Dipl.-Inf. Felix Beier

Technische Universität Ilmenau
Fakultät für Informatik und Automatisierung
FG Datenbanken und Informationssysteme

Postfach 100 565
D-98684 Ilmenau

Phone: +49 (0) 3677 69 4555
FAX: +49 (0) 3677 69 4541
E-Mail: felix.beier[at]tu-ilmenau.de
Office: ZB 3088

Short CV

Felix Beier was born 1986 in Schlema, Germany. Beginning of 2010, he received his Diploma (M.Sc.) in Computer Science from the Ilmenau University of Technology, Germany. Since May 2010, he is a Ph.D. student in the Database and Information Systems group at the Faculty of Computer Science and Automation and the Graduate School for Image Processing and Image Interpretation of the Ilmenau University of Technology.

Research

Topics

Query coprocessing on GPGPUs
Benchmarking realtime out-of-core 3D model visualization tools

Publications

2012

Felix Beier, Torsten Kilias, Kai-Uwe Sattler
GiST Scan Acceleration using Coprocessors
Proc. of 8th Int. Workshop on Data Management on New Hardware (DaMoN 2012), Scottsdale, AZ, To appear, 2012
[Bibtex]

@Inproceedings{DBIS:BeiKilSat2012,
	Author 	= {Beier, Felix and Kilias, Torsten and Sattler, Kai-Uwe},
	Booktitle	= {Proc. of 8th Int. Workshop on Data Management on New Hardware (DaMoN 2012), Scottsdale, AZ},
	Note	= {To appear},
	Title	= {GiST Scan Acceleration using Coprocessors},
	Year	= {2012}
}

2011

Felix Beier, Knut Stolze, Kai-Uwe Sattler
Autonomous Workload-driven Reorganization of Column Groupings in MMDBS
Proceedings of the 2011 IEEE 27th International Conference on Data Engineering Workshops, April, Pages 22-27, 2011, Publisher IEEE
[Bibtex]
```
@Inproceedings{DBIS:BeiStoSat11,
	Author 	= {Beier, Felix and Stolze, Knut and Sattler, Kai-Uwe},
	Booktitle	= {Proceedings of the 2011 IEEE 27th International Conference on Data Engineering Workshops},
	ISBN	= {978-1-4244-9194-0},
	Month	= {April},
	Pages	= {22-27},
	Title	= {Autonomous Workload-driven Reorganization of Column Groupings in MMDBS},
	Url	= {http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=05767654},
	Year	= {2011},
	Publisher	= {IEEE}
}
```
[Abstract]

A current trend to achieve high query performance even for huge data warehouse and business intelligence systems is to exploit main-memory-based processing techniques such as compression, cache-conscious strategies, and optimized data structures. However, update processing and skews in data distribution might lead to degenerations in such densely packed and highly compressed data structures affecting the memory efficiency and query performance negatively. Thus, reorganization tasks for repairing these data structures are necessary but should be carefully applied in order to not impact query execution or even system availability significantly. In this paper, we consider the special problem of tuple layout in banked storage structures. Based on runtime statistics capturing typical access patterns in the current workload, we present a bank reassignment approach that can be piggybacked to maintenance tasks without any administration overhead. We have implemented this approach in IBM Smart Analytics Optimizer (ISAOPT). The results of our experimental evaluation show that a simple automatic restructuring of the considered hybrid row-column-store structures offers opportunities to improve query runtimes when a slight memory overhead is acceptable.
[Download]
Knut Stolze, Felix Beier, Oliver Koeth, Kai-Uwe Sattler
Integrating Cluster-Based Main-Memory Accelerators in Relational Data Warehouse Systems
Datenbank-Spektrum, July, 2011, Publisher Springer, 1618-2162
[Bibtex]
```
@Article{DBIS:StoBeiKoe11+,
	Author 	= {Stolze, Knut and Beier, Felix and Koeth, Oliver and Sattler, Kai-Uwe},
	Editor	= {R. Schenkel; T. H{\"a}rder},
	Journal	= {Datenbank-Spektrum},
	Month	= {July},
	Title	= {Integrating Cluster-Based Main-Memory Accelerators in Relational Data Warehouse Systems},
	Url	= {http://www.springerlink.com/content/g727656075006883/},
	Year	= {2011},
	Publisher	= {Springer},
	ISSN	= {1618-2162}
}
```
[Abstract]

