Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 26
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A communication-driven routing technique for application-specific NoCs

  • Tornero, Rafael
  • Orduña Huertas, Juan Manuel
  • Mejia, Andres
  • Duato Marín, José Francisco
  • Flich Cardo, José
Networks on Chip (NoCs) have been shown as an efficient solution to the complex on-chip communication problems derived from the increasing number of processor cores. One of the key issues in the design of NoCs is the reduction of both area and power dissipation. As a result, two-dimensional meshes have become the preferred topology, since it offers low and constant link delay. Unfortunately, manufacturing defects or even real-time failures often make the resulting topology to become irregular, preventing the use of traditional routing algorithms. This scenario shows the need for topology-agnostic routing algorithms that provide a valid routing solution when applied over any topology. This paper proposes a new communication-driven routing technique that optimizes the network performance for Application-Specific NoCs. This technique combines a flexible, topology-agnostic routing algorithm with a communication-aware mapping technique that matches the traffic generated by the application with the available network bandwidth. Since the mapping technique can be pruned as needed in order to fit either quality function values or time constraints, this technique can be adapted to fit with different computational costs. The evaluation results show that it significantly improves network performance in terms of both latency and power consumption., This work has been jointly supported by the Spanish MICINN, the European Commission FEDER funds, and the University of Valencia under grants Consolider-Ingenio 2010 CSD2006-00046, TIN2009-14475-C04-04, and V_SEGLES_PIE.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Leveraging task-parallelism in message-passing dense matrix factorizations using SMPSs

  • Martín, Alberto F.
  • Reyes, Ruyman
  • Badia, Rosa M.
  • Quintana-Ortí, Enrique S.
In this paper, we investigate how to exploit task-parallelism during the execution of the Cholesky factorization on clusters of multicore processors with the SMPSs programming model. Our analysis reveals that the major difficulties in adapting the code for this operation in ScaLAPACK to SMPSs lie in algorithmic restrictions and the semantics of the SMPSs programming model, but also that they both can be overcome with a limited programming effort. The experimental results report considerable gains in performance and scalability of the routine parallelized with SMPSs when compared with conventional approaches to execute the original ScaLAPACK implementation in parallel as well as two recent message-passing routines for this operation. In summary, our study opens the door to the possibility of reusing message-passing legacy codes/libraries for linear algebra, by introducing up-to-date techniques like dynamic out-of-order scheduling that significantly upgrade their performance, while avoiding a costly rewrite/reimplementation. © 2014 Elsevier B.V. All rights reserved., This research was supported by Project EU INFRA-2010-1.2.2 ‘‘TEXT: Towards EXaflop applicaTions’’. The researcher at BSC-CNS was supported by the HiPEAC-2 Network of Excellence (FP7/ICT 217068), the Spanish Ministry of Education (CICYT TIN2011-23283, TIN2007-60625 and CSD2007-00050), and theGeneralitat de Catalunya (2009-SGR-980). The researcher at CIMNE was partially funded by the UPC postdoctoral grants under the programme ‘‘BKC5-Atracció i Fidelització de talent al BKC’’. The researcher at UJI was supported by project CICYT TIN2008-06570-C04-01 and FEDER. We thank Jesus Labarta, from BSC-CNS, for helpful discussions on SMPSs and his help with the performance analysis of the codes with Paraver . We thank Vladimir Marjanovic, also from BSC-CNS, for his help in the set-up and tuning of the MPI/SMPSs tools on JUROPA . Finally, we thank Rafael Mayo, from UJI, for his support in the preliminary stages of this work. The authors gratefully acknowledge the computing time granted on the supercomputer JUROPA at Jülich Supercomputing Centre., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

