Skip to main content

-

IJCI-2015-23266

•
Funding agency name Ministerio de Ciencia e Innovación
Funding agency acronym MICINN
Program Programa Estatal de Promoción del Talento y su Empleabilidad en I+D+i
Subprogram Subprograma Estatal de Incorporación
Call Juan de la Cierva Incorporación
Call year 2015
State Plan 2013-2016 Plan Estatal de Investigación Científica y Técnica y de Innovación
Management unit Dirección General de Investigación Científica y Técnica
Beneficiary centre BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACION

Resultados relacionados

Resultados totales (Incluyendo duplicados): 20
Encontrada(s) 1 página(s)

UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Integrating memory perspective into the BSC performance tools

  • Servat, Harald
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
  • Hoppe, Hans-Christian
  • Gimenez, Judit
  • Peña, Antonio J.
The growing gap between processor and memory speeds results in complex memory hierarchies as processors evolve to mitigate such differences by taking advantage of locality of reference. In this direction, the BSC performance analysis tools have been recently extended to provide insight relative the application memory accesses depicting their temporal and spatial characteristics, correlating with the source-code and the achieved performance simultaneously. These extensions rely on the Precise Event-Based Sampling (PEBS) mechanism available in recent Intel processors to capture information relative to the application memory accesses. The sampled information is processed with the Folding mechanism to provide a detailed temporal evolution of the memory accesses and in conjunction with the achieved performance and the source-code counterpart. The results obtained from the combination of these tools help application developers to understand better how the application behaves and how the system performs. We demonstrate the value of the complete work-flow by exploring an already optimized state-of-the-art benchmark, providing detailed insight of their memory access behavior., This work has been performed in the Intel-BSC Exascale Lab. We would like to thank Forschungszentrum Julich for the compute time on the Jureca system. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Efficient Scalable Computing through Flexible Applications and Adaptive Workloads

  • Iserte, Sergio
  • Mayo, Rafael
  • Quintana Ortí, Enrique Salvador
  • Beltran Querol, Vicenç
  • Peña, Antonio J.
In this paper we introduce a methodology for dynamic job reconfiguration driven by the programming model runtime in collaboration with the global resource manager. We improve the system throughput by exploiting malleability techniques (in terms of number of MPI ranks) through the reallocation of resources assigned to a job during its execution. In our proposal, the OmpSs runtime reconfigures the number of MPI ranks during the execution of an application in cooperation with the Slurm workload manager. In addition, we take advantage of OmpSs offload semantics to allow application developers deal with data redistribution. By combining these elements a job is able to expand itself in order to exploit idle nodes or be shrunk if other queued jobs could be initiated. This novel approach adapts the system workload in order to increase the throughput as well as make a smarter use of the underlying resources. Our experiments demonstrate that this approach can reduce the total execution time of a practical workload by more than 40% while reducing the amount of resources by 30%., This work is supported by the Project TIN2014-53495-R and TIN2015-65316-P from MINECO and FEDER. Antonio J. Peña is cofinanced by MINECO under Juan de la Cierva fellowship number IJCI-2015-23266. Special thanks to José I. Aliaga for the conjugate gradient code., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Automating the application data placement in hybrid memory systems

  • Servat, Harald
  • Peña, Antonio J.
  • Llort, German
  • Mercadal, Estanislao
  • Hoppe, Hans-Christian
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
Multi-tiered memory systems, such as those based on Intel® Xeon Phi™processors, are equipped with several memory tiers with different characteristics including, among others, capacity, access latency, bandwidth, energy consumption, and volatility. The proper distribution of the application data objects into the available memory layers is key to shorten the time– to–solution, but the way developers and end-users determine the most appropriate memory tier to place the application data objects has not been properly addressed to date.In this paper we present a novel methodology to build an extensible framework to automatically identify and place the application’s most relevant memory objects into the Intel Xeon Phi fast on-package memory. Our proposal works on top of inproduction binaries by first exploring the application behavior and then substituting the dynamic memory allocations. This makes this proposal valuable even for end-users who do not have the possibility of modifying the application source code. We demonstrate the value of a framework based in our methodology for several relevant HPC applications using different allocation strategies to help end-users improve performance with minimal intervention. The results of our evaluation reveal that our proposal is able to identify the key objects to be promoted into fast on-package memory in order to optimize performance, leading to even surpassing hardware-based solutions., This work has been performed in the Intel-BSC Exascale Lab. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. We would like to thank the Intel’s DCG HEAT team for allowing us to access their computational resources. We also want to acknowledge this team, especially Larry Meadows and Jason Sewall, as well as Pardo Keppel for the productive discussions. We thank Raphaël Léger for allowing us to access the MAXW-DGTD application and its input., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

