Cette page appartient aux archives web de l'EPFL et n'est plus tenue à jour.
This page belongs to EPFL's web archive and is no longer updated.

DIAS News

ViDa: Transforming Raw Data Into Information through Virtualization

The excellence of the research performed at EPFL has once again been recognized at an international level. Anastasia Ailamaki, head of Data-Intensive Applications and Systems Laboratory DIAS has been awarded a CONSOLIDATOR GRANT 2013 from the European Research Council (ERC) for

ViDa: Transforming Raw Data into Information through Virtualization

More on http://actu.epfl.ch/news/anastasia-ailamaki-awarded-a-consolidator-grant-20/

 

Posted by Dimitra Tsaoussis Melissargos at 9:16
News IC - Information and Computing Technology is Essential to the HBP

Neuroinformatics, brain simulations or supercomputing, these are research areas that are dear to Professor Anastasia Ailamaki Recently nominated co-director (with Prof. Richard Frackowiak) of the Medical Informatics Division and a workpackage leader in the High-Performance Computing Division of the Human Brain Project (HPB), Prof. Ailamaki describes the challenges that await her.

Posted by Dimitra Tsaoussis Melissargos at 11:04
Natassa in "Science! on tourne"

More and more data churned out by even more powerful supercomputers. Science has entered the era of "Very Big Data". But what for, exactly? Anastasia Ailamaki, head of the Data-intensive Applications and Systems Laboratory gives an answer to the above and other questions in "Science! on tourne" (broadcasted live on 27.03.2013)

Posted by Dimitra Tsaoussis Melissargos at 11:39
We welcome 5 new members!

We are happy to welcome 5 new members in DIAS lab!!

Mira Pavlovic started her intership this September.

Ivo Dimitrov, Manos Karpathiotakis and Stelios Styianou joined us for this semester as fellowship PhD students.

Erietta Liarou joined us as a post-doctoral researcher after spending the last few years as a PhD student at CWI, Netherlands.

Posted by Manoussos Gavriil Athanassoulis at 1:06
DIAS @ VDLB 2012

Four papers were presented in VLDB 2012 and the collocated workshop ADMS:

Porobic, D., Pandis, I., Branco, M., Tözün, P. and Ailamaki, A. OLTP on Hardware Islands. VLDB 2012

Tauheed, F., Heinis, T., Schürmann, F., Markram, H. and Ailamaki, A. SCOUT: Prefetching for Latent Structure Following Queries. VLDB 2012

Alagiannis, I., Borovica, R., Branco, M., Idreos, S. and  Ailamaki A. NoDB in Action: Adaptive Query Processing on Raw Data. VLDB 2012 (Demonstration)

Athanassoulis, M., Bhattacharjee, B., Canim, M. and Ross, K.A. Path Processing using Solid State Storage. ADMS 2012

 

Click to get all DIAS publications

Posted by Manoussos Gavriil Athanassoulis at 1:00
New publications

Two paper were recently accepted in VLDB journal and in the 45th Annual IEEE/ACM International Symposium on Microarchitecture, 2012:

Atta, I., Tözün, P., Ailamaki, A. and Moshovos, A. SLICC: Self-Assembly of Instruction Cache Collectives for OLTP Workloads. MICRO 2012

Tözün, P., Pandis, I., Johnson, F.R. and Ailamaki, A. Scalable and Dynamically Settled Shared-Everything OLTP with Physiological Partitioning. The VLDB Journal. 

 

Click to get all DIAS publications

Posted by Manoussos Gavriil Athanassoulis at 0:59
DIAS @ SIGMOD 2012

Three papers were presented in SIGMOD 2012 and the collocated workshops DaMoN and DBTest:

Alagiannis, I., Borovica, R., Branco, M., Idreos, S. and Ailamaki, A. NoDB: Efficient Query Execution on Raw Data Files. SIGMOD 2012

Atta, I., Tözün, P., Ailamaki, A. and Moshovos, A. Reducing OLTP Instruction Misses With Thread Migration. DaMoN 2012

Borovica, R., Alagiannis, I. and Ailamaki, A. Automated Physical Designers: What You See is (Not) What You Get. DBTest 2012