ACOTES project: Advanced compiler technologies for embedded streaming

  • Duranton, M.
  • Munk, H.
  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
  • Bastoul, C.
  • Carpenter, Paul Matthew
  • Chamski, Z.
  • Cohen, A.
  • Cornero, M.
  • Dumont, P.
  • Pop, S.
  • Pop, A.
  • Ornstein, A.
  • Nuzman, D.
  • Miranda, C.
  • Martorell Bofill, Xavier|||0000-0002-0417-3430
  • Lindwer, M.
  • Ladelsky, R.
  • Ferrer, Roger
  • Fellahi, M.
  • Pouchet, L. N
  • Zaks, A.
  • Shvadron, U.
  • Trifunovic, K.
  • Rohou, E.
  • Rosen, I.
  • Ramírez Bellido, Alejandro
  • Ródenas, D.
Streaming applications are built of data-driven, computational components, consuming and producing unbounded data streams. Streaming oriented systems have become dominant in a wide range of domains, including embedded applications and DSPs. However, programming efficiently for streaming architectures is a challenging task, having to carefully partition the computation and map it to processes in a way that best matches the underlying streaming architecture, taking into account the distributed resources (memory, processing, real-time requirements) and communication overheads (processing and delay). These challenges have led to a number of suggested solutions, whose goal is to improve the programmer’s productivity in developing applications that process massive streams of data on programmable, parallel embedded architectures. StreamIt is one such example. Another more recent approach is that developed by the ACOTES project (Advanced Compiler Technologies for Embedded Streaming). The ACOTES approach for streaming applications consists of compiler-assisted mapping of streaming tasks to highly parallel systems in order to maximize cost-effectiveness, both in terms of energy and in terms of design effort. The analysis and transformation techniques automate large parts of the partitioning and mapping process, based on the properties of the application domain, on the quantitative information about the target systems, and on programmer directives. This paper presents the outcomes of the ACOTES project, a 3-year collaborative work of industrial (NXP, ST, IBM, Silicon Hive, NOKIA) and academic (UPC, INRIA, MINES ParisTech) partners, and advocates the use of Advanced Compiler Technologies that we developed to support Embedded Streaming., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Social based layouts for the increase of locality in graph operations

  • Prat Pérez, Arnau
  • Domínguez Sal, David
  • Larriba Pey, Josep|||0000-0002-7070-9256
Graphs provide a natural data representation for analyzing the relationships among entities in many application areas. Since the analysis algorithms perform memory intensive operations, it is important that the graph layout is adapted to take advantage of the memory hierarchy. Here, we propose layout strategies based on community detection to improve the in-memory data locality of generic graph algorithms. We conclude that the detection of communities in a graph provides a layout strategy that improves the performance of graph algorithms consistently over other state of the art strategies.
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Leveraging task-parallelism in message-passing dense matrix factorizations using SMPSs

  • Martín Huertas, Alberto Francisco|||0000-0001-5751-4561
  • Reyes, Ruyman
  • Badia Sala, Rosa Maria|||0000-0003-2941-5499
  • Quintana Ortí, Enrique Salvador
In this paper, we investigate how to exploit task-parallelism during the execution of the Cholesky factorization on clusters of multicore processors with the SMPSs programming model. Our analysis reveals that the major difficulties in adapting the code for this operation in ScaLAPACK to SMPSs lie in algorithmic restrictions and the semantics of the SMPSs programming model, but also that they both can be overcome with a limited programming effort. The experimental results report considerable gains in performance and scalability of the routine parallelized with SMPSs when compared with conventional approaches to execute the original ScaLAPACK implementation in parallel as well as two recent message-passing routines for this operation. In summary, our study opens the door to the possibility of reusing message-passing legacy codes/libraries for linear algebra, by introducing up-to-date techniques like dynamic out-of-order scheduling that significantly upgrade their performance, while avoiding a costly rewrite/reimplementation., This research was supported by Project EU INFRA-2010-1.2.2 \TEXT:Towards EXa op applicaTions". The researcher at BSC-CNS was supported by the HiPEAC-2 Network of Excellence (FP7/ICT 217068), the Spanish Ministry of Education (CICYT TIN2011-23283, TIN2007-60625 and CSD2007- 00050), and the Generalitat de Catalunya (2009-SGR-980). The researcher at CIMNE was partially funded by the UPC postdoctoral grants under the programme \BKC5-Atracció i Fidelització de talent al BKC". The researcher at UJI was supported by project CICYT TIN2008-06570-C04-01 and FEDER. We thank Jesus Labarta, from BSC-CNS, for helpful discussions on SMPSs and his help with the performance analysis of the codes with Paraver. We thank Vladimir Marjanovic, also from BSC-CNS, for his help in the set-up and tuning of the MPI/SMPSs tools on JuRoPa. Finally, we thank Rafael Mayo, from UJI, for his support in the preliminary stages of this work. The authors gratefully acknowledge the computing time granted on the supercomputer JuRoPa at Jülich Supercomputing Centrer., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Extending OpenMP to survive the heterogeneous multi-core era