GLT: A Unified API for Lightweight Thread Libraries

  • Castelló, Adrián
  • Seo, Sangmin
  • Mayo, Rafael
  • Balaji, Pavan
  • Quintana Ortí, Enrique Salvador
  • Peña, Antonio J.
In recent years, several lightweight thread (LWT) libraries have emerged to tackle exascale challenges. These offer programming models (PMs) based on user-level threads and incorporate their own lightweight mechanisms. However, each library proposes its own PM, exposing different semantics and hindering portability. To address this drawback, we have designed Generic Lightweight Thread (GLT), an application programming interface that frames the functionality of the most popular LWT libraries for high-performance computing under a single PM. We implement GLT on top of Argobots, MassiveThreads, and Qthreads. We provide GLT as a dynamic library, as well as in the form of a static version based on macro preprocessing resolution to reduce overhead. This paper discusses the GLT PM and demonstrates its minimal performance impact., Researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO, the Generalitat Valenciana fellowship programme Vali+d 2015, and FEDER. Antonio J. Peña is cofinancied by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

On the adequacy of lightweight thread approaches for high-level parallel programming models

  • Castelló, Adrián
  • Mayo Gual, Rafael
  • Sala Penadés, Kevin|||0000-0001-8233-1185
  • Beltran Querol, Vicenç
  • Balaji, Pavan
  • Peña, Antonio J.
High-level parallel programming models (PMs) are becoming crucial in order to extract the computational power of current on-node multi-threaded parallelism. The most popular PMs, such as OpenMP or OmpSs, are directive-based: the complexity of the hardware is hidden by the underlying runtime system, improving coding productivity. The implementations of OpenMP usually rely on POSIX threads (pthreads), offering excellent performance for coarse-grained parallelism and a perfect match with the current hardware. OmpSs is a task oriented PM based on an ad hoc runtime solution called Nanos++; it is the precursor of the tasking parallelism in the OpenMP tasking specification. A recent trend in runtimes and applications points to leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. In this paper we analyze the behavior of the OpenMP and OmpSs PMs on top of the recently emerged Generic Lightweight Threads (GLT) API. GLT exposes a common API for lightweight thread (LWT) libraries that offers the possibility of running the same application over different native LWT solutions. We describe the design details of those high-level PMs implemented on top of GLT and analyze different scenarios in order to assess where the use of LWTs may benefit application performance. Our work reveals those scenarios where LWTs overperform pthread-based solutions and compares the performance between an ad hoc solution and a generic implementation., The researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO, Spain and FEDER, Spain, the Generalitat Valenciana fellowship programme, Spain Vali+d 2015. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness, Spain under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357. We gratefully acknowledge Enrique S. Quintana-Ortí (Universitat Jaume I) and Sangmin Seo (Samsung Corp.) for their advice in this work and the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

