direkt zum Inhalt springen

direkt zum Hauptnavigationsmenü

Sie sind hier

TU Berlin

Inhalt des Dokuments

Nexus++: a Hardware Task Management System for Multicore Systems

Summary

Since the trend of improving processor’s performance has shifted toward in- tegrating more cores on the same chip, several parallel programming models have been proposed that aim at relieving parallel programming. Examples include Google’s MapReduce, Intel’s TBB, and OmpSs.

OmpSs
is a promising approach, as it requires minimal effort of the pro- grammer to program in parallel since he or she can annotate the sequential application. The current software OmpSs runtime system, however, can be a bottleneck that limits the scalability of applications for large number of cores. Roughly speaking, the runtime system cannot compute task depen- dencies and attend to finished tasks fast enough to keep all cores busy. In this project we therefore develop Nexus++, a hardware system that provides architectural support for task-based parallel programming models such as OmpSs. The main goal of Nexus++ is to accelerate the task management responsibilities of the OmpSs runtime system.

Objectives

Nexus++
Nexus++ in a multicore system
Lupe [1]

To develop a hardware task management system that

  • Accelerates task dependency resolution
  • Improves the performance of application with fine-grained tasks
  • Works for arbitrary task dependency graphs
  • Requires no or minimal support from the programmer.
  • Can be integrated in any type of multicore system.

Contact

  • Tamer Dallou [2]
  • Prof. Dr. Ben Juurlink [3]
  • Dr. Cor Meenderinck [4]
  • Nina Engelhardt [5]
  • Ahmed Elhossini [6]

Funding

This research is funded by the European Community’s Seventh Framework Programme [FP7/2007-2013] under the ENCORE Project [7], grant agreement n248647. It is also funded by the German-Jordanian university [8].

Student Projects

MSc project Design of a hardware task management unit for multicore systems, Mauricio Fayula Gonzalez.

Publications

Tamer Dallou and Ben Juurlink (2012). Hardware-Based Task Dependency Resolution for the StarSs Programming Model [12]. 41st International Conference on Parallel Processing Workshops (ICPPW), SRMPDS'2012, 367 -374.


C.H. Meenderinck and Ben Juurlink (2010). A Case for Hardware Task Management Support for the StarSS Programming Model [13]. Proceedings Conference on Digital System Design Architectures, Methods and Tools


Tamer Dallou and Ahmed Elhossini and Ben Juurlink (2013). FPGA-Based Prototype of Nexus++ Task Manager [14]. Proc. 6th Workshop on Many-Task Computing on Clouds, Grids, and Supercomputers (MTAGS) 2013, Co-located with SC 2013


Nina Engelhardt and Tamer Dallou and Ahmed Elhossini and Ben Juurlink (2014). An Integrated Hardware-Software Approach to Task Graph Management [15]. Proc. 16th IEEE International Conference on High Performance and Communications HPCC-2014


Cor Meenderinck and Ben Juurlink (2011). Nexus: Hardware Support for Task-Based Programming [16]. Digital Systems Design, Euromicro Symposium on. IEEE Computer Society, 442-445.


Tamer Dallou and Ben Juurlink and Cor Meenderinck (2011). Improving the Scalability and Capabilities of the Nexus Hardware Task Management System [17]. First International Workshop on Future Architectural Support for Parallel Programming


------ Links: ------

Zusatzinformationen / Extras

Direktzugang:

Schnellnavigation zur Seite über Nummerneingabe

Copyright TU Berlin 2008