  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
  • Badia Sala, Rosa Maria|||0000-0003-2941-5499
  • Bellens, Pieter
  • Cabrera, Daniel
  • Duran González, Alejandro
  • Ferrer, Roger
  • González Tallada, Marc|||0000-0002-3780-1106
  • Igual Peña, Francisco D.
  • Jiménez González, Daniel|||0000-0001-6064-7883
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
  • Martinell Andreu, Luis
  • Martorell Bofill, Xavier|||0000-0002-0417-3430
  • Mayo Gual, Rafael
  • Pérez Cáncer, Josep Maria|||0000-0002-0558-7600
  • Planas, Judit
  • Quintana Ortí, Enrique Salvador
This paper advances the state-of-the-art in programming models for exploiting task-level parallelism on heterogeneous many-core systems, presenting a number of extensions to the OpenMP language inspired in the StarSs programming model. The proposed extensions allow the programmer to write portable code easily for a number of different platforms, relieving him/her from developing the specific code to off-load tasks to the accelerators and the synchronization of tasks. Our results obtained from the StarSs instantiations for SMPs, theCell, and GPUs report reasonable parallel performance. However, the real impact of our approach in is the productivity gains it yields for the programmer.
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

On the simulation of large-scale architectures using multiple application abstraction levels

  • Rico Carro, Alejandro
  • Cabarcas, Felipe
  • Villavieja Prados, Carlos
  • Pavlovic, Milan
  • Vega, Augusto
  • Etsion, Yoav
  • Ramírez Bellido, Alejandro
  • Valero Cortés, Mateo|||0000-0003-2917-2482
Simulation is a key tool for computer architecture research. In particular, cycle-accurate simulators are extremely important for microarchitecture exploration and detailed design decisions, but they are slow and, so, not suitable for simulating large-scale architectures, nor are theymeant for this. Moreover,microarchitecture design decisions are irrelevant, or even misleading, for early processor design stages and high-level explorations. This allows one to raise the abstraction level of the simulated architecture, and also the application abstraction level, as it does not necessarily have to be represented as an instruction stream. In this paper we introduce a definition of different application abstraction levels, and how these are employed in TaskSim, a multi-core architecture simulator, to provide several architecture modeling abstractions, and simulate large-scale architectures with hundreds of cores. We compare the simulation speed of these abstraction levels to the ones in existing simulation tools, and also evaluate their utility and accuracy. Our simulations show that a very high-level abstraction, which may be even faster than native execution, is useful for scalability studies on parallel applications; and that just simulating explicit memory transfers, we achieve accurate simulations for architectures using non-coherent scratchpad memories, with just a 25xslowdown compared to native execution. Furthermore, we revisit trace memory simulation techniques, that are more abstract than instruction-by-instruction simulations and provide an 18x simulation speedup., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Simulating whole supercomputer applications

  • González García, Juan
  • Casas, Marc|||0000-0003-4564-2093
  • Giménez Lucas, Judit
  • Moretó Planas, Miquel|||0000-0002-9848-8758
  • Ramírez Bellido, Alejandro
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
  • Valero Cortés, Mateo|||0000-0003-2917-2482
Detailed simulations of large scale message-passing interface parallel applications are extremely time consuming and resource intensive. A new methodology that combines signal processing and data mining techniques plus a multilevel simulation reduces the simulated data by various orders of magnitude. This reduction makes possible detailed software performance analysis and accurate performance predictions in a reasonable time., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Evaluation of OpenMP task scheduling strategies

  • Duran González, Alejandro
  • Corbalán González, Julita|||0000-0002-3926-5634
  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
OpenMP is in the process of adding a tasking model that allows the programmer to specify independent units of work, called tasks, but does not specify how the scheduling of these tasks should be done (although it imposes some restrictions). We have evaluated different scheduling strategies (schedulers and cut-offs) with several applications and we found that work-first schedules seem to have the best performance but because of the restrictions that OpenMP imposes a breadth-first scheduler is a better choice to have as a default for an OpenMP runtime., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Nebelung: execution environment for transactional OpenMP