cuHinesBatch: solving multiple hines systems on GPUs Human Brain Project

  • Valero-Lara, Pedro
  • Martinez-Perez, Ivan
  • Peña, Antonio J.
  • Martorell Bofill, Xavier|||0000-0002-0417-3430
  • Sirvent, Raul
  • Labarta Mancho, Jesús José|||0000-0002-7489-4727
The simulation of the behavior of the Human Brain is one of the most important challenges today in computing. The main problem consists of finding efficient ways to manipulate and compute the huge volume of data that this kind of simulations need, using the current technology. In this sense, this work is focused on one of the main steps of such simulation, which consists of computing the Voltage on neurons’ morphology. This is carried out using the Hines Algorithm. Although this algorithm is the optimum method in terms of number of operations, it is in need of non-trivial modifications to be efficiently parallelized on NVIDIA GPUs. We proposed several optimizations to accelerate this algorithm on GPU-based architectures, exploring the limitations of both, method and architecture, to be able to solve efficiently a high number of Hines systems (neurons). Each of the optimizations are deeply analyzed and described. To evaluate the impact of the optimizations on real inputs, we have used 6 different morphologies in terms of size and branches. Our studies have proven that the optimizations proposed in the present work can achieve a high performance on those computations with a high number of neurons, being our GPU implementations about 4× and 8× faster than the OpenMP multicore implementation (16 cores), using one and two K80 NVIDIA GPUs respectively. Also, it is important to highlight that these optimizations can continue scaling even when dealing with number of neurons., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 720270 (HBP SGA1), from the Spanish Ministry of Economy and Competitiveness under the project Computación de Altas Prestaciones VII (TIN2015-65316-P) and the Departament d’Innovació, Universitats i Empresa de la Generalitat de Catalunya, under project MPEXPAR: Models de Programació i Entorns d’Execució Paral·lels (2014-SGR-1051). We thank the support of NVIDIA through the BSC/UPC NVIDIA GPU Center of Excellence. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Efficient data sharing on heterogeneous systems

  • García-Flores, Víctor
  • Ayguadé Parra, Eduard|||0000-0002-5146-103X
  • Peña, Antonio J.
General-purpose computing on GPUs has become more accessible due to features such as shared virtual memory and demand paging. Unfortunately it comes at a price, and that is performance. Automatic memory management is convenient but suffers from many drawbacks, preventing heterogeneous systems from achieving their full potential. In this work we analyze the challenges and inefficiencies of demand paging in GPUs, in particular on collaborative computations where data migrates multiple times between host and device. We establish that demand paging on GPUs introduces significant overheads for these kind of computations, and identify the issues of false sharing and unnecessary data transfers derived from the granularity at which data is migrated. In order to alleviate these problems we propose a memory organization and dynamic migration scheme to efficiently share data between host and device at fine granularities and without software intervention. We evaluate our design with a set of collaborative heterogeneous benchmarks and find it achieves 15% lower execution times on average with cache line-sized migrations, but severely degrading performance on benchmarks that access large blocks of contiguous memory. Page-sized migrations, although inefficient, provide on average a 47% execution time reduction with our design over a baseline system implementing demand paging. Our results suggest that cache line-sized migrations are not feasible in systems using a PCI-Express interconnect. In order to understand how future interconnect technologies will impact the feasibility of fine-grained migrations, we evaluate our scheme with various link latencies. We find interconnect latencies four to five times lower than PCI-Express are sufficient to effectively share data at finer granularities., The authors would like to thank Ivan Tanasic and Lluc Alvarez for their insightful suggestions and Javier López Ovalle for his help proofreading the document. This work has been supported by the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P) and by the BSC/UPC NVIDIA GPU Center of Excellence. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Supporting automatic recovery in offloaded distributed programming models through MPI-3 techniques

  • Peña, Antonio J.
  • Beltran Querol, Vicenç
  • Clauss, Carsten
  • Moschny, Thomas
In this paper we describe the design of fault tolerance capabilities for general-purpose offload semantics, based on the OmpSs programming model. Using ParaStation MPI, a production MPI-3.1 implementation, we explore the features that, being standard compliant, an MPI stack must support to provide the necessary fault tolerance guarantees, based on MPI's dynamic process management. Our results, including synthetic benchmarks and applications, reveal low runtime overhead and efficient recovery, demonstrating that the existing MPI standard provided us with sufficient mechanisms to implement an effective and efficient fault-tolerant solution., This research received funding from the European Community’s 7th Framework Programme via the DEEP-ER project under Grant Agreement no. 610476. This work has also been supported by the Spanish Ministry of Science and Innovation (contract TIN2012-34557) and by Generalitat de Catalunya (contracts 2014-SGR-1051 and 2014-SGR-1272). Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. The authors thank Jorge Bell´on, from BSC, for his technical support with the Nanos++ internals., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

cuThomasBatch and cuThomasVBatch, CUDA Routines to compute batch of tridiagonal systems on NVIDIA GPUs

