboration in the Action and actually achieve the establishment of a true European research platform in the field, the following networking activities will be set up: Set up and maintain an ACROSS website and provide associated collaboration tools. Organization of regular coordination meetings (at least two two-day meetings per year) providing ample room for technical presentations and discussions. COST 024/13 8
Organization of yearly open workshops with invited speakers from related other scientific fields. Organization of Summer schools for young researchers. Support of visits (e.g. Short Term Scientific Missions STMS) for senior and young researchers participating in the Action. C.4 Potential impact of the Action The expected scientific impacts can be summarized as follows (cf. also Sections B.2 and B.3): 1. Fundamental scientific advances in performance optimization and autonomous control regarding service creation (composition) and delivery in the Internet of Services, including related business aspects like pricing strategies, resource usage, SLAs, etc. 2. New research directions inspired by the work in the present Action (e.g., regarding the convergence with the Internet of Things and the Internet of Contents), and reuse of the Action s results in other application areas (e.g., smartgrids, logistics, telecommunications). 3. Closing the gap with the scientific advances in the field in the US. Bundling of research efforts and cross-fertilisation of top-quality expertise groups will bring Europe to the forefront of scientific research in the area. Expected benefits from a societal and economical point of view are: 1. Increase in the level of trust by the society in Internet-based services, leading to higher usage of these services and improved quality of life. 2. Humus layer and eco-system with pricing schemes providing the right for the development of new reliable services, boosting economic innovations. COST 024/13 9
3. Affordable access to ICT services in many societal areas (e.g., healthcare, entertainment). C.5 Target groups/end users The likely stakeholders and end-users that will exploit the expected results of this COST Action are: 1. System- and service integrators: Their R&D labs will bring the Action s results to the next step by developing prototype QoS-control, monitoring and pricing mechanisms (taking into account practical constraints), prior to commercial roll-out. 2. Providers of (composite) services: The main benefit for (composite) service providers is increased flexibility in the creation and delivery of new services at sharp price/quality ratios, thereby enhancing their competitive edge. 3. Universities: For academia the Action results may be the starting point or trigger for further and deeper elaboration of the addressed problems and proposed solutions. Representatives from all of these three stakeholder categories have been involved in the preparation of the Action. D. SCITIFIC PROGRAMME D.1 Scientific focus The scientific work program in this Action focuses on the development, validation and evaluation of autonomous control methods and algorithms for a reliable, quality-aware and economically viable IoS. Recently, a variety of quality-control frameworks for de IoS have been proposed, However, despite the enormous potential, today only little is understood about how to effectively exploit the possibilities of these frameworks (e.g., in terms of proper parameter setting, decision-threshold values). Motivated by this, this Action aims at taking this hurdle by developing, validating and evaluating quantitative models and scalable and robust solution algorithms. COST 024/13 10
D.2 Scientific work plan methods and means The main sub-areas of interest (outlined in Section B.2) will be covered by three corresponding Working Groups (WGs): WG1: Methods for autonomous management and real-time control WG2: Methods and tools for monitoring and service prediction WG3: Smart pricing and competition in many-domain systems The workplan for these WGs are outlined below. WG1: Autonomous management and real-time control WG1 will develop algorithms for static, semi-dynamic and fully dynamic service compositions, which are subject to end-to-end user-perceived QoS constraints, and where third-party and cloud services may be subject to negotiated service-level agreements. This will include methods for determining cost-optimal service compositions, and optimal decision tables for dynamic service (re)composition aiming to achieve the optimal tradeoff between static and dynamic service composition, both having their pros and cons in terms of simplicity and effectiveness. Starting with small-scale examples with a single composite service provider and a single priority, WG1 will gradually increase complexity, and consider complex settings with multiple composite service providers and priority mechanisms, where the intricate interaction between control techniques will be subject to study. Methods to be used in this WG: dynamic programming, approximate dynamic programming, statistical learning, robust optimization, heuristics and simulations. WG2: Methods and tools for monitoring and service prediction For autonomous real-time QoS control in large, dynamic, complex many-domain environments like the IoS, there is a great need for scalable, non-intrusive monitoring and measurement of service demands, service performance and resource usage. Additional constraints regarding e.g. privacy and integrity will further complicate the challenges for monitoring and measurement. In addition, proactive service adaptation capabilities are rapidly becoming increasingly important for serviceoriented systems like IoS. In this context, there is a need for online quality prediction methods in combination with self-adaptation capabilities (e.g., dynamic service re-composition). Motivated by this, several approaches have been studied, including data-mining and run-time verification techniques, online testing and statistical analysis approaches to detect anomalies and QoS COST 024/13 11
