Two Accepted Papers@BTW’19

We are happy to announce that our papers and “Fast Approximated Nearest Neighbor Joins For Relational Database Systems” written by Michael Günther, Maik Thiele and Wolfgang Lehner and “Fighting the Duplicates in Hashing: Conflict Detection-aware Vectorization of Linear Probing” written by Johannes Pietrzyk, Annett Ungethüm, Dirk Habich and Wolfgang Lehner has been accepted as full paper at BTW 2019.

Klausureinsicht

Die Klausureinsicht zur Datenbankprüfung in Sommersemester 2018 findet am 11. Oktober 14:00 Uhr im Raum APB 3105 statt. Bitte die Teilnahme an der Einsicht per Email an maik.thiele@tu-dresden.de anzeigen.

Master/Diploma Thesis Topic

We are looking for an enthusiastic student for our Topic “AsynchronousPageRankcalculationon a NUMA-Aware, In-Memory Graph Processing Engine”.

Click here for the Flyer de | en

Best Paper Award@DATA 2018

We are happy to announce that our paper „Column Scan Optimization by Increasing Intra-Instruction Parallelism“ written by Nusrat Jahan Lisa, Annett Ungethüm, Dirk Habich, Nguyen Duy Anh Tuan, Akash Kumar, and Wolfgang Lehner has been awarded with the Best Paper of the 2018 DATA conference in Porto, Portugal.

Hinweise zur Prüfung “Datenbanken – Grundlagen”

Hinweise zur Prüfung “Datenbanken – Grundlagen”, insbesondere die Raumzuordnung finden Sie hier.

Paper@ISITA 2018

We are delighted to announce that our paper „Lower Bound-oriented Parameter Calculation for AN coding“ written by Juliana Hildebrandt, Till Kolditz, Dirk Habich, and Wolfgang Lehner has been accepted at the International Symposium on Information Theory and Its Applications (ISITA). ISITA is the leading conference on information theory.

 

Abstract: The hardware as well as software communities have recently experienced a shift towards mitigating bit flips issues in software, rather than completely mitigating only in hardware. For this software error mitigation, arithmetic error coding schemes like AN coding are increasingly applied because arithmetic operations can be directly executed without decoding and bit flip detection is provided in an end-to-end fashion. In this case, the encoded data words are computed by multiplying a constant integer value A onto each original data word. To reliably detect b bit flips in each code word, the value A has to be well-chosen, so that a minimum Hamming distance of b+1 can be guaranteed. However, the value A depends on the data word length as well as on the desired minimum Hamming distance. Up to now, a very expensive brute force approach for computation of the value for A is applied. To tackle that in a more efficient way, we present a lower bound-oriented approach for this calculation in this paper. 

Paper@ADMS 2018

We are happy to announce that our joint paper with the processor design group of Akash Kumar has been accepted at the ninth International Workshop on Accelerating Analytics and Data Management Systems Using Modern Processor and Storage Architectures (co-located to VLDB). The paper presents a column scan accleration approach for hybrid CPU-FPGA systems.

2nd place in the ACM SIGMOD 2018 Student Research Competition (SRC)

Michael Günther has been awarded 2nd place in the ACM SIGMOD 2018 Student Research Competition (SRC). Congratulations!!!

 

SIGARCH Post

In a recent post on SIGARCH about Near Data Computing from a Database Systems Perspective written by Spyros Blanas, our work on resilient data storage and query processing is mentioned as an highlight and important future topic.

Paper@GvDB-Workshop

We are happy to announce that our paper “Beyond Straightforward Vectorization of Lightweight Data Compression Algorithms for Larger Vector Sizes” written by Johannes Pietrzyk, Annett Ungethüm, Dirk Habich and Wolfgang Lehner has been accepted at the GvDB-Workshop. Johannes Pietrzyk is currently writing his diploma thesis at our chair and summarized already some results for the GvDB-workshop.

Klausureinsicht

Die Klausureinsicht zur Datenbankprüfung (Wiederholungsklausur im WS 17/18) wird am 11. April um 11:00 Uhr im Raum APB 3043 stattfinden.

Demo@SIGMOD 2018

We are happy to announce that our demo „Energy-Utility Function-Based Resource Control for In-Memory Database Systems LIVE“ has been accepted at SIGMOD 2018. This is a joint demo with the operating system group of Prof. Härtig.

Paper@ICDE-Workshop

We are happy to announce that our paper “Conflict Detection-based Run-Length Encoding — AVX512-CD Instruction Set in Action” written by Annett Ungethüm, Johannes Pietrzyk, Patrick Damme, Dirk Habich, and Wolfgang Lehner has been accepted at the joint ICDE workshop of HardBD and Active.

Database Systems meet Deep Learning

Im kommenden Sommersemester 2018 bieten wir ein Praktikum (Komplex-, Forschungs-, bzw. Masterpraktikum) zu einem neuen Ansatz an, der die Gebiete Datenbanksysteme und Deep Learning verbinden soll. Eine Beschreibung dazu findet ihr hier. Bei Interesse bitte bei Dirk Habich bzw. Claudio Hartmann melden.

DB-SyA Komplex- bzw. Forschungspraktikum im SS2018

Im kommenden Sommersemester bieten wir wieder ein spannendes SyA-Komplex- bzw. Forschungspraktikum an. Eine Beschreibung findet ihr hier: DB-SyA Praktikum.

Paper@DAS 2018

We are happy to announce that our paper “Table Recognitions in Spreadsheets via a Graph Representation” written by: Elvis Koci, Maik Thiele, Oscar Romero and Wolfgang Lehner has been accepted as full paper at DAS 2018.

Lightning Talk@ICDE 2018

We are happy to announce that we will give a lightning talk at ICDE 2018 about „Teaching In-Memory Database Systems the Detection of Hardware Errors“.