  • Valero Lara, Pedro
  • Martinez Pérez, Ivan
  • Sirvent, Raül
  • Martorell Bofill, Xavier|||0000-0002-0417-3430
  • Peña, Antonio J.
The solving of tridiagonal systems is one of the most computationally expensive parts in many applications, so that multiple studies have explored the use of NVIDIA GPUs to accelerate such computation. However, these studies have mainly focused on using parallel algorithms to compute such systems, which can efficiently exploit the shared memory and are able to saturate the GPUs capacity with a low number of systems, presenting a poor scalability when dealing with a relatively high number of systems. The gtsvStridedBatch routine in the cuSPARSE NVIDIA package is one of these examples, which is used as reference in this article. We propose a new implementation (cuThomasBatch) based on the Thomas algorithm. Unlike other algorithms, the Thomas algorithm is sequential, and so a coarse-grained approach is implemented where one CUDA thread solves a complete tridiagonal system instead of one CUDA block as in gtsvStridedBatch. To achieve a good scalability using this approach, it is necessary to carry out a transformation in the way that the inputs are stored in memory to exploit coalescence (contiguous threads access to contiguous memory locations). Different variants regarding the transformation of the data are explored in detail. We also explore some variants for the case of variable batch, when the size of the systems of the batch has different size (cuThomasVBatch). The results given in this study prove that the implementations carried out in this work are able to beat the reference code, being up to 5× (in double precision) and 6× (in single precision) faster using the latest NVIDIA GPU architecture, the Pascal P100., This project was funded from the European Union's Horizon 2020 research and innovation programme under grant agreement 720270 (HBPSGA1), from the Spanish Ministry of Economy and Competitiveness under the project Computación de Altas Prestaciones VII (TIN2015-65316-P)and the Departament d'Innovació, Universitats i Empresa de la Generalitat de Catalunya, under project MPEXPAR: Models de Programació iEntorns d'Execució Paral·lels (2014-SGR-1051). We thank the support of NVIDIA through the BSC/UPC NVIDIA GPU Center of Excellence andthe valuable feedback provided by Lung Sheng Chien and Alex Fit-Florea. Antonio J. Peña was cofinanced by the Spanish Ministry of Economy andCompetitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

GLTO: On the Adequacy of Lightweight Thread Approaches for OpenMP Implementations

  • Castelló, Adrián
  • Mayo, Rafael
  • Quintana Ortí, Enrique Salvador
  • Seo, Sangmin
  • Balaji, Pavan
  • Peña, Antonio J.
OpenMP is the de facto standard application programming interface (API) for on-node parallelism. The most popular OpenMP runtimes rely on POSIX threads (pthreads) implementations that offer an excellent performance for coarse-grained parallelism and match perfectly with the current hardware. However, a recent trend in runtimes/applications points in the direction of leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. It has been demonstrated that lightweight thread (LWT) solutions are more appropriate for these new parallel paradigms. We have developed GLTO, an OpenMP implementation over the recently-emerged Generic Lightweight Threads (GLT) API. GLT exports a common API for LWT libraries that offers the possibility of running the same application over different native LWT solutions. In this paper we use GLTO to analyze different scenarios where OpenMP implementations may benefit from the use of either LWT or pthreads. Our study reveals that none of the threading approaches obtains the best performance in all the scenarios, but that there are important gaps among them., The Researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO and FEDER, the Generalitat Valenciana fellowship programme Vali+d 2015. Antonio J. Peña is cofinancied by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357. We gratefully acknowledge the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Automating the application data placement in hybrid memory systems

