Ivan Breskovic

Ivan Brešković

MSc

Research assistant, PhD student

 

About Me

I am a PhD student and research assistant at the Distributed Systems Group, Institute of Information Systems, Vienna University of Technology. I did my bachelor study in Software Engineering at the University of Zagreb, Faculty of Electrical Engineering and Computing in Zagreb, Croatia and master’s study in Software Engineering and Information Systems at the same faculty. I am currently involved in the Holistic Energy Efficient Approach for the Management of Hybrid Clouds (HALEY) project, which received the Research Award from the Vienna University of Technology. My areas of interest include cloud computing, autonomic computing, SLA and QoS management.

Research Areas

My research interest includes the following:

  • Cloud computing
  • Service Level Agreement (SLA) management
  • Quality of Service (QoS) in clouds
  • Autonomic computing
  • Service-Oriented Architectures

Research Projects

HALEY - Holistic Energy Efficient Approach for the Management of Hybrid Clouds

FoSII - Foundations of Self-governing ICT Infrastructures (FoSII)

Teaching

I supervise bachelor’s theses, master’s theses, and the following practical courses:

Recent students:

  • Christoph Redl: "Autonomic Management of SLA Mappings in Cloud Markets" (master's thesis, April 2012)

Currently open topics:

  • Automatic Service Provider Selection in Cloud Computing Marketplaces
     (TISS)

    Cloud computing is a novel computing paradigm that offers resources (i.e., data, software, and hardware services) in a manner similar to traditional utilities such as water, electricity, gas, and telephony. However, the current cloud market is fragmented and static, preventing the successful implementation of ubiquitous computing on demand. In order to address the issue of fragmentation, commodity cloud computing markets have been suggested.

    With the growth and expansion of the cloud computing paradigm and participation of various service providers, specifications of quality parameters, metrics, and units start to diverse and even contradict. This ambiguity does not only result in various QoS interoperability problems, but also in the distraction of service consumers who find themselves unable to match their quality requirements with the providers’ offers. In order to counteract this problem, SLA mappings were introduced [1]. They are documents used to map the differences between differing SLAs. Unlike other matching techniques, which mostly include QoS ontologies, SLA mappings are immune to the dynamic nature of market’s user base and the actions users perform.

    Currently, SLA mappings are used not only as a matching technique, but also in a process of creating standardized products for electronic markets [2][3]. Standardized products reduce the number of resource types offered to the users and therefore decrease the cost incurred by searching for an appropriate service on the market. In this thesis, this work will be extended by defining and implementing methods for cloud provider service selection using several methodologies, such as cost-benefit analysis, performance evaluation, and comprehensive set of well-defined quality aspects.

    In particular, the thesis should define and combine (existing) quality aspects of SLAs used to describe service offers (in case of providers) and service requirements (in case of consumers). Furthermore, a utility/cost function should be defined for assessing the quality of a match between a consumer’s requirement and a provider’s offer. The thesis should also propose a mathematical model addressing the optimization problem of cloud service provider selection based on the proposed quality aspects. Finally, the efficiency of the approach should be validated through simulation studies and several evaluation scenarios.

    [1] Michael Maurer, Vincent C. Emeakaroha, Ivona Brandic, Jörn Altmann. Cost-Benefit Analysis of an SLA Mapping Approach for Defining Standardized Cloud Computing Goods. Future Generation Computer Systems, 2011.
    [2] Ivan Breskovic, Michael Maurer, Vincent C. Emeakaroha, Ivona Brandic, Jörn Altmann. Towards Autonomic Market Management in Cloud Computing Infrastructures. International Conference on Cloud Computing and Services Science (CLOSER 2011), 2011.
    [3] Ivan Breskovic, Michael Maurer, Vincent C. Emeakaroha, Ivona Brandic, Schahram Dustdar. Cost-Efficient Utilization of Public SLA Templates in Autonomic Cloud Markets. 4th IEEE International Conference on Utility and Cloud Computing (UCC 2011), 2011.

  • Knowledge Management in Self-Aware Cloud Marketplaces
     (TISS)

    Cloud computing is a novel computing paradigm that offers resources (i.e., data, software, and hardware services) in a manner similar to traditional utilities such as water, electricity, gas, and telephony. However, the current cloud market is fragmented and static, preventing the successful implementation of ubiquitous computing on demand. In order to address the issue of fragmentation, commodity cloud computing markets have been suggested.

