The growing gap between processor and memory speeds has lead to complex memory hierarchies as pro... more The growing gap between processor and memory speeds has lead to complex memory hierarchies as processors evolve to mitigate such divergence by exploiting the locality of reference. In this direction, the BSC performance analysis tools have been recently extended to provide insight into the application memory accesses by 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 regarding the application memory accesses. The sampled information is later combined with the Folding technique to represent a detailed temporal evolution of the memory accesses and in conjunction with the achieved performance and the source-code counterpart. The reports generated by the latter tool help not only application developers but also processor architects to understand better how the application behaves and how the system performs. In this paper, we describe a tighter integration of the sampling mechanism into the monitoring package. We also demonstrate the value of the complete workflow by exploring already optimized state-of-the-art benchmarks, providing detailed insight of their memory access behavior. We have taken advantage of this insight to apply small modifications that improve the applications' performance.
Python is progressively consolidating itself within the HPC community with its simple syntax, lar... more Python is progressively consolidating itself within the HPC community with its simple syntax, large standard library, and powerful third-party libraries for scientific computing that are especially attractive to domain scientists. Despite Python lowering the bar for accessing parallel computing, utilizing the capacities of HPC systems efficiently remains a challenging task, after all. Yet, at the moment only few supporting tools exist and provide merely basic information in the form of summarized profile data. In this paper, we present our efforts in developing eventbased tracing support for Python within the performance monitor Extrae to provide detailed information and enable a profound performance analysis. We present concepts to record the complete communication behavior as well as to capture entry and exit of functions in Python to provide the according application context. We evaluate our implementation in Extrae by analyzing the well-established electronic structure simulation package GPAW and demonstrate that the recorded traces provide equivalent information as for traditional C or Fortran applications and, therefore, offering the same profound analysis capabilities now for Python, as well.
2022 International Joint Conference on Neural Networks (IJCNN)
The majority of density-based clustering algorithms can not perform properly when data expose ver... more The majority of density-based clustering algorithms can not perform properly when data expose very different density through the feature space. These algorithms implicitly presume that all clusters almost have the same density, therefore, they normally use global parameters. Consequently, they are often biased towards finding dense clusters in front of sparse ones. In this paper, we propose a parametric multilinear transformation method to homogenize cluster densities while preserving the topological structure of the dataset. The transformed clusters have approximately the same density while all inter-cluster regions become globally low-density. In our method, the feature space is locally bent by dense data point concentrations the same way as stars bend the space-time dimensions in Theory of Relativity. We present a new Gravitational Self-organization Map to model the feature space curvature by plugging the concepts of gravity and fabric of space into the Self-organization Map algorithm to mathematically describe the density structure of the data. To homogenize the cluster density, we introduce a novel mapping mechanism to project the data from a non-Euclidean curved space to a new Euclidean flat space. Specifically, this mechanism transfers the basis vectors instead of the feature vectors to guarantee the continuity of the mapping function and optimize the computation cost of the algorithm. As a result, our method can efficiently and explicitly homogenize the density of any dataset globally to then apply existing clustering algorithms without modification. Our experimental results over both realworld and synthetic datasets show that our approach outperforms the current statistical-based methods.