  • Servat, Harald
  • Peña, Antonio J.
  • Llort, German
  • Mercadal, Estanislao
  • Hoppe, Hans-Christian
  • Labarta Mancho, Jesús José
Multi-tiered memory systems, such as those based on Intel® Xeon Phi™processors, are equipped with several memory tiers with different characteristics including, among others, capacity, access latency, bandwidth, energy consumption, and volatility. The proper distribution of the application data objects into the available memory layers is key to shorten the time– to–solution, but the way developers and end-users determine the most appropriate memory tier to place the application data objects has not been properly addressed to date.In this paper we present a novel methodology to build an extensible framework to automatically identify and place the application’s most relevant memory objects into the Intel Xeon Phi fast on-package memory. Our proposal works on top of inproduction binaries by first exploring the application behavior and then substituting the dynamic memory allocations. This makes this proposal valuable even for end-users who do not have the possibility of modifying the application source code. We demonstrate the value of a framework based in our methodology for several relevant HPC applications using different allocation strategies to help end-users improve performance with minimal intervention. The results of our evaluation reveal that our proposal is able to identify the key objects to be promoted into fast on-package memory in order to optimize performance, leading to even surpassing hardware-based solutions., This work has been performed in the Intel-BSC Exascale Lab. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. We would like to thank the Intel’s DCG HEAT team for allowing us to access their computational resources. We also want to acknowledge this team, especially Larry Meadows and Jason Sewall, as well as Pardo Keppel for the productive discussions. We thank Raphaël Léger for allowing us to access the MAXW-DGTD application and its input., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

GLTO: On the Adequacy of Lightweight Thread Approaches for OpenMP Implementations

  • Castelló, Adrián
  • Mayo, Rafael
  • Quintana Ortí, Enrique Salvador
  • Seo, Sangmin
  • Balaji, Pavan
  • Peña, Antonio J.
OpenMP is the de facto standard application programming interface (API) for on-node parallelism. The most popular OpenMP runtimes rely on POSIX threads (pthreads) implementations that offer an excellent performance for coarse-grained parallelism and match perfectly with the current hardware. However, a recent trend in runtimes/applications points in the direction of leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. It has been demonstrated that lightweight thread (LWT) solutions are more appropriate for these new parallel paradigms. We have developed GLTO, an OpenMP implementation over the recently-emerged Generic Lightweight Threads (GLT) API. GLT exports a common API for LWT libraries that offers the possibility of running the same application over different native LWT solutions. In this paper we use GLTO to analyze different scenarios where OpenMP implementations may benefit from the use of either LWT or pthreads. Our study reveals that none of the threading approaches obtains the best performance in all the scenarios, but that there are important gaps among them., The Researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO and FEDER, the Generalitat Valenciana fellowship programme Vali+d 2015. Antonio J. Peña is cofinancied by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357. We gratefully acknowledge the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Supporting automatic recovery in offloaded distributed programming models through MPI-3 techniques

  • Peña, Antonio J.
  • Beltran Querol, Vicenç
  • Clauss, Carsten
  • Moschny, Thomas
In this paper we describe the design of fault tolerance capabilities for general-purpose offload semantics, based on the OmpSs programming model. Using ParaStation MPI, a production MPI-3.1 implementation, we explore the features that, being standard compliant, an MPI stack must support to provide the necessary fault tolerance guarantees, based on MPI's dynamic process management. Our results, including synthetic benchmarks and applications, reveal low runtime overhead and efficient recovery, demonstrating that the existing MPI standard provided us with sufficient mechanisms to implement an effective and efficient fault-tolerant solution., This research received funding from the European Community’s 7th Framework Programme via the DEEP-ER project under Grant Agreement no. 610476. This work has also been supported by the Spanish Ministry of Science and Innovation (contract TIN2012-34557) and by Generalitat de Catalunya (contracts 2014-SGR-1051 and 2014-SGR-1272). Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. The authors thank Jorge Bell´on, from BSC, for his technical support with the Nanos++ internals., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

cuThomasBatch and cuThomasVBatch, CUDA Routines to compute batch of tridiagonal systems on NVIDIA GPUs

