About me
I am Cagri Balkesen and this is my personal homepage. I am currently studying Computer Science (Informatik) as a graduate student at ETH Zurich (Swiss Federal Institute of Technology), Switzerland. Before coming to ETH Zurich, I studied Computer Engineering at Middle East Technical University, Turkey. At ETH Zurich, I am working as a research assistant in Systems Group at ETH Zurich.
In June 2009, I graduated from ETH Computer Science Department as a Master of Computer Science ETH in Information Systems. Since then, I pursue my Ph.D. studies at ETH Zurich.
Research Interests
Main interest areas of research: Data Management, Database Systems, Architecture and Internals of a DBMS, Parallel Data Processing on Modern Hardware, Data Stream Management, Market Data Processing, Complex Event Processing, Parallel Stream Processing.
Publications
2011- C. Balkesen, N. Tatbul "Scalable Data Partitioning Techniques for Parallel Sliding Window Processing over Data Streams", VLDB International Workshop on Data Management for Sensor Networks (DMSN'11), Seattle, WA, USA, August 2011. (Slides: pdf, pptx)
- L. Aders, R. Buffat, Z. Chothia, M. Wetter, C. Balkesen, P. M. Fischer "DEBS 2011 Grand Challenge: Streams, Rules, or a Custom Solution?", Technical Report TR-749, ETH Zurich Department of Computer Science, July 2011.
- N. Dindar, C. Balkesen, K. Kromwijk, N. Tatbul, "Event Processing Support for Cross-Reality Environments", IEEE Pervasive Computing Magazine 8(3),Special Issue on Cross-Reality Environments, July 2009.
- K. Kromwijk, C. Balkesen, G. Boder, N. Dindar, F. Keusch, A. Sengul, N. Tatbul, "Connecting the Real World with the Virtual World: The SmartRFLib RFID-Supported Library System on Second Life", Book chapter in "Handbook of Research on Web 2.0, 3.0, and X.0: Technologies, Business, and Social Applications", edited by S. Murugesan, IGI-Global, to appear.
Articles, presentations
- EPC-Network - an internet of things infrastructure. (Paper, slides, website)
- SmartRFLib: RFID Supported Library. (SmartRFLib Website, SmartRFLib Final Report and Presentation Slides, Various links: ETH Life Magazine, March 20, 2008; Science Daily, April 1, 2008; ACM TechNews, April 7, 2008; 2L Italia, April 21, 2008; Le Matin Dimanche, May 11, 2008; Computer Power User, September 2008)
Teaching
| Fall Term '09: | TA for 252-0201-00 Information Systems [link] (Prof. Dr. Nesime Tatbul) |
| Fall Term '10: | TA for 252-0383-00 Networked Information Systems [link] (Dr. Peter Fischer, Dr. Maxim Grinev , Prof. Dr. Nesime Tatbul) |
| Fall Term '11: | TA for 252-0383-00 Networked Information Systems [link] (Prof. Dr. Nesime Tatbul) |
Projects
- Parallel Data Processing on Modern Hardware: Investigates techniques for parallel processing of database operators such as joins on modern multi-core processors by utilizing fine-granular parallelism at various levels such as thread-level, instruction-level and data-level parallelism.
- Parallel Data Stream Processing: Investigates techniques for efficiently and scalably processing multiple continuous queries on a shared-nothing architecture. The goal of the project is supporting end users' QoS requirements by using minimal number of computing nodes where the load of the queries are highly fluctuating. This is an ongoing work.
- Market Data Processing: Financial market data has a trend of ever increasing growth in volumes. In this project,
we analyzed a large-scale production-level market data processing system with the goal of optimizing its performance in both latency and throughput.
We focused on analyzing and benchmarking the FAST (FIX Adapted for STreaming) Protocol implementation component of the market
data processing system. Based on real data and workload traces, we have then carefully investigated the potential performance bottlenecks in this system, and came up with
a number of software optimization ideas to deal with these problems, which we have then implemented in our lab setup. Our improvement results have been
reported in the following Master's Thesis:
- Cagri Balkesen, "Analysis, Benchmarking, and Performance Improvement of a FAST Protocol Implementation", Master Thesis, ETH Zurich, Department of Computer Science, Systems Group, May 2009.
- ACM SIGMOD Programming Contest: Implementation of a Main-Memory DB The task in the contest was designing and implementing a main-memory transactional index. The goal was to be the fastest implementation for handling highest number of concurrent transactions per time unit. My submission was one of the top 5 performers and I was invited to attend SIGMOD conference with a travel scholarship. On this page you can find more information about my approach and implementation details as well as the source code.
Previous Research and Work Experience
| IBM Extreme Blue Internship | IBM Germany Boblingen, July 2008 - October 2008 |
| I worked as an intern at IBM Germany as a part of a global internship program called IBM Extreme Blue. I worked on a project about hardware acceleration of a data management application. In this project, we took Index AND'ing in XML Databases as a usecase application and tried to offload CPU intensive processing to an FPGA-based hardware platform. We demonstrated a speed-up of 16, which showed the potential of hardware acceleration in data management applications. Meanwhile, we also filed 4 patent applications as a result of this project work. Citation on IBM Germany's webpage | |
| Junior Software Engineer | METU Technopolis, Ankara, Turkey, June 2006 - September 2006 |
| I worked as a junior Software Engineer to develop some modules of a learning management system with C# and ASP.NET. I was involved in design and development of a web-based E-Learning Management System. I worked with Visual Studio, MS SQL Server, and Visual Source Safe. | |
| Software Research Intern | Software Research and Development Center (SRDC), METU, June 2005 - September 2005 |
Contact
|
ETH Zurich, CAB E 77.2 Universitaetstrasse 6 CH-8092 Zurich, Switzerland Email: cagri.balkesen_at_inf.ethz.ch Phone: +41 44 632 2496 |
|
|
|
|
"Imagination is more important than knowledge." - Albert Einstein
"You see things; and you say 'Why?' But I dream things that never were; and I say 'Why not?'." - George Bernard Shaw