Today, data warehouse systems are faced with challenges for providing nearly realtime response times even for complex analytical queries on enormous data volumes. Highly scalable computing clusters in combination with parallel in-memory processing of compressed data are valuable techniques to address these challenges. In this paper, we give an overview on core techniques of the IBM Smart Analytics Optimizer -- an accelerator engine for IBM's mainframe database system DB2 for z/OS. We particularly discuss aspects of a seamless integration between the two worlds and describe techniques exploiting features of modern hardware such as parallel processing, cache utilization, and SIMD. We describe issues encountered during the development and evaluation of our system and outline current research activities for solving them.
[Download]
Felix Beier, Knut Stolze, Kai-Uwe Sattler
Online Reorganization in Read Optimized MMDBS
SIGMOD 2011, June, Pages 1125-1136, 2011, Publisher ACM
[Bibtex]
```
@Inproceedings{DBIS:BeiStoSat11Reorg,
	Author 	= {Beier, Felix and Stolze, Knut and Sattler, Kai-Uwe},
	Booktitle	= {SIGMOD 2011},
	Month	= {June},
	Pages	= {1125-1136},
	Title	= {Online Reorganization in Read Optimized MMDBS},
	Url	= {http://portal.acm.org/citation.cfm?id=1989443},
	Year	= {2011},
	Publisher	= {ACM}
}
```
[Abstract]

Query performance is a critical factor in modern business intelligence and data warehouse systems. An increasing number of companies uses detailed analyses for conducting daily business and supporting management decisions. Thus, several techniques have been developed for achieving near realtime response times - techniques which try to alleviate I/O bottlenecks while increasing the throughputs of available processing units, i.e. by keeping relevant data in compressed main-memory data structures and exploiting the read-only characteristics of analytical workloads. However, update processing and skews in data distribution result in degenerations in these densely packed and highly compressed data structures affecting the memory efficiency and query performance negatively. Reorganization tasks can repair these data structures, but -- since these are usually costly operations -- require a well-considered decision which of several possible strategies should be processed and when, in order to reduce system downtimes. In this paper, we address these problems by presenting an approach for online reorganization in main-memory database systems (MMDBS). Based on a discussion of necessary reorganization strategies in IBM Smart Analytics Optimizer, a read optimized parallel MMDBS, we introduce a framework for executing arbitrary reorganization tasks online, i.e. in the background of normal user workloads without disrupting query results or performance.
[Download]
Knut Stolze, Felix Beier, Kai-Uwe Sattler, Sebastian Sprenger, Carlos Caballero Grolimund, Marco Czech
Architecture of a Highly Scalable Data Warehouse Appliance Integrated to Mainframe Database Systems
BTW 2011, February, Pages 628-639, 2011
[Bibtex]
```
@Inproceedings{DBIS:StoBeiSat11+,
	Author 	= {Stolze, Knut and Beier, Felix and Sattler, Kai-Uwe and Sprenger, Sebastian and Caballero Grolimund, Carlos and Czech, Marco},
	Booktitle	= {BTW 2011},
	Month	= {February},
	Pages	= {628-639},
	Title	= {Architecture of a Highly Scalable Data Warehouse Appliance Integrated to Mainframe Database Systems},
	Url	= {http://www.dbis.prakinf.tu-ilmenau.de/publications/files/DBIS:Bei10.pdf},
	Year	= {2011}
}
```
[Abstract]

Main memory processing and data compression are valuable techniques to address the new challenges of data warehousing regarding scalability, large data volumes, near realtime response times, and the tight connection to OLTP. The IBM Smart Analytics Optimizer (ISAOPT) is a data warehouse appliance that implements a main memory database system for OLAP workloads using a cluster-based architecture. It is tightly integrated with IBM DB2 for z/OS (DB2) to speed up complex queries issued against DB2. In this paper, we focus on autonomic cluster management, high availability, and incremental update mechanisms for data maintenance in ISAOPT.
[Download]

2010

Felix Beier
Masterthesis: Parallel and non-disruptive reorganization of distributed in-memory database tables
Ilmenau University of Technology, April, 2010
[Bibtex]
```
@Masterthesis{DBIS:Bei10,
	Author 	= {Beier,  Felix},
	Institution	= {Ilmenau University of Technology},
	Month	= {April},
	Title	= {Parallel and non-disruptive reorganization of distributed in-memory database tables},
	Year	= {2010}
}
```
[Abstract]

This diploma thesis is about the automatic reorganization of database tables in a distributed environment. For fast data scans, database tables are pre-joined and replicated to an acceleration appliance consisting of several cluster nodes which keep the data highly compressed in their fast main memory. In this context, optimal data reorganization strategies are analyzed, prototypically implemented and evaluated. Strategies range from removing deleted data, deciding whether tables are to be pre-joined or joined at run-time, or if a full recompression of the data with newly determined parameters is in order. The optimization is performed under resource restrictions and based on the online evaluation of steadily changing statistics about data access patterns and compression values of different columns. The developed algorithm runs in parallel without disrupting currently running processes.

TU Ilmenau (Homelink)