  • Valero Lara, Pedro
  • Martinez Pérez, Ivan
  • Sirvent, Raül
  • Martorell Bofill, Xavier
  • Peña, Antonio J.
The solving of tridiagonal systems is one of the most computationally expensive parts in many applications, so that multiple studies have explored the use of NVIDIA GPUs to accelerate such computation. However, these studies have mainly focused on using parallel algorithms to compute such systems, which can efficiently exploit the shared memory and are able to saturate the GPUs capacity with a low number of systems, presenting a poor scalability when dealing with a relatively high number of systems. The gtsvStridedBatch routine in the cuSPARSE NVIDIA package is one of these examples, which is used as reference in this article. We propose a new implementation (cuThomasBatch) based on the Thomas algorithm. Unlike other algorithms, the Thomas algorithm is sequential, and so a coarse-grained approach is implemented where one CUDA thread solves a complete tridiagonal system instead of one CUDA block as in gtsvStridedBatch. To achieve a good scalability using this approach, it is necessary to carry out a transformation in the way that the inputs are stored in memory to exploit coalescence (contiguous threads access to contiguous memory locations). Different variants regarding the transformation of the data are explored in detail. We also explore some variants for the case of variable batch, when the size of the systems of the batch has different size (cuThomasVBatch). The results given in this study prove that the implementations carried out in this work are able to beat the reference code, being up to 5× (in double precision) and 6× (in single precision) faster using the latest NVIDIA GPU architecture, the Pascal P100., This project was funded from the European Union's Horizon 2020 research and innovation programme under grant agreement 720270 (HBPSGA1), from the Spanish Ministry of Economy and Competitiveness under the project Computación de Altas Prestaciones VII (TIN2015-65316-P)and the Departament d'Innovació, Universitats i Empresa de la Generalitat de Catalunya, under project MPEXPAR: Models de Programació iEntorns d'Execució Paral·lels (2014-SGR-1051). We thank the support of NVIDIA through the BSC/UPC NVIDIA GPU Center of Excellence andthe valuable feedback provided by Lung Sheng Chien and Alex Fit-Florea. Antonio J. Peña was cofinanced by the Spanish Ministry of Economy andCompetitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Efficient data sharing on heterogeneous systems

  • García-Flores, Víctor
  • Ayguadé Parra, Eduard
  • Peña, Antonio J.
General-purpose computing on GPUs has become more accessible due to features such as shared virtual memory and demand paging. Unfortunately it comes at a price, and that is performance. Automatic memory management is convenient but suffers from many drawbacks, preventing heterogeneous systems from achieving their full potential. In this work we analyze the challenges and inefficiencies of demand paging in GPUs, in particular on collaborative computations where data migrates multiple times between host and device. We establish that demand paging on GPUs introduces significant overheads for these kind of computations, and identify the issues of false sharing and unnecessary data transfers derived from the granularity at which data is migrated. In order to alleviate these problems we propose a memory organization and dynamic migration scheme to efficiently share data between host and device at fine granularities and without software intervention. We evaluate our design with a set of collaborative heterogeneous benchmarks and find it achieves 15% lower execution times on average with cache line-sized migrations, but severely degrading performance on benchmarks that access large blocks of contiguous memory. Page-sized migrations, although inefficient, provide on average a 47% execution time reduction with our design over a baseline system implementing demand paging. Our results suggest that cache line-sized migrations are not feasible in systems using a PCI-Express interconnect. In order to understand how future interconnect technologies will impact the feasibility of fine-grained migrations, we evaluate our scheme with various link latencies. We find interconnect latencies four to five times lower than PCI-Express are sufficient to effectively share data at finer granularities., The authors would like to thank Ivan Tanasic and Lluc Alvarez for their insightful suggestions and Javier López Ovalle for his help proofreading the document. This work has been supported by the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P) and by the BSC/UPC NVIDIA GPU Center of Excellence. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