degradation, and simulation-based approaches to simulate future behavior of service-oriented systems. In this context, the main challenges include handling the heterogeneity (e.g., in service types and in time scales) and assessing the relevance of predictions. Methods to be used in this WG: statistical learning, forecasting models, time-series analysis, datamining techniques, verification models and anomaly detection. WG3: Smart pricing and competition in many-domain systems In the context of multiple composite service providers and multiple single-service providers, where each provider implements some pricing strategy, which can range from simplistic flat-rate pricing to advanced dynamic pricing schemes based on real-time information about sales and stock levels. In this context, WG3 will study the implications of these strategies in terms of the existence and characterization of game-theoretical equilibriums, using for example simulations or techniques form algorithmic game theory. Methods to be used in this WG: algorithmic game-theoretical models, revenue management techniques, unconstraining techniques, heuristics and simulations. Means: For each of the three WGs, the methods and algorithms to be developed in the Action will be evaluated and validated through extensive system simulations and, where possible, experiments with real equipment. The resulting implementations may be further developed towards Proof-of- Concepts and used for demonstration purposes. E. ORGANISATION E.1 Coordination and organisation According to the COST Rules and Procedures, the Management Committee (MC) will be in charge of the overall supervision of the Action. The MC will have one Chair, one Vice-Chair and a Secretariat. There will be at least two MC meetings per year. Each MC meeting will be open to all the involved experts as non-voting members. The MC meetings (usually two days) will be mainly dedicated to technical presentations and discussions. The program of these plenary technical presentations and discussions will be aligned according to the Working Group (WG) structure; if needed, the technical sessions may be (partly) held in parallel. WG-specific organizational matters will be discussed in short parallel meetings, if needed. The internal organization of the Action s technical activities is divided in three Working Groups COST 024/13 12
(WG), see also Section D.2. The Working Group Chairs and Co-Chairs will be proposed by the Action Chair and approved by the MC during the first meeting. In the first meeting, the MC will form a special task force to setup and maintain the ACROSS website and provide additional electronic collaboration tools (e-mail lists and shared working space). The website and collaboration tools will be made available within 3 months after the start of the Action. See Section H for more details about their role within the Action and externally. The Action Chair and the Working Group Chairs will be in charge of coordinating the delivery of the annual reports due at month 12, 24, 36 and 48. The first Annual Report will contain an extensive state-of-the-art review of ACROSS research, which will also serve as a basis for refinement of the planning of future activities. In the preparations for the Annual Report, the Action will carry out a self-evaluation and determine required follow-up actions in order to ensure achievement of the objectives or timely identify deviations and properly adapt the working program. The selfevaluation will be prepared by the Action Chair, Action Vice-Chair and WG Chairs, and be discussed and finalized in the MC. Besides the Action s website several other instruments will be used to stimulate and coordinate the exchange of knowledge and experience among the participants and with other researchers in the field, in particular: Once per year an MC meeting will contain an open workshop, possibly connected to a suitable conference or held jointly with another event. These workshops will be actively announced in the scientific/engineering community in the field. Short Term Scientific Missions (STSMs) will be strongly encouraged, in particular to young researchers; they will be coordinated by an STSM manager to be appointed by the MC. Individual STSM proposals should be approved by the MC. A Summer School will be organized every year starting from Year 2. Target audience are young researchers and practitioners working in the field. External key speakers, also from other disciplines and/or from industry, will be regularly invited to attend the technical meetings and to give presentations. Liaisons and interactions with related research projects and programmes (see E.3). COST 024/13 13