  • Milovanovic, M
  • Ferrer, Roger
  • Gajinov, Vladimir
  • Unsal, Osman Sabri
  • Cristal Kestelman, Adrián|||0000-0003-1277-9296
  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
  • Valero Cortés, Mateo|||0000-0003-2917-2482
Future generations of Chip Multiprocessors (CMP) will provide dozens or even hundreds of cores inside the chip. Writing applications that benefit from the massive computational power offered by these chips is not going to be an easy task for mainstream programmers who are used to sequential algorithms rather than parallel ones. This paper explores the possibility of using Transactional Memory (TM) in OpenMP, the industrial standard for writing parallel programs on shared-memory architectures, for C, C++ and Fortran. One of the major complexities in writing OpenMP applications is the use of critical regions (locks), atomic regions and barriers to synchronize the execution of parallel activities in threads. TM has been proposed as a mechanism that abstracts some of the complexities associated with concurrent access to shared data while enabling scalable performance. The paper presents a first proof-of-concept implementation of OpenMP with TM. Some language extensions to OpenMP are proposed to express transactions. These extensions are implemented in our source-to-source OpenMP Mercurium compiler and our Software Transactional Memory (STM) runtime system Nebelung that supports the code generated by Mercurium. Hardware Transactional Memory (HTM) or Hardware-assisted STM (HaSTM) are seen as possible paths to make the tandem TM-OpenMP more scalable. In the evaluation section we show the preliminary results. The paper finishes with a set of open issues that still need to be addressed, either in OpenMP or in the hardware/software implementations of TM., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Migration of a generic multi-physics framework to HPC environments

  • Dadvand, Pooyan|||0000-0001-9515-9273
  • Rossi, Riccardo|||0000-0003-0528-7074
  • Gil, Marisa|||0000-0001-9755-3311
  • Martorell Bofill, Xavier|||0000-0002-0417-3430
  • Cotela Dalmau, Jordi|||0000-0002-4635-0423
  • Juanpere, E.
  • Idelsohn Barg, Sergio Rodolfo
  • Oñate Ibáñez de Navarra, Eugenio|||0000-0002-0804-7095
Creating a highly parallelizable code is a challenge specially for distributed memory machines (DMMs). Moreover, algorithms and data structures suitable for these platforms can be very different from the ones used in serial code. For this reason, many programmers in the field prefer to start their own code from scratch. However, for an already existing framework supported by a long-time expertise the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics [1] is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the x86 architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show high scalability, from which we present the Telescope problem in this paper.
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Hybrid transactional memory with pessimistic concurrency control

  • Vallejo, Enrique
  • Sanyal, Sutirtha
  • Harris, Tim
  • Vallejo, Fernando
  • Beivide, Ramón
  • Unsal, Osman Sabri
  • Cristal Kestelman, Adrián|||0000-0003-1277-9296
  • Valero Cortés, Mateo|||0000-0003-2917-2482
Transactional Memory (TM) intends to simplify the design and implementation of the shared-memory data structures used in parallel software. Many Software TM systems are based on writer-locks to protect the data being modified. Such implementations can suffer from the “privatization” problem, in which transactional and non-transactional accesses to the same location can lead to inconsistent results. One solution is the use of Pessimistic Concurrency Control, but it entails an important performance penalty due to the need of reader-writer locking. In this paper a hybrid TM design is proposed to reduce the performance overheads caused by the use of these locks while combining three desirable features: i) full TM functionality whether or not the architectural support is present; ii) execution of a single common code path in software or hardware; and, iii) immunity from the privatization problem. The analysis shows how a Hybrid TM can lose important properties, such as starvation freedom. To overcome this issue, Directory Reservations is presented, a low-cost mechanism improving existent solutions designed for Hardware TM., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Extending the OpenMP tasking model to allow dependent tasks

  • Duran González, Alejandro
  • Pérez Cáncer, Josep Maria|||0000-0002-0558-7600
  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
  • Badia Sala, Rosa Maria|||0000-0003-2941-5499
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
Tasking in OpenMP 3.0 has been conceived to handle the dynamic generation of unstructured parallelism. New directives have been added allowing the user to identify units of independent work (tasks) and to define points to wait for the completion of tasks (task barriers). In this paper we propose an extension to allow the runtime detection of dependencies between generated tasks, broading the range of applications that can benefit from tasking or improving the performance when load balancing or locality are critical issues for performance. Furthermore the paper describes our proof-of-concept implementation (SMP Superscalar) and shows preliminary performance results on an SGI Altix 4700., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