cuHinesBatch: solving multiple hines systems on GPUs Human Brain Project

  • Valero-Lara, Pedro
  • Martinez-Perez, Ivan
  • Peña, Antonio J.
  • Martorell Bofill, Xavier
  • Sirvent, Raul
  • Labarta Mancho, Jesús José
The simulation of the behavior of the Human Brain is one of the most important challenges today in computing. The main problem consists of finding efficient ways to manipulate and compute the huge volume of data that this kind of simulations need, using the current technology. In this sense, this work is focused on one of the main steps of such simulation, which consists of computing the Voltage on neurons’ morphology. This is carried out using the Hines Algorithm. Although this algorithm is the optimum method in terms of number of operations, it is in need of non-trivial modifications to be efficiently parallelized on NVIDIA GPUs. We proposed several optimizations to accelerate this algorithm on GPU-based architectures, exploring the limitations of both, method and architecture, to be able to solve efficiently a high number of Hines systems (neurons). Each of the optimizations are deeply analyzed and described. To evaluate the impact of the optimizations on real inputs, we have used 6 different morphologies in terms of size and branches. Our studies have proven that the optimizations proposed in the present work can achieve a high performance on those computations with a high number of neurons, being our GPU implementations about 4× and 8× faster than the OpenMP multicore implementation (16 cores), using one and two K80 NVIDIA GPUs respectively. Also, it is important to highlight that these optimizations can continue scaling even when dealing with number of neurons., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 720270 (HBP SGA1), from the Spanish Ministry of Economy and Competitiveness under the project Computación de Altas Prestaciones VII (TIN2015-65316-P) and the Departament d’Innovació, Universitats i Empresa de la Generalitat de Catalunya, under project MPEXPAR: Models de Programació i Entorns d’Execució Paral·lels (2014-SGR-1051). We thank the support of NVIDIA through the BSC/UPC NVIDIA GPU Center of Excellence. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

On the adequacy of lightweight thread approaches for high-level parallel programming models

  • Castelló, Adrián
  • Mayo Gual, Rafael
  • Sala Penadés, Kevin
  • Beltran Querol, Vicenç
  • Balaji, Pavan
  • Peña, Antonio J.
High-level parallel programming models (PMs) are becoming crucial in order to extract the computational power of current on-node multi-threaded parallelism. The most popular PMs, such as OpenMP or OmpSs, are directive-based: the complexity of the hardware is hidden by the underlying runtime system, improving coding productivity. The implementations of OpenMP usually rely on POSIX threads (pthreads), offering excellent performance for coarse-grained parallelism and a perfect match with the current hardware. OmpSs is a task oriented PM based on an ad hoc runtime solution called Nanos++; it is the precursor of the tasking parallelism in the OpenMP tasking specification. A recent trend in runtimes and applications points to leveraging massive on-node parallelism in conjunction with fine-grained and dynamic scheduling paradigms. In this paper we analyze the behavior of the OpenMP and OmpSs PMs on top of the recently emerged Generic Lightweight Threads (GLT) API. GLT exposes a common API for lightweight thread (LWT) libraries that offers the possibility of running the same application over different native LWT solutions. We describe the design details of those high-level PMs implemented on top of GLT and analyze different scenarios in order to assess where the use of LWTs may benefit application performance. Our work reveals those scenarios where LWTs overperform pthread-based solutions and compares the performance between an ad hoc solution and a generic implementation., The researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO, Spain and FEDER, Spain, the Generalitat Valenciana fellowship programme, Spain Vali+d 2015. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness, Spain under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357. We gratefully acknowledge Enrique S. Quintana-Ortí (Universitat Jaume I) and Sangmin Seo (Samsung Corp.) for their advice in this work and the computing resources provided and operated by the Joint Laboratory for System Evaluation (JLSE) at Argonne National Laboratory., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