Industrial Forum In order to provide additional support for the exchange of knowledge and experience with the stakeholders an Industrial Forum will be established consisting of individuals from industry and end-user communities that are not directly involved in the Action. The Industrial Forum will be regularly informed about the activities and progress of the Action, and will be asked to give feedback in order to evaluate results and (based on this) to refine the objectives. In Year 1 the MC will actively contact 10-15 individuals from industry and end-user communities to establish the Industrial Forum, which will be extended during the four year period of the Action. E.2 Working Groups The three main sub-areas of interest will be covered by three corresponding Working Groups (WGs), see also Section D.2: WG1: Methods for autonomous management and real-time control WG2: Methods and tools for monitoring and service prediction WG3: Smart pricing and competition in many-domain systems Each WG will be dedicated to specific activities and goals. WG Chairs will be responsible for the coordination of particular WG activities, and will be assisted by a WG Vice-Chair. WG Chairs and Co-Chairs will be proposed by the Action Chair and approved by the MC. Any Action participant can take part in a WG. The program of the technical presentation and discussion sessions at the MC meetings will be aligned according to the WG structure; if needed (depending on the number of presentations) these sessions will (partly) be held in parallel. For WG-specific (organizational) matters short parallel meetings will be held during the MC meetings, if needed. WGs will have their own email list; the Action s website and shared working space will be (partly) organized according to the WG structure in order to facilitate collaboration at WG level. E.3 Liaison and interaction with other research programmes ACROSS will establish liaisons and interact with several related research programmes, platforms and projects. In particular, liaisons will be established with the European NESSI (Networked European Software and Services Initiative) platform, with IFIP WG 2.14 (Service Oriented Systems), and with the World Wide Web Consortium (W3C) and IETF being the main standardization bodies in the field. Establishment of liaisons with other forums (e.g., ITU-T FG COST 024/13 14
Cloud, Open Cloud Consortium) will be investigated. Further, interaction is foreseen with several recently started FP7 projects on (autonomous/dynamic) control in the context of the IoS, e.g. Broker@Cloud, CELAR and MODAClouds, which have a focus on cloud services. Broker@Cloud aims at development of methods for continuous quality assurance and optimization in future enterprise cloud service brokers, CELAR is on autonomous control of cloud resources, and MODAClouds aims at a model-driven approach for design and execution of applications on multiple Clouds. Other interesting (but almost finished) related research projects are FP7 OPTIMIS and FP7 VISION Cloud. Coordination will take place through the exchange of information about ongoing and planned work, invitation of people as speaker at the technical presentation sessions during MC meetings, as well as by seeking to join forces for the organization of workshops and summer schools. These activities will secure proper alignment and synergy with related projects and research programmes. Interactions and liaisons with research programs in complementary areas and disciplines such as intelligent systems, dynamic control, statistics, quality of experience (e.g., in COST Action IC1003 [41]) will also be part of coordination and liaison efforts. The List of Experts participating in the Action already includes representatives involved in several of the referenced initiatives. For example, one of the Experts serves as vice-chair of IFIP WG 2.14. These existing connections will facilitate the communication and coordination. It also indicates that the Action does have the critical mass to play a central role in the research community. E.4 Gender balance and involvement of early-stage researchers This COST Action will respect an appropriate gender balance in all its activities and the Management Committee will place this as standard item on all its MC agendas. The action will also be committed to considerably involve early-stage researchers. This item will also be placed as a standard item on all MC agendas. Gender balance and involvement of early-stage researchers will get particular attention in (future) nominations of MC members, in the nominations for WG (Co-)Chairs, and in encouraging participation in STSMs. For more details, see Part II, Section E. F. TIMETABLE The expected duration of the Action is four years. At least two MC meetings will take place every year (in Q1 and Q3). MC meetings will take place during two days and will typically contain a COST 024/13 15
general management assembly (half-day) and plenary technical sessions (presentations plus discussions) arranged according to the WG structure. If needed, the technical sessions may be (partly) held in parallel. WG-specific organizational matters will be discussed in short parallel meetings, if needed. WG chairs will be elected at the kick-off MC meeting. After 3 months from the kick-off MC meeting the first version of the Action s website will be launched At that time also a first detailed version of the Action s dissemination plan will be delivered, to be updated yearly as part of the Annual Report. And also the Industrial Forum (see Section E.1) will be established in the first 3 months of the Action. In Year 1, the Action will work on an extensive state-of-the-art review of ACROSS research. This overview will be presented in the first Annual Report of the Action at the end of Year 1, which will also serve as basis for refinement of the planning of future activities. The Action s progress will then be reported in two additional Annual Reports at the end of Year 2 and Year 3, and in the Final Report at the end of Year 4. Every year, starting from Year 2, a Summer School for PhD students and (young) practitioners will be organized. Another yearly event is the open workshop, which is meant to give external visibility to the Action. It may be co-located to related