|
|
Integrierte Hard- and Software Mechanismen für datenintensive Anwendungen auf Heterogenen Mehrkernsystemen
Zeitraum
01.08.2012 - 31.12.2014
Motivation
Mit dem Einzug von Grafikbeschleunigern in herkömmliche Desktoprechner
beginnt für diese die im Bereich mobiler eingebetteter Systeme und
High-Performance Computing schon länger angebrochene Zeit des heterogenen
Rechnens. Bei diesem Prinzip werden komplexe Berechnungen von herkömmlichen,
voll ausgestatteten Prozessorkernen auf speziell optimierte und dadurch wesentlich
effizientere Beschleunigerkerne verlagert.
Ziel der Nachwuchsforschergruppe IMData ist es, komplementär,
jedoch in Anlehnung an die Aktivitäten des Exzellenzclusters
Center for Advancing
Electronics Dresden (cfAED), intelligente, adaptive und
anwendungsbewusste Mechanismen zu entwickeln, die es ermöglichen, ein
breiteres Spektrum an komplexen Anwendungen in diesem Sinne auf Spezialhardware
zu verlagern.
Als vertikal integrierendes Projekt spiegelt IMData dabei die am
Standort Dresden beheimatete, einmalige Firmenstruktur wider. Die Forschergruppe
ist deshalb in besonderem Maße geeignet, Nachwuchsforscherinnen und
Nachwuchsforschern an etablierte lokale Forschungs- und Industrienetzwerke
heranzuführen und in der Region zu verankern.
Beteiligte Lehrstühle
Publikationen
2014
-
L2_ISA++: Instruction Set Architecture Extensions for 4G and LTE-Advanced MPSoCs
O. Arnold, F. Neumaerker and G. Fettweis
Erscheint in: In Proceedings of the International Symposium on System-on-Chip (SOC'14), Tampere, Finland, 27.10. - 29.10.2014
-
Adaptive Runtime Management of Heterogeneous MPSoCs: Analysis, Acceleration and Silicon Prototype
O. Arnold and G. Fettweis
Erscheint in: In Proceedings of the International Symposium on System-on-Chip (SOC'14), Tampere, Finland, 27.10. - 29.10.2014
-
CM_ISA++: An Instruction Set for Dynamic Task Scheduling Units for More Than 1000 Cores
O. Arnold, B. Nöthen and G. Fettweis
In Proceedings of the IEEE SOC Conference (SOCC'14), Las Vegas, Nevada, USA, 2.9. - 5.9.2014
PDF
-
Heterogeneity-Aware Operator Placement in Column-Store DBMS
Karnagel, T.; Habich, D.; Schlegel, B.; Lehner, W.
Journal: Datenbank-Spektrum, September 2014
PDF
-
HASHI: An Application-Specific Instruction Set Extension for Hashing
Arnold, O.; Haas, S.; Fettweis, G.; Schlegel, B.; Kissinger, T.; Karnagel, T.; Lehner, W.
In Proceedings of the VLDB Endowment 2014 (ADBS'14), September 1-5, 2014, Hangzhou, China
PDF
-
Locks: Picking key methods for a scalable quantitative analysis
C. Baier, M. Daum, B. Engel, H. Härtig, J. Klein, S. Klüppelholz, S. Märker, H. Tews, M. Völp
Erscheint in: Journal of Computer and System Sciences (in print), avail. online, July 2014
-
From silicon nanowire reconfigurable transistors to operating-system controlled circuit space (Poster)
M. Raitza, M. Völp
2nd Dresden Nanoanalysis Symposium (Poster Session) Dresden, Germany, July 2014
-
An Application-Specific Instruction Set for Accelerating Set-Oriented Database Primitives
O. Arnold, S. Haas, G. Fettweis, B. Schlegel, T. Kissinger and W. Lehner
in Proceedings of the SIGMOD'14 Special Interest Group on Management of Data (SIGMOD'14), Snowbird, Utah, USA, 22.6. - 27.6.2014
PDF
-
Towards Computation with Microchemomechanical Systems
A. Voigt, R. Greiner, M. Allerdissen, A. Richter, S. Henker, M. Völp
International Journal of Foundations of Computer Science , Vol 25, Issue 04, June 2014
PDF
-
Demonstrating Efficient Query Processing in Heterogeneous Environments
Karnagel, T.; Hille, M.; Ludwig, M.; Habich, D.; Lehner, W.; Heimel, M.; Markl, V.