GLT: A Unified API for Lightweight Thread Libraries

  • Castelló, Adrián
  • Seo, Sangmin
  • Mayo, Rafael
  • Balaji, Pavan
  • Quintana Ortí, Enrique Salvador
  • Peña, Antonio J.
In recent years, several lightweight thread (LWT) libraries have emerged to tackle exascale challenges. These offer programming models (PMs) based on user-level threads and incorporate their own lightweight mechanisms. However, each library proposes its own PM, exposing different semantics and hindering portability. To address this drawback, we have designed Generic Lightweight Thread (GLT), an application programming interface that frames the functionality of the most popular LWT libraries for high-performance computing under a single PM. We implement GLT on top of Argobots, MassiveThreads, and Qthreads. We provide GLT as a dynamic library, as well as in the form of a static version based on macro preprocessing resolution to reduce overhead. This paper discusses the GLT PM and demonstrates its minimal performance impact., Researchers from the Universitat Jaume I de Castelló were supported by project TIN2014-53495-R of the MINECO, the Generalitat Valenciana fellowship programme Vali+d 2015, and FEDER. Antonio J. Peña is cofinancied by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266. This work was partially supported by the U.S. Dept. of Energy, Office of Science, Office of Advanced Scientific Computing Research (SC-21), under contract DE-AC02-06CH11357., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Integrating memory perspective into the BSC performance tools

  • Servat, Harald
  • Labarta Mancho, Jesús José
  • Hoppe, Hans-Christian
  • Gimenez, Judit
  • Peña, Antonio J.
The growing gap between processor and memory speeds results in complex memory hierarchies as processors evolve to mitigate such differences by taking advantage of locality of reference. In this direction, the BSC performance analysis tools have been recently extended to provide insight relative the application memory accesses depicting their temporal and spatial characteristics, correlating with the source-code and the achieved performance simultaneously. These extensions rely on the Precise Event-Based Sampling (PEBS) mechanism available in recent Intel processors to capture information relative to the application memory accesses. The sampled information is processed with the Folding mechanism to provide a detailed temporal evolution of the memory accesses and in conjunction with the achieved performance and the source-code counterpart. The results obtained from the combination of these tools help application developers to understand better how the application behaves and how the system performs. We demonstrate the value of the complete work-flow by exploring an already optimized state-of-the-art benchmark, providing detailed insight of their memory access behavior., This work has been performed in the Intel-BSC Exascale Lab. We would like to thank Forschungszentrum Julich for the compute time on the Jureca system. Antonio J. Peña is cofinanced by the Spanish Ministry of Economy and Competitiveness under Juan de la Cierva fellowship number IJCI-2015-23266., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Efficient Scalable Computing through Flexible Applications and Adaptive Workloads

  • Iserte, Sergio
  • Mayo, Rafael
  • Quintana Ortí, Enrique Salvador
  • Beltran Querol, Vicenç
  • Peña, Antonio J.
In this paper we introduce a methodology for dynamic job reconfiguration driven by the programming model runtime in collaboration with the global resource manager. We improve the system throughput by exploiting malleability techniques (in terms of number of MPI ranks) through the reallocation of resources assigned to a job during its execution. In our proposal, the OmpSs runtime reconfigures the number of MPI ranks during the execution of an application in cooperation with the Slurm workload manager. In addition, we take advantage of OmpSs offload semantics to allow application developers deal with data redistribution. By combining these elements a job is able to expand itself in order to exploit idle nodes or be shrunk if other queued jobs could be initiated. This novel approach adapts the system workload in order to increase the throughput as well as make a smarter use of the underlying resources. Our experiments demonstrate that this approach can reduce the total execution time of a practical workload by more than 40% while reducing the amount of resources by 30%., This work is supported by the Project TIN2014-53495-R and TIN2015-65316-P from MINECO and FEDER. Antonio J. Peña is cofinanced by MINECO under Juan de la Cierva fellowship number IJCI-2015-23266. Special thanks to José I. Aliaga for the conjugate gradient code., Peer Reviewed, Postprint (author's final draft)