Last updated: 1.03.2013 |
[Lehrziele] |
Lehrziele (Goals) |
Ziel des Seminars ist es, grundlegende Techniken des wissenschaftlichen Arbeitens im Bereich Verteilte Systeme mit einem Fokus auf Grid/Cloud Computing, SLAs and Negotiation, Autonomic Computing, und Green Computing zu üben: Literatursuche, selbständiges Erarbeiten wissenschaftlicher Inhalte, Präsentation eines Themas, Vortragstechnik sowie rege Diskussionsteilnahme sind daher besonders wichtig.
Goal of this seminar is to discuss and learn foundational techniques required in scientific work in the area of distributed systems with a focus on Grid/Cloud Computing, SLAs and Negotiation, Autonomic Computing, and Green Computing; in particular, literatur search, presentation of scientific results, presentation. Therefore, active participation is highly encouraged.
Termine |
Datum
Zeit
Hörsaal
Inhalt
22 März 2013 9-11 Uhr Seminarraum Argentinierstrasse 1. Oganisatorische Vorbesprechung (Slides)
2. Einführung in das Thema
3. Themenvergabe21 Juni 2013 (Track 1) 9 - 14 Uhr (bzw. kürzer/länger; abhängig von der Teilnehmer-Anzahl) Bibliothek, Institut f. Informatiossysteme, Argentinierstraße 8, 3. Stock 1. Präsentation der Seminararbeiten
2. Diskussion21 Juni 2013 (Track 2) 9 - 14 Uhr (bzw. kürzer/länger; abhängig von der Teilnehmer-Anzahl) Seminarraum 384 (Neues EI, Stiege I, 4. Stock, Gusshausstr. 27-29) 1. Präsentation der Seminararbeiten
2. Diskussion
Registration |
Registration at TISS: 7.3.2013, 12.00h - 27.3.2013, 12:00h
Inhalte (Content) |
Das Seminar diskutiert den aktuellen Entwicklungsstand und die Perspektiven von Verteilten Systemen. Dieses Jahr wirden die Themen Grid Computing, SLAs and Negotiation, Autonomic Computing, und Green Computing im Mittelpunkt stehen. Die besprochenen Technologien und zugrundeliegenden Konzepte bilden auch eine wesentliche Grundlage für Service-oriented Computing und Internet Computing.
This seminar discusses current state-of-the art and perspectives in distributed systems. This year the focus is on Grid Computing, SLAs and Negotiation, Autonomic Computing, and Green Computing. The discussed technologies and concepts constitute a paramount basis for Service-oriented Computing and Internet Computing.
List of topics
Services are no longer just based on software. The rise of crowds and teams in the Internet offers new service models based on individual and team capabilities, establishing the socalled human-based services. Furthermore, relationships among services are not longer limited to typical software relationships. This topic studies state-of-the-art techniques and models that can be used to manage service ecosystems that include both software-based and human-based services in social and cloud computing platforms.
Currently, several solutions have been provided for supporting facility sustainability and analysis, for example, monitoring and analysis energy consumptions and CO2 emissions in smart buildings and houses. This is the place where IoTs have been integrated to IoSs. Several techniques have been used to integrated sensors and smart meters to provide information for different analyses in business services. Several complex analyses have been developed based on sustainability rules. Furthermore, many cloud computing providers have offered solutions for supporting the integration of IoTs into IoSs. The goal of this seminar topic is to study the ecosystem, business models, system architectures, integration middleware, analysis techniques, and cloud computing platforms for building/facility monitoring. This topic is divided into the following connected subtopics
As these subtopics are interconnected, it is expected that the overlapping discussions among these subtopics are minimum. Students will be asked to present their discussions in that sequence of subtopics and to collaborate to avoid overlapping work. The discussions are expected to focus on: (i) real-world applications and systems, (ii)qualitative and quantitative evaluation - presenting comparison and metrics, and (iii)from specific tools/systems to generalization concepts - see how the same concepts but utilized differently in different systems.
- Subtopic 1 -- the ecosystem of facility monitoring with IoTs/IoSs: stakeholders and roles, the interactions among stakeholders, basic components in end-to-end facility monitoring, architectures of facility monitoring
- Subtopic 2: potential applications, business models and community roles of facility monitoring with IoTs/IoSs: applications for different stakeholders, different pricing models, deployment models in practices, community roles
- Subtopic 3: sensors, smart meters and integration middleware: typical sensors and meters, existing middleware, sensor integration techniques, programming platforms for sensors and middleware
- Subtopic 4: data interoperability and data management for sustainability monitoring and analysis, NoSQL data management systems for sustainability data
- Subtopic 5: cloud computing platforms for facility monitoring: which platforms? their roles and architectures, integration among cloud platforms for sustainability monitoring and analysis
- Subtopic 6: sustainability analysis and measurements for facility: analysis techniques and metrics
- Subtopic 7: Efficient application development (& deployment) for heterogenous IoT environments, including gateway environments, generative programming, comparison between gateway application environments, programming and deployment languages on gateways, and existing application development and deployment approaches.