Proceedings of the 2014 ACM SIGMOD (Demo Track), June 22-27, 2014, Snowbird, Utah, USA
PDF
-
Computing Conditional Probabilities in Markovian Models Efficiently
Christel Baier, Joachim Klein, Sascha Klüppelholz, Steffen Märcker
Proceedings of the 20th International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS'14), Grenoble, France, April 2014
PDF
-
Has Energy Surpassed Timeliness? - Scheduling Energy-Constrained Mixed-Criticality Systems
M. Völp, M. Hähnel, A. Lackorzynski
20th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2014) Berlin, Germany, April 2014
PDF
-
What if we would degrade LO tasks in mixed-criticality systems?
M. Völp
20th IEEE Real-Time and Embedded Technology and Applications Symposium - Work in Progress Session (RTAS-WIP 2014) Berlin, Germany, April 2014
PDF
-
Improving In-Memory Database Index Performance with Intel TSX
Karnagel, T.; Dementiev, R.; Rajwar, R.; Lai, K.; Legler, T.; Schlegel, B.; Lehner, W.
Proceedings of the 20th IEEE International Symposium On High Performance Computer Architecture (HPCA-2014), February 15-19, 2014, Orlando, Florida, USA
PDF
2013
-
On the Expressiveness of Fixed-Priority Scheduling Contexts for Mixed-Criticality Scheduling
M. Völp, A. Lackorzynski, H. Härtig
1st International Workshop on Mixed Criticality Systems (WMC), co-located with RTSS 2013 Vancouver, Canada, December 2013
PDF
-
Towards Elastic SDR Architectures Using Dynamic Task Management
O. Arnold, E. Matus, B. Nöthen, F. Pauls and G. Fettweis
Proceedings of the Global Conference on Signal and Information Processing (GlobalSIP'13), Austin, Texas, December 2013
PDF
-
Getting More Cores by Minimalism w/o Losing Spatial Isolation (Poster)
N. Asmussen, H. Härtig, M. Völp
SOSP 2013 - Poster Session, Farmington, PA, USA, November 2013
-
Tomahawk - Parallelism and Heterogeneity in Communications Signal Processing MPSoCs
O. Arnold, E. Matus, B. Nöthen, M. Winter, T. Limberg and G. Fettweis
ACM Transactions on Embedded Computing Systems (TECS), 2013
PDF
-
Predictable Coherent Caching with Incoherent Caches
A. Lackorzynski, B. Engel, M. Völp
Real-Time Linux WorkshopLugano-Manno, Switzerland, October 2013
PDF
-
On the Impact of Dynamic Data Management for Distributed Local Memories in Heterogeneous MPSoCs
Nöthen, Benedikt; Arnold, Oliver; Fettweis, Gerhard
Proceedings of the SoC International Symposium on System-on-Chip (SOC'13), Tampere, Finland, October 2013
PDF
-
A Flexible Analytic Model for a Dynamic Task-Scheduling Unit for Heterogeneous MPSoCs
O. Arnold, B. Nöthen and G. Fettweis
Proceedings of the International Conference on Advances in System Simulation (SIMUL'13), Venice, Italy, October 2013
PDF
-
Elastic Manycores: How to bring the OS back into the scheduling game?