Heterogeneous systems are gaining more importance in supercomputing, yet they are challenging to ... more Heterogeneous systems are gaining more importance in supercomputing, yet they are challenging to program and developers require support tools to understand how well their accelerated codes perform and how they can be improved. The OpenMP Tools Interface (OMPT) is a new performance monitoring interface that is being considered for integration into the OpenMP standard. OMPT allows monitoring the execution of heterogeneous OpenMP applications by revealing the activity of the runtime through a standardized API as well as facilitating the exchange of performance information between devices with accelerated codes, and the analysis tool. In this paper we describe our efforts implementing parts of the OMPT specification necessary to monitor accelerators. In particular, the integration of the OMPT features to our parallel runtime system and instrumentation framework helps to obtain detailed performance information about the execution of the accelerated tasks issued to the devices to allow an insightful analysis. As a result of this analysis, the parallel runtime of the programming model has been improved. We focus on the evaluation of monitoring FPGA devices studying the performance of a common kernel in scientific algorithms: matrix multiplication. Nonetheless, this development is as well applicable to monitor GPU accelerators and Intel R Xeon Phi TM co-processors operating under the OmpSs programming model. 1 Introduction High performance computing (HPC) systems provide tremendous computational resources, however, nowadays, the gain in performance comes with a gain in complexity as well. One of the contributing factors is the growing heterogeneity
To maximise the scientific output of a high-performance computing system, different stakeholders ... more To maximise the scientific output of a high-performance computing system, different stakeholders pursue different strategies. While individual application developers are trying to shorten the time to solution by optimising their codes, system administrators are tuning the configuration of the overall system to increase its throughput. Yet, the complexity of today's machines with their strong interrelationship between application and system performance presents serious challenges to achieving these goals. The HOPSA project (HOlistic Performance System Analysis) therefore sets out to create an integrated diagnostic infrastructure for combined application and system-level tuning-with the former provided by the EU and the latter by the Russian project partners. Starting from system-wide basic performance screening of individual jobs, an automated workflow routes findings on potential bottlenecks either to application developers or system administrators with recommendations on how to identify their root cause using more powerful diagnostic tools. Developers can choose from a variety of mature performance-analysis
2020 19th International Symposium on Parallel and Distributed Computing (ISPDC)
Hybrid parallelization may be the only path for most codes to use HPC systems on a very large sca... more Hybrid parallelization may be the only path for most codes to use HPC systems on a very large scale. Even within a small scale, with an increasing number of cores per node, combining MPI with some shared memory thread-based library allows to reduce the application network requirements. Despite the benefits of a hybrid approach, it is not easy to achieve an efficient hybrid execution. This is not only because of the added complexity of combining two different programming models, but also because in many cases the code was initially designed with just one level of parallelization and later extended to a hybrid mode. This paper presents our model to diagnose the efficiency of hybrid applications, distinguishing the contribution of each parallel programming paradigm. The flexibility of the proposed methodology allows us to use it for different paradigms and scenarios, like comparing the MPI+OpenMP and MPI+CUDA versions of the same code.
All in-text references underlined in blue are linked to publications on ResearchGate, letting you... more All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
Performance analysis tools are essential in the process of understanding application behavior, id... more Performance analysis tools are essential in the process of understanding application behavior, identifying critical performance issues and adapting applications to new architectures and increasingly scaling HPC systems. State-of-the-art tools provide extensive functionality and a plenitude of specialized analysis capabilities. At the same time, the complexity of the potential performance issues and sometimes the tools themselves remains a challenging task, especially for non-experts. In particular, identifying the main issues in the overwhelming amount of data and tool opportunities as well as quantifying their impact and potential for improvement can be tedious and time consuming. In this paper we present a structured approach to performance analysis used within the EU Centre of Excellence for Performance Optimization and Productivity (POP). The structured approach features a method to get a general overview, determine the focus of the analysis, and identify the main issues and areas for potential improvement with a statistical performance model that leads to starting points for a subsequent in-depth analysis. All steps of the structured approach are accompanied with according tools from the BSC tool suite and underlined with an exemplary performance analysis.
Feature selection techniques are enormously applied in a variety of data analysis tasks in order ... more Feature selection techniques are enormously applied in a variety of data analysis tasks in order to reduce the dimensionality. According to the type of learning, feature selection algorithms are categorized to: supervised or unsupervised. In unsupervised learning scenarios, selecting features is a much harder problem, due to the lack of class labels that would facilitate the search for relevant features. The selecting feature difficulty is amplified when the data is corrupted by different noises. Almost all traditional unsupervised feature selection methods are not robust against the noise in samples. These approaches do not have any explicit mechanism for detaching and isolating the noise thus they can not produce an optimal feature subset. In this article, we propose an unsupervised approach for feature selection on noisy data, called Robust Independent Feature Selection (RIFS). Specifically, we choose feature subset that contains most of the underlying information, using the same criteria as the Independent component analysis (ICA). Simultaneously, the noise is separated as an independent component. The isolation of representative noise samples is achieved using factor oblique rotation whereas noise identification is performed using factor pattern loadings. Extensive experimental results over divers real-life data sets have showed the efficiency and advantage of the proposed algorithm.