References:
- The pacific controls system,http://www.pcs.net/
- Pachube, http://www.pachube.com/
- Google powermeter, http://www.google.com/powermeter
- Amee, http://www.amee.com/
- R. Acker and M. Massoth, Secure ubiquitous house and facility control solution, in Proceedings of the 2010 Fifth International Conference on Internet and Web Applications and Services, ser. ICIW 10. Washington, DC, USA: IEEE Computer Society, 2010, pp. 262- 267. [Online]. Available: http://dx.doi.org/10.1109/ICIW.2010.45
- A. Lipnickas, R. Rutkauskas, and R. Cerkauskas, "Interoperability of scada system applications with web services," in Intelligent Data Acquisition and Advanced Computing Systems: Technology and Applications, 2009. IDAACS 2009. IEEE International Workshop on, 2009, pp. 196- 200.
- B. L. Capehart and L. C. Capehart, Eds., Web Based Energy Information and Control Systems: Case Studies and Applications. Fairmont Press, Inc., 2005.
S. Tompros, N. Mouratidis, M. Draaijer, A. Foglar, and H. Hrasnica, "Enabling applicability of energy saving applications on the appliances of the home environment", Netwrk. Mag. of Global Internetwkg., vol. 23, pp. 8-16, November 2009. [Online]. Available: http://dx.doi.org/10.1109/MNET.2009.5350347- S. Krishnamurthy, O. Anson, L. Sapir, C. Glezer, M. Rois, I. Shub, and K. Schloeder, "Automation of facility management processes using machine-to-machine technologies," in Proceedings of the 1st international conference on The internet of things, ser. IOTâ08. Berlin, Heidelberg: Springer-Verlag, 2008, pp. 68-86. [Online]. Available: http://portal.acm.org/citation.cfm?id=1793060.1793067
- J. Choi, D. Shin, and D. Shin, "Research and implementation of the context-aware middleware for controlling home appliances," in Consumer Electronics, 2005. ICCE. 2005 Digest of Technical Papers. International Conference on, 2005, pp. 161 - 162.
- D. Guinard, V. Trifa, S. Karnouskos, P. Spiess, and D. Savio, "Interacting with the soa-based internet of things: Discovery, query, selection, and on-demand provisioning of web services," IEEE Transactions on Services Computing, vol. 3, pp. 223-235, 2010.
- A. Broering, T. Foerster, S. Jirka, and C. Priess, "Sensor bus: an intermediary layer for linking geosensors and the sensor web," in Proceedings of the 1st International Conference and Exhibition on Computing for Geospatial Research & Application, ser. COM.Geo â10. New York, NY, USA: ACM, 2010, pp. 12:1-12:8. [Online]. Available: http://doi.acm.org/10.1145/1823854.1823870
- B. Consortium, "Deliverable d2.2: End-to-end platform specification beywatch data model (annex)," June 2010, http://www.beywatch.eu/docs/D22 Annex.pdf.
- Alertme, http://www.alertme.com.
- G. Thyagarajan, V. Sarangan, A. Sivasubramaniam, R. Suriyanarayanan, P. Chitra, and A. Y, "Managing carbon footprint of buildings with ecview," Computer, vol. 99, no. PrePrints, 2010.
- Open building information exchange, http://www.obix.org/.
- Extended environments markup language (eeml), http://www.eeml.org.
Due to the virtualization overhead and the required massive scalability Cloud computing infrastructures exhibits high rate of energy waste. There are several approaches to tackle efficient management of energy consumption in clouds including management of VMs, application workload management or autonomic management of Cloud infrastructures. Following topics for the comprehensive overview about the current trends in energy efficiency have been identified:
- Cloud Energy efficiency at the VM (or VM Manager) level. Overview about the current development for the energy efficient management of VMs, including workload prediction and Cloud management concepts like federation.
- Autonomic management. Overview about the current development, technologies and recent research issues in autonomic management of Cloud infrastructures considering energy efficiency and generation of Service Level Agreements (SLAs).
Relevant Publications:
- Anton Beloglazov Rajkumar Buyya 577-578 2010 CCGRID http://dx.doi.org/10.1109/CCGRID.2010.45 conf/ccgrid/2010 db/conf/ccgrid/ccgrid2010.html#BeloglazovB10
- Andreas Berl, Erol Gelenbe, Marco Di Girolamo, Giovanni Giuliani, Hermann De Meer, Minh Quan Dang, Kostas Pentikousis. Energy-Efficient Cloud Computing, The Computer Journal (2010) 53(7): 1045-1051.doi: 10.1093/comjnl/bxp080First published online: August 19, 2009
- Borja Sotomayor, Rubén S. Montero, Ignacio M. Llorente, and Ian Foster Virtual Infrastructure Management in Private and Hybrid Clouds EEE Internet Computing, September/October 2009, pp. 14-22
- Eduard Ayguadé and Jordi Torres. Holistic Management for a more Energy-Efficient Cloud Computing, ERCIM News 83, Special Issue Cloud Computing, 2010.