Marcus Völp, Michael Roitzsch
1st Workshop on Runtime and Operating Systems for the Many-core Era (ROME 2013), Aachen, Germany, August 2013
PDF
-
Flat but Trustworthy: Security Aspects in Flattened Hierarchical Scheduling
Adam Lackorzynski, Marcus Völp and Alexander Warg
Workshop on Virtualization for Real-Time Embedded Systems (VtRES'13, colocated with RTCSA'13), Taipeh, Taiwan, August 2013
PDF
-
The Case for Practical Multi-Resource and Multi-Level Scheduling Based on Energy/Utility
Hermann Härtig, Marcus Völp, Marcus Hähnel
19th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA), Taipeh, Taiwan, August 2013
PDF
-
The HELLS-Join: A Heterogeneous Stream join for ExtremeLy Large windows
Karnagel, T.; Habich, D.; Schlegel, B.; Lehner, W.
Proceedings of the Ninth International Workshop on Data Management on New Hardware (DaMoN 2013), June 24, 2013, New York, New York, USA
PDF
-
Scalable Frequent Itemset Mining on Many-core Processors
Schlegel, B.; Karnagel, T.; Kiefer, T.; Lehner, W.
Proceedings of the Ninth International Workshop on Data Management on New Hardware (DaMoN 2013), June 24, 2013, New York, New York, USA
PDF
-
eBond: Energy Saving in Heterogeneous R.A.I.N
Marcus Hähnel, Björn Döbel, Hermann Härtig, Marcus Völp
ACM eEnergy (e-Energy 2013), Berkeley, CA, USA, May 2013
PDF
-
A Probabilistic Quantitative Analysis of Probabilistic-Write/Copy-Select
Christel Baier, Benjamin Engel, Sascha Klüppelholz, Steffen Märcker, Hendrik Tews, Marcus Völp
5th NASA Formal Methods Symposium (NFM'13), May 2013
Lecture Notes in Computer Science 7871, pp 307-321, 2013
SpringerLink
-
Turning x86 into a Hardware Simulator for Future Manycores
Nils Asmussen, Hermann Härtig, Marcus Völp
3rd Workshop on Systems for Future Multicore Architectures (SFMA'13), Prague, Czech Republic, April 2013
PDF
-
On Confidentiality-Preserving Real-Time Locking Protocols
Marcus Völp, Benjamin Engel, Claude-Joachim Hamann, Hermann Härtig
In Proceedings of the 19th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2013), April 2013
PDF
-
Consolidate-to-Idle: The Second Dimension is Almost for Free
Marcus Völp, Johannes Steinmetz, Marcus Hähnel
19th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS 2013) WiP Session, Philadelphia, USA, April 2013
PDF
-
Stream Join Processing on Heterogeneous Processors
Karnagel, T.; Schlegel, B.; Habich, D.; Lehner, W.
In Proceedings of BTW-Workshops 2013, Data Streams and Event Processing (DSEP), pages 17-26, Magdeburg, Germany
PDF
-
The Potential of Energy/Utility-Accrual Scheduling
Marcus Hähnel, Marcus Völp, Björn Döbel, Hermann Härtig
First International WorkShop on Energy-Aware Systems, Communications and Security (EASyCoSe 2013), Barcelona, Spain, March 2013
PDF
2012
-
Chiefly Symmetric: Results on the Scalability of Probabilistic Model Checking for Operating-System Code
Christel Baier, Marcus Daum, Benjamin Engel, Hermann Härtig, Joachim Klein, Sascha Klüppelholz, Steffen Märcker, Hendrik Tews, Marcus Völp
7th Conference on Systems Software Verification, Sydney, Australia, November 2012
PDF
-
Who is Going to Program This?
Marcus Völp, Michael Roitzsch, Hermann Härtig
OSDI 2012 - Poster Session, Hollywood, CA, USA, October 2012
PDF
-
The IMData Approach to Accelerate Data-Intensive Workloads
Marcus Völp, Nils Asmussen, Hermann Härtig
Facing the Multicore-Challendge III - Poster, Stuttgart, Germany, September 2012
PDF
-
Instruction Set Architecture Extensions for a Dynamic Task Scheduling Unit
O. Arnold, B. Nöthen and G. Fettweis
Proceedings of the IEEE Annual Symposium on VLSI (ISVLSI'12), August 2012
PDF
|
|