international conferences or other events. Typically, the open workshops will take place jointly with an MC meeting. Short Term Scientific Missions (STSM), to be approved by the MC, can take place anytime during the Action period. In Year 3 and Year 4 there will be special attention for continuation and consolidation of (part of) the Action s activities through establishment of a European Community of Practice, coupled with a yearly workshop or conference where scientists and practitioners meet each other and exchange ideas and experiences. G. ECONOMIC DIMSION The following COST countries have actively participated in the preparation of the Action or otherwise indicated their interest: AT, BE, CH, DE, ES, FI, FR, HU, IE, IT, NL, PL, RS, SE, UK. On the basis of national estimates, the economic dimension of the activities to be carried out under the Action has been estimated at 60 Million for the total duration of the Action. This estimate is valid under the assumption that all the countries mentioned above but no other countries will participate in the Action. Any departure from this will change the total cost accordingly. H. DISSEMINATION PLAN H.1 Who? COST 024/13 16
The primary target audiences for dissemination of the Action s results are: International research community working in the ACROSS field of (autonomous) service control. Researchers from related fields, for example system control, intelligent systems, data analysis & statistics, performance modeling and analysis. Industrial players, in particular system integrators, service providers and service brokers, telecom operators and equipment vendors. European projects and research platforms in the area of Internet of Services, cloud Computing and advanced software engineering, e.g. NESSI, IFIP WG 2.14, etc. (see also Section E.3). Standardization, in particular the World Wide Web Consortium (W3C) and Internet Engineering Task Force (IETF) (the latter one, in particular, regarding cloud services) and, possibly, ITU-T FG Cloud and the Open Cloud Consortium (Telecommunication Standardization Sector of the International Telecommunication Union). Government agencies and other (legal) institutes in charge of regulation and/or certification of (the quality of) Internet services. H.2 What? The following dissemination methods will be used: ACROSS website, discussion forums and (thematic) mailing lists. Workshops (once a year) open for participation by all researchers in the field. COST 024/13 17
Publication of articles in peer reviewed journals and conference proceedings, as well as (more popular) publications in magazines. Presentations and demonstrations at conferences, workshops and other events. The Action s Annual Reports. Open data bases containing system usage characteristics and quality measurements collected by the Action s participants, which can be used as reference by the community. Contributions to standardization bodies, like W3C and IETF as a possible spin-off of the research within the Action, e.g. regarding system interfaces needed for autonomous control and/or required measurements. H.3 How? Together with the ACROSS website, a first detailed version of the Action s dissemination plan will be delivered after 3 months from the kick-off MC meeting. This plan also contains the participants names responsible for specific tasks, and will be updated yearly as part of the Annual Report. Important elements of the detailed dissemination plan are briefly described below. Besides serving as a collaboration platform within ACROSS the Action's website will make publically available all the material delivered by the Action: scientific papers, technical reports, presentations, workshop materials, etc. It will include a wiki, host discussion forums and provide the possibility to post public comments about published articles. A special task force will be established at the kick-off MC meeting to build and maintain the website. The first version of the website will be launched within three months after the Action s kick-off. COST 024/13 18
Dissemination will also be achieved through the Summer School that will be organized yearly, and the Short Term Scientific Missions (STSM) strongly encouraged by the Action. Articles and papers to be published in international journals (e.g. IEEE Transactions on Services Computing, ACM TWeb) and conference proceedings (e.g. ICSOC, ICWS- SCC-Cloud, ECOWS) will explicitly acknowledge the Action s support. The Action will also pursue the organization of sessions within international conferences on specific topics within the ACROSS scope. For example, special sessions could be devoted to autonomous control methods, service monitoring, service prediction, etc. The technical sessions at the MC meetings and the annual open workshops organized by the Action will include keynote lectures by and discussions with external experts, particularly scientists from complementary disciplines and/or representatives from industry, thus serving as a dissemination channel. The interaction with the Industrial Forum, that will be established in the initial phase of the Action (see Section E.1), does also provide a way to disseminate the Action s results to industry and other end-users. The Annual Reports (public parts) produced by the Action will be published on the website. The technical part of the Final Report will be published as a book (e.g. LNCS volume of Springer). Contributions to standardization are seen as important possible spin-offs of the research planned within the Action, and will be encouraged by the Action. The Action will seek a liaison with selected standardization bodies, in particular W3C and IETF. COST 024/13 19