 

Click to get all DIAS publications

Posted by Manoussos Gavriil Athanassoulis at 0:51
DIAS @ ICDE 2012

Two papers were presented in ICDE 2012 and the collocated workshop SMBD:

Tauheed, F., Biveinis, L., Heinis, T., Schürmann, F., Markram, H. and Ailamaki, A. Accelerating Range Queries For Brain Simulations. ICDE 2012

Popescu, A.D., Ercegovac, V., Balmin, A., Branco, M. and Ailamaki, A. Same Queries, Different Data: Can we Predict Query Performance?. SMDB 2012

 

Click to get all DIAS publications

Posted by Manoussos Gavriil Athanassoulis at 0:43
Summer Interns 2012

We welcome our 2012 summer interns!
Alexandros Stougiannis already started and we wait in a few days Farzaneh Mahdisoltani. They will work with us until September.

Posted by Manoussos Gavriil Athanassoulis at 15:42
New publications

"Scalability of write-ahead logging on multicore and multisocket hardware" and "Challenges and Opportunities in Self-Managing Scientific Databases" were published early in December 2011.

Scalability of write-ahead logging on multicore and multisocket hardware: The shift to multi-core and multi-socket hardware brings new challenges to database systems, as the software parallelism determines performance. Even though database systems traditionally accommodate simultaneous requests, a multitude of synchronization barriers serialize execution. Write-ahead logging is a fundamental, omnipresent component in ARIES-style concurrency and recovery, and one of the most important yet-to-be addressed potential bottlenecks, especially in OLTP workloads making frequent small changes to data. In this paper, we identify four loggingrelated impediments to database system scalability. Each issue challenges different level in the software architecture: (a) the high volume of small-sized I/O requests may saturate the disk, (b) transactions hold locks while waiting for the log ?ush, (c) extensive context switching overwhelms the OS scheduler with threads executing log I/Os, and (d) contention appears as transactions serialize accesses to in-memory log data structures. We demonstrate these problems and address them with techniques that, when combined, comprise a holistic, scalable approach to logging. Our solution achieves a 20–69% speedup over a modern database system when running log-intensive workloads, such as the TPC-B and TATP benchmarks, in a single-socket multiprocessor server. Moreover, it achieves log insert throughput over 2.2 GB/s for small log records on the single-socket server, roughly 20 times higher than the traditional way of accessing the log using a single mutex. Furthermore, we investigate techniques on scaling the performance of logging to multi-socket servers. We present a set of optimizations which partly ameliorate the latency penalty that comes with multi-socket hardware, and then we investigate the feasibility of applying a distributed log buffer design at the socket level.

Challenges and Opportunities in Self-Managing Scientific Databases: Advances in observation instruments and abundance of computational power for simulations encourage scientists to gather and produce unprecedented amounts of increasingly complex data. Organizing data automatically to enable ef?cient and unobstructed access is pivotal for the scientists. Organizing these vast amounts of complex data, however, is particularly dif?cult for scientists who have little experience in data management; hence they spend considerable amounts of time dealing with data analysis and computing problems rather than answering scienti?c questions or developing new hypotheses. Therefore scienti?c experiments are in many ways ideal targets for research in self-managing database systems. In this paper, we describe challenges and opportunities for research in automating scienti?c data management. We ?rst discuss the problems faced in particular scienti?c domains using concrete examples of large-scale applications from neuroscience and high-energy physics. As we will show, the scienti?c questions are evolving ever more rapidly while datasets size and complexity increases. Scientists struggle to organize and reorganize the data whenever their hypothesis change and therefore their queries and their data changes as well. We identify research challenges in large-scale scienti?c data management related to self-management. By addressing these research challenges we can relieve the burden of organizing the data off the scientists, thereby ensuring that they can access it in the most ef?cient way and ultimately enabling the scientists to focus on their science.

External links:
http://www.springerlink.com/content/k747mw3213l3p472/
ftp://ftp.research.microsoft.com/pub/debull/A11dec/issue1.htm

 

Click to get all DIAS publications

Posted by Manos Athanassoulis at 15:53
Page : « Previous 1 2 3 4 5 Next »