    A large body of research into the cloud paradigm has yielded the technological development of cloud infrastructures, such as development of the appropriate resource management models, solutions for the energy efficient management of clouds as well as security and privacy solutions. Yet, very little research exists on the development of appropriate market platforms in a similar way to other commodities like energy, stocks, and assets. In the case of public and hybrid clouds, flexible, dynamic, and low entry-barrier markets with sufficient stability are crucial if computation (in general) is to become a commodity.

    Today the static nature of the cloud market means it cannot adapt adequately to dynamic changes in user requirements or new services. Moreover, due to the large variability in resource types and still low number of traders, markets suffer from low liquidity (the ability to easily and quickly sell or purchase a service at a certain price), repelling potential consumers and disadvantaging new providers. Thus, appropriate methodologies and techniques for the definition and management of Cloud market platforms are crucial and will affect whether cloud computing can emerge as a new state of the art technology for utility computing. In [1], it is shown that autonomic capabilities (self-optimization, self-configuration, self-healing and self-protection) are essential for the creation of such market platforms. To create a market platform with each of these capabilities harmoniously working together is, however, a very challenging task.

    In this thesis, a knowledge management mechanism for self-aware cloud markets should be developed. In particular, the methodology for the monitoring of cloud markets presented in [1] should be used to analyze the behavior of a cloud market platform from both the economical and infrastructural perspectives. Using a market simulator (e.g., JASA [2]), several allocation mechanisms (e.g., double, English, Dutch, first-price sealed-bid, and continuous double auction) should be implemented, monitored, and mutually compared over various trading scenarios. A utility/cost model should be developed to quantify the performance of allocation mechanisms and the underlying infrastructure. Furthermore, a rule-based knowledge management component should be developed that will utilize the utility/cost model in order to automatically change various properties of a market mechanism with the goal of improving market performance. Using the same simulation environment, the efficiency of the approach should be validated.

    [1] Ivan Breskovic, Christian Haas, Simon Caton, Ivona Brandic. Towards Self-Awareness in Cloud Markets: A Monitoring Methodology. 9th IEEE International Conference on Dependable, Autonomic and Secure Computing (DASC 2011), 2011.
    [2] JASA: a high-performance auction simulator, http://freecode.com/projects/jasa

Other topics related to cloud computing and SLA management are also possible. If you are interested in any of my research topics, drop me a mail.

Professional Activities

Research visits

Scientific talks

  • "Towards Self-Awareness in Cloud Markets: A Monitoring Methodology". 9th IEEE International Conference on Dependable, Autonomic and Secure Computing (DASC 2011), December 13, 2011, Sydney, Australia
  • "Cost-Efficient Utilization of Public SLA Templates in Autonomic Cloud Markets". 4th IEEE International Conference on Utility and Cloud Computing (UCC 2011), December 8, 2011, Melbourne, Australia
  • "Towards Autonomic Market Management in Cloud Computing Infrastructures". International Conference on Cloud Computing and Services Science (CLOSER 2011), May 7, 2011, Noordwijkerhout, The Netherlands

Reviewer for journals

Publications

Several papers are available for download. By following these links you agree to respect the copyrights of the papers. The papers obtained from this web page are included by the contributing authors as a means to ensure timely dissemination of scholarly and technical work on a non-commercial basis. Copyright and all rights therein are maintained by the authors or by other copyright holders, notwithstanding that they have offered their works here electronically. It is understood that all persons copying this information will adhere to the terms and constraints invoked by each author’s copyright. These works may not be reposted without the explicit permission of the copyright holder.

Articles in academic magazines

Publications in conference proceedings

Book chapters

Other publications

  • Ivan Breskovic. Towards Self-Awareness in Cloud Markets. Aware Project - Self Awareness in Autonomic Systems. August, 2011
  • Vincent C. Emeakaroha, Michael Maurer, Ivan Breskovic, Ivona Brandic. Time Shared VMs and Monitoring of Time Shared VMs. Proceedings of the COST Action IC0804 on Energy Efficiency in Large Scale Distributed Systems, 2nd Year, J. Pierson, H. Hlavacs (Ed.), COST Office, 2011, p. 47-51.

Theses

  • Master’s thesis: "Extending Spatial Database Systems to Support Moving Objects" (in Croatian). University of Zagreb, Faculty of Electrical Engineering and Computing, July 2010. Advisor: Prof. dr. sc. Mirta Baranović

Contact

email:
phone:
fax:


+43-1-58801-18458
+43-1-58801-918458

Distributed Systems Group
Institute of Information Systems
Vienna University of Technology
Argentinierstrasse 8/184-1, 5th floor, room 0503
A-1040 Vienna, Austria