A highly scalable parallel implementation of H.264

  • Azevedo, Arnaldo
  • Juurlink, Ben
  • Meenderinck, Cor
  • Terechko, Andrei
  • Hoogerbrugge, Jan
  • Álvarez Mesa, Mauricio|||0000-0002-1751-0993
  • Ramírez Bellido, Alejandro
  • Valero Cortés, Mateo|||0000-0003-2917-2482
Developing parallel applications that can harness and efficiently use future many-core architectures is the key challenge for scalable computing systems. We contribute to this challenge by presenting a parallel implementation of H.264 that scales to a large number of cores. The algorithm exploits the fact that independent macroblocks (MBs) can be processed in parallel, but whereas a previous approach exploits only intra-frame MB-level parallelism, our algorithm exploits intra-frame as well as inter-frame MB-level parallelism. It is based on the observation that inter-frame dependencies have a limited spatial range. The algorithm has been implemented on a many-core architecture consisting of NXP TriMedia TM3270 embedded processors. This required to develop a subscription mechanism, where MBs are subscribed to the kick-off lists associated with the reference MBs. Extensive simulation results show that the implementation scales very well, achieving a speedup of more than 54 on a 64-core processor, in which case the previous approach achieves a speedup of only 23. Potential drawbacks of the 3D-Wave strategy are that the memory requirements increase since there can be many frames in flight, and that the frame latency might increase. Scheduling policies to address these drawbacks are also presented. The results show that these policies combat memory and latency issues with a negligible effect on the performance scalability. Results analyzing the impact of the memory latency, L1 cache size, and the synchronization and thread management overhead are also presented. Finally, we present performance requirements for entropy (CABAC) decoding. This work was performed while the fourth author was with NXP Semiconductors., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Evaluation of OpenMP task scheduling strategies

  • Duran González, Alejandro
  • Corbalán González, Julita
  • Ayguadé Parra, Eduard
OpenMP is in the process of adding a tasking model that allows the programmer to specify independent units of work, called tasks, but does not specify how the scheduling of these tasks should be done (although it imposes some restrictions). We have evaluated different scheduling strategies (schedulers and cut-offs) with several applications and we found that work-first schedules seem to have the best performance but because of the restrictions that OpenMP imposes a breadth-first scheduler is a better choice to have as a default for an OpenMP runtime., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Hybrid transactional memory with pessimistic concurrency control

  • Vallejo, Enrique
  • Sanyal, Sutirtha
  • Harris, Tim
  • Vallejo, Fernando
  • Beivide, Ramón
  • Unsal, Osman Sabri
  • Cristal Kestelman, Adrián
  • Valero Cortés, Mateo
Transactional Memory (TM) intends to simplify the design and implementation of the shared-memory data structures used in parallel software. Many Software TM systems are based on writer-locks to protect the data being modified. Such implementations can suffer from the “privatization” problem, in which transactional and non-transactional accesses to the same location can lead to inconsistent results. One solution is the use of Pessimistic Concurrency Control, but it entails an important performance penalty due to the need of reader-writer locking. In this paper a hybrid TM design is proposed to reduce the performance overheads caused by the use of these locks while combining three desirable features: i) full TM functionality whether or not the architectural support is present; ii) execution of a single common code path in software or hardware; and, iii) immunity from the privatization problem. The analysis shows how a Hybrid TM can lose important properties, such as starvation freedom. To overcome this issue, Directory Reservations is presented, a low-cost mechanism improving existent solutions designed for Hardware TM., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

On the simulation of large-scale architectures using multiple application abstraction levels