Knowledge management is necessary in autonomic Cloud systems in order to learn about the system behavior and to develop strategies for the reaction to prevent system failures and to optimize resource consumption. There are two distinct areas for the knowledge management n Clouds: SLA management and system learning approaches. Following topics for the comprehensive overview about the current trends in knowledge management in Clouds have been identified:
- SLA Management includes generation of SLA agreements considering system utilization and requested resources of an applications executed in the Clouds.
- Learning approaches in clouds include techniques and methodologies for learning from past experiences in order to prevent violation of established SLA agreements in clouds.
Relevant Publications:
- [M. Maurer et al. 2010b] M. Maurer, I. Brandic, and R. Sakellariou. Simulating Autonomic SLA Enactment in Clouds using Case Based Reasoning. ServiceWave 2010, Ghent, Belgium, December 2010.
- [Wood et al. 2009] T. Wood, P. J. Shenoy, A. Venkataramani, M. S. Yousif. Sandpiper: Black-box and gray-box resource management for virtual machines, Computer Networks 53(17): 2923-2938, 2009.
Due to the virtualization layer overhead, monitoring of Cloud systems remain a challenging research issue. Thereby, monitoring of hardware resources and appropriate mapping of measured metrics to virtualization layer represent the emerging research issues. Following topics for the comprehensive overview about the current trends in energy efficiency have been identified:
- Monitoring of Cloud Hardware and mapping of the measured metrics to the virtualization and software layer
- SLA management in compliance with the monitoring metrics
Relevant Publications:
- [Fu et al. 2006] W. Fu, Q. Huang. GridEye: A service-oriented grid monitoring system with improved forecasting algorithm. GCCW 2006.
- [V. Emeakaroha et al. 2010a] V.C. Emeakaroha, I. Brandic, M. Maurer, S. Dustdar. Low Level Metrics to High Level SLAs - LoM2HiS framework: Bridging the gap between monitored metrics and SLA parameters in Cloud environments. HPCS 2010, Caen, France, June, 2010
Methodik |
Unsere Vorgehensweise im Rahmen des Seminars ist wie folgt:
1. Recherche
Der Seminararbeit und dem Vortrag sollen mindestens 5 wissenschaftliche Quellen zugrunde liegen. Die Quellen sind durch Recherche in wissenschaftlichen Bibliotheken und im Internet zu finden (siehe z.B. die Linksammlung unten). Sie sollten dokumentieren, dass Sie in der Lage sind wissenschaftliche Publikationen zu vestehen und in ihrer Arbeit zu referenzieren.
2. Seminararbeit
Die gesammelten Unterlagen sind in einer 15 seitigen Seminararbeit in englischer Sprache zusammenzufassen. Die Formatierung des Beitrags soll den Springer Verlag LNCS Richtlinien entsprechen. Senden Sie uns die PDF Version Ihrer Seminararbeit bis zum Semesterende.
3. Vortrag
In einem 20 minütigen Vortrag wird das gewählte Thema präsentiert. Anschliessend gibt es noch bis zu 10 Minuten Zeit für Diskussion. Die Folien können mittels gedruckten Overheadfolien oder mit Powerpoint bzw.ähnlichen Programmen (falls möglich, bitte eigenen Notebook mitbringen) präsentiert werden (Beamer ist vorhanden).
4. Anwesenheit
Grundsätzlich besteht bei den Seminarterminen Anwesenheitspflicht.
Bei Fragen oder Problemen können Sie uns jederzeit eine email senden oder nach Vereinbarung persönlich vorbeikommen.
Prüfung |
Die Qualität der Recherche (Auswahl der Referenzen), die Seminararbeit sowie Ihr Vortrag, bilden die Kriterien für die Benotung.
Digital Libraries |
In den hier angeführten Digital Libraries können Sie nach weiteren Artikel suchen (und finden). Falls Ihr Computer eine IP-Adresse von der TU-Wien hat (z.B. Zugriff über einen Computer in einem Informatik-Labor), haben Sie bei IEEE, ACM, Springer, Elsevier und Wiley kostenlosen Zugriff auf die dort angebotenen Artikel.
In the digital libraries listed below you can find more research papers. If your computer has an IP address of TU Wien, you have free access to journal papers at IEEE, ACM, Springer, Elsevier and Wiley.