  • Rico Carro, Alejandro
  • Cabarcas, Felipe
  • Villavieja Prados, Carlos
  • Pavlovic, Milan
  • Vega, Augusto
  • Etsion, Yoav
  • Ramírez Bellido, Alejandro
  • Valero Cortés, Mateo
Simulation is a key tool for computer architecture research. In particular, cycle-accurate simulators are extremely important for microarchitecture exploration and detailed design decisions, but they are slow and, so, not suitable for simulating large-scale architectures, nor are theymeant for this. Moreover,microarchitecture design decisions are irrelevant, or even misleading, for early processor design stages and high-level explorations. This allows one to raise the abstraction level of the simulated architecture, and also the application abstraction level, as it does not necessarily have to be represented as an instruction stream. In this paper we introduce a definition of different application abstraction levels, and how these are employed in TaskSim, a multi-core architecture simulator, to provide several architecture modeling abstractions, and simulate large-scale architectures with hundreds of cores. We compare the simulation speed of these abstraction levels to the ones in existing simulation tools, and also evaluate their utility and accuracy. Our simulations show that a very high-level abstraction, which may be even faster than native execution, is useful for scalability studies on parallel applications; and that just simulating explicit memory transfers, we achieve accurate simulations for architectures using non-coherent scratchpad memories, with just a 25xslowdown compared to native execution. Furthermore, we revisit trace memory simulation techniques, that are more abstract than instruction-by-instruction simulations and provide an 18x simulation speedup., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Extending OpenMP to survive the heterogeneous multi-core era

  • Ayguadé Parra, Eduard
  • Badia Sala, Rosa Maria
  • Bellens, Pieter
  • Cabrera, Daniel
  • Duran González, Alejandro
  • Ferrer, Roger
  • González Tallada, Marc
  • Igual Peña, Francisco D.
  • Jiménez González, Daniel
  • Labarta Mancho, Jesús José
  • Martinell Andreu, Luis
  • Martorell Bofill, Xavier
  • Mayo Gual, Rafael
  • Pérez Cáncer, Josep Maria
  • Planas, Judit
  • Quintana Ortí, Enrique Salvador
This paper advances the state-of-the-art in programming models for exploiting task-level parallelism on heterogeneous many-core systems, presenting a number of extensions to the OpenMP language inspired in the StarSs programming model. The proposed extensions allow the programmer to write portable code easily for a number of different platforms, relieving him/her from developing the specific code to off-load tasks to the accelerators and the synchronization of tasks. Our results obtained from the StarSs instantiations for SMPs, theCell, and GPUs report reasonable parallel performance. However, the real impact of our approach in is the productivity gains it yields for the programmer., Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Migration of a generic multi-physics framework to HPC environments

  • Dadvand, Pooyan
  • Rossi, Riccardo
  • Gil, Marisa
  • Martorell Bofill, Xavier
  • Cotela Dalmau, Jordi
  • Juanpere, E.
  • Idelsohn Barg, Sergio Rodolfo
  • Oñate Ibáñez de Navarra, Eugenio
Creating a highly parallelizable code is a challenge specially for distributed memory machines (DMMs). Moreover, algorithms and data structures suitable for these platforms can be very different from the ones used in serial code. For this reason, many programmers in the field prefer to start their own code from scratch. However, for an already existing framework supported by a long-time expertise the idea of transformation becomes attractive in order to reuse the effort done during years of development. In this presentation we explain how a relatively complex framework but with modular structure can be prepared for high performance computing with minimum modification. Kratos Multi-Physics [1] is an open source generic multi-disciplinary platform for solution of coupled problems consist of fluid, structure, thermal and electromagnetic fields. The parallelization of this framework is performed with objective of enforcing the less possible changes to its different solver modules and encapsulate the changes as much as possible in its common kernel. This objective is achieved thanks to the Kratos design and also innovative way of dealing with data transfers for a multi-disciplinary code. This work is completed by the migration of the framework from the x86 architecture to the Marenostrum Supercomputing platform. The migration has been verified by a set of benchmarks which show high scalability, from which we present the Telescope problem in this paper., Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Social based layouts for the increase of locality in graph operations

  • Prat Pérez, Arnau
  • Domínguez Sal, David
  • Larriba Pey, Josep
Graphs provide a natural data representation for analyzing the relationships among entities in many application areas. Since the analysis algorithms perform memory intensive operations, it is important that the graph layout is adapted to take advantage of the memory hierarchy. Here, we propose layout strategies based on community detection to improve the in-memory data locality of generic graph algorithms. We conclude that the detection of communities in a graph provides a layout strategy that improves the performance of graph algorithms consistently over other state of the art strategies., Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

A highly scalable parallel implementation of H.264

  • Azevedo, Arnaldo
  • Juurlink, Ben
  • Meenderinck, Cor
  • Terechko, Andrei
  • Hoogerbrugge, Jan
  • Álvarez Mesa, Mauricio
  • Ramírez Bellido, Alejandro
  • Valero Cortés, Mateo
Developing parallel applications that can harness and efficiently use future many-core architectures is the key challenge for scalable computing systems. We contribute to this challenge by presenting a parallel implementation of H.264 that scales to a large number of cores. The algorithm exploits the fact that independent macroblocks (MBs) can be processed in parallel, but whereas a previous approach exploits only intra-frame MB-level parallelism, our algorithm exploits intra-frame as well as inter-frame MB-level parallelism. It is based on the observation that inter-frame dependencies have a limited spatial range. The algorithm has been implemented on a many-core architecture consisting of NXP TriMedia TM3270 embedded processors. This required to develop a subscription mechanism, where MBs are subscribed to the kick-off lists associated with the reference MBs. Extensive simulation results show that the implementation scales very well, achieving a speedup of more than 54 on a 64-core processor, in which case the previous approach achieves a speedup of only 23. Potential drawbacks of the 3D-Wave strategy are that the memory requirements increase since there can be many frames in flight, and that the frame latency might increase. Scheduling policies to address these drawbacks are also presented. The results show that these policies combat memory and latency issues with a negligible effect on the performance scalability. Results analyzing the impact of the memory latency, L1 cache size, and the synchronization and thread management overhead are also presented. Finally, we present performance requirements for entropy (CABAC) decoding. This work was performed while the fourth author was with NXP Semiconductors., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Extending the OpenMP tasking model to allow dependent tasks

  • Duran González, Alejandro
  • Pérez Cáncer, Josep Maria
  • Ayguadé Parra, Eduard
  • Badia Sala, Rosa Maria
  • Labarta Mancho, Jesús José
Tasking in OpenMP 3.0 has been conceived to handle the dynamic generation of unstructured parallelism. New directives have been added allowing the user to identify units of independent work (tasks) and to define points to wait for the completion of tasks (task barriers). In this paper we propose an extension to allow the runtime detection of dependencies between generated tasks, broading the range of applications that can benefit from tasking or improving the performance when load balancing or locality are critical issues for performance. Furthermore the paper describes our proof-of-concept implementation (SMP Superscalar) and shows preliminary performance results on an SGI Altix 4700., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Nebelung: execution environment for transactional OpenMP

  • Milovanovic, M
  • Ferrer, Roger
  • Gajinov, Vladimir
  • Unsal, Osman Sabri
  • Cristal Kestelman, Adrián
  • Ayguadé Parra, Eduard
  • Valero Cortés, Mateo
Future generations of Chip Multiprocessors (CMP) will provide dozens or even hundreds of cores inside the chip. Writing applications that benefit from the massive computational power offered by these chips is not going to be an easy task for mainstream programmers who are used to sequential algorithms rather than parallel ones. This paper explores the possibility of using Transactional Memory (TM) in OpenMP, the industrial standard for writing parallel programs on shared-memory architectures, for C, C++ and Fortran. One of the major complexities in writing OpenMP applications is the use of critical regions (locks), atomic regions and barriers to synchronize the execution of parallel activities in threads. TM has been proposed as a mechanism that abstracts some of the complexities associated with concurrent access to shared data while enabling scalable performance. The paper presents a first proof-of-concept implementation of OpenMP with TM. Some language extensions to OpenMP are proposed to express transactions. These extensions are implemented in our source-to-source OpenMP Mercurium compiler and our Software Transactional Memory (STM) runtime system Nebelung that supports the code generated by Mercurium. Hardware Transactional Memory (HTM) or Hardware-assisted STM (HaSTM) are seen as possible paths to make the tandem TM-OpenMP more scalable. In the evaluation section we show the preliminary results. The paper finishes with a set of open issues that still need to be addressed, either in OpenMP or in the hardware/software implementations of TM., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Leveraging task-parallelism in message-passing dense matrix factorizations using SMPSs

  • Martín Huertas, Alberto Francisco
  • Reyes, Ruyman
  • Badia Sala, Rosa Maria
  • Quintana Ortí, Enrique Salvador
In this paper, we investigate how to exploit task-parallelism during the execution of the Cholesky factorization on clusters of multicore processors with the SMPSs programming model. Our analysis reveals that the major difficulties in adapting the code for this operation in ScaLAPACK to SMPSs lie in algorithmic restrictions and the semantics of the SMPSs programming model, but also that they both can be overcome with a limited programming effort. The experimental results report considerable gains in performance and scalability of the routine parallelized with SMPSs when compared with conventional approaches to execute the original ScaLAPACK implementation in parallel as well as two recent message-passing routines for this operation. In summary, our study opens the door to the possibility of reusing message-passing legacy codes/libraries for linear algebra, by introducing up-to-date techniques like dynamic out-of-order scheduling that significantly upgrade their performance, while avoiding a costly rewrite/reimplementation., This research was supported by Project EU INFRA-2010-1.2.2 \TEXT:Towards EXa op applicaTions". The researcher at BSC-CNS was supported by the HiPEAC-2 Network of Excellence (FP7/ICT 217068), the Spanish Ministry of Education (CICYT TIN2011-23283, TIN2007-60625 and CSD2007- 00050), and the Generalitat de Catalunya (2009-SGR-980). The researcher at CIMNE was partially funded by the UPC postdoctoral grants under the programme \BKC5-Atracció i Fidelització de talent al BKC". The researcher at UJI was supported by project CICYT TIN2008-06570-C04-01 and FEDER. We thank Jesus Labarta, from BSC-CNS, for helpful discussions on SMPSs and his help with the performance analysis of the codes with Paraver. We thank Vladimir Marjanovic, also from BSC-CNS, for his help in the set-up and tuning of the MPI/SMPSs tools on JuRoPa. Finally, we thank Rafael Mayo, from UJI, for his support in the preliminary stages of this work. The authors gratefully acknowledge the computing time granted on the supercomputer JuRoPa at Jülich Supercomputing Centrer., Peer Reviewed, Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

ACOTES project: Advanced compiler technologies for embedded streaming

  • Duranton, M.
  • Munk, H.
  • Ayguadé Parra, Eduard
  • Bastoul, C.
  • Carpenter, Paul Matthew
  • Chamski, Z.
  • Cohen, A.
  • Cornero, M.
  • Dumont, P.
  • Pop, S.
  • Pop, A.
  • Ornstein, A.
  • Nuzman, D.
  • Miranda, C.
  • Martorell Bofill, Xavier
  • Lindwer, M.
  • Ladelsky, R.
  • Ferrer, Roger
  • Fellahi, M.
  • Pouchet, L. N
  • Zaks, A.
  • Shvadron, U.
  • Trifunovic, K.
  • Rohou, E.
  • Rosen, I.
  • Ramírez Bellido, Alejandro
  • Ródenas, D.
Streaming applications are built of data-driven, computational components, consuming and producing unbounded data streams. Streaming oriented systems have become dominant in a wide range of domains, including embedded applications and DSPs. However, programming efficiently for streaming architectures is a challenging task, having to carefully partition the computation and map it to processes in a way that best matches the underlying streaming architecture, taking into account the distributed resources (memory, processing, real-time requirements) and communication overheads (processing and delay). These challenges have led to a number of suggested solutions, whose goal is to improve the programmer’s productivity in developing applications that process massive streams of data on programmable, parallel embedded architectures. StreamIt is one such example. Another more recent approach is that developed by the ACOTES project (Advanced Compiler Technologies for Embedded Streaming). The ACOTES approach for streaming applications consists of compiler-assisted mapping of streaming tasks to highly parallel systems in order to maximize cost-effectiveness, both in terms of energy and in terms of design effort. The analysis and transformation techniques automate large parts of the partitioning and mapping process, based on the properties of the application domain, on the quantitative information about the target systems, and on programmer directives. This paper presents the outcomes of the ACOTES project, a 3-year collaborative work of industrial (NXP, ST, IBM, Silicon Hive, NOKIA) and academic (UPC, INRIA, MINES ParisTech) partners, and advocates the use of Advanced Compiler Technologies that we developed to support Embedded Streaming., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Simulating whole supercomputer applications

  • González García, Juan
  • Casas, Marc
  • Giménez Lucas, Judit
  • Moretó Planas, Miquel
  • Ramírez Bellido, Alejandro
  • Labarta Mancho, Jesús José
  • Valero Cortés, Mateo
Detailed simulations of large scale message-passing interface parallel applications are extremely time consuming and resource intensive. A new methodology that combines signal processing and data mining techniques plus a multilevel simulation reduces the simulated data by various orders of magnitude. This reduction makes possible detailed software performance analysis and accurate performance predictions in a reasonable time., Peer Reviewed, Postprint (published version)