2017 46th International Conference on Parallel Processing Workshops (ICPPW), Aug 1, 2017
In this paper, we address the decreasing performance of the FFTXlib, the Fast Fourier Transformat... more In this paper, we address the decreasing performance of the FFTXlib, the Fast Fourier Transformation (FFT) kernel of Quantum ESPRESSO, when scaling to a full KNL node. An increased performance in the FFTXlib will likewise increase the performance of the entire Quantum ESPRESSO code one of the most used plane-wave DFT codes in the community of material science. Our approach focuses on, first, overlapping computation and communication and, second, decreasing resource contention for higher compute efficiency. In order to achieve this we use the OmpSs programming model based on task dependencies. We allow overlapping of computation and communication by converting all steps of the FFT into tasks following a flow dependency. In the same way, we decrease resource contention by converting each FFT into an individual task that can be scheduled asynchronously. In both cases, multiple FFTs can be computed in parallel. The task-based optimizations are implemented in the FFTXlib and show up to 10 % runtime reduction on the already highly optimized version. Since the task scheduling is done dynamically during execution by the parallel runtime, not statically by the user, it also frees the user from finding the ideal parallel configuration himself.
Abstract. Knowing the performance of an application in a Grid environment is an important issue i... more Abstract. Knowing the performance of an application in a Grid environment is an important issue in application development and for scheduling decisions. In this paper we describe the analysis and optimisation of a computation- and communication-intensive application from the field of bioinformatics, which was demonstrated at the HPC-Challenge of Supercomputing 2002 at Baltimore. This application has been adapted to be run on an heterogeneous computational Grid by means of PACX-MPI. The analysis and optimisation is based on trace driven tools, mainly Dimemas and Vampir. All these methodologies and tools are being extended in the frame of the DAMIEN IST project. 1
Life Science is developing into one of the largest e- Infrastructure users in Europe, in part due... more Life Science is developing into one of the largest e- Infrastructure users in Europe, in part due to the ever-growing amount of biological data. Modern drug design typically includes both sequence bioinformatics, in silico virtual screening, and free energy calculations, e.g. of drug binding. This development will accelerate tremendously, and puts high demands on simulation software and support services. e-Infrastructure projects such as PRACE/DEISA have made important advances on hardware and scalability, but have largely been focused on theoretical scalability for large systems, while typical life science applications rather concern small-to-medium size molecules. Here, we propose to address this with by implementing new techniques for efficient small-system parallelization combined with throughput and ensemble computing to enable the life science community to exploit the largest next-generation e-Infrastructures. We will also build a new cross-disciplinary Competence Network for ...
The growing gap between processor and memory speeds results in complex memory hierarchies as proc... more 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 s...
The growing gap between processor and memory speeds has lead to complex memory hierarchies as pro... more The growing gap between processor and memory speeds has lead to complex memory hierarchies as processors evolve to mitigate such divergence by exploiting the locality of reference. In this direction, the BSC performance analysis tools have been recently extended to provide insight into the application memory accesses by 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 regarding the application memory accesses. The sampled information is later combined with the Folding technique to represent a detailed temporal evolution of the memory accesses and in conjunction with the achieved performance and the source-code counterpart. The reports generated by the latter tool help not only application developers but also processor architects to understand better how the application behaves and how the system performs. In this paper, we describe a tighter integration of the sampling mechanism into the monitoring package. We also demonstrate the value of the complete workflow by exploring already optimized state-of-the-art benchmarks, providing detailed insight of their memory access behavior. We have taken advantage of this insight to apply small modifications that improve the applications' performance.
Python is progressively consolidating itself within the HPC community with its simple syntax, lar... more Python is progressively consolidating itself within the HPC community with its simple syntax, large standard library, and powerful third-party libraries for scientific computing that are especially attractive to domain scientists. Despite Python lowering the bar for accessing parallel computing, utilizing the capacities of HPC systems efficiently remains a challenging task, after all. Yet, at the moment only few supporting tools exist and provide merely basic information in the form of summarized profile data. In this paper, we present our efforts in developing eventbased tracing support for Python within the performance monitor Extrae to provide detailed information and enable a profound performance analysis. We present concepts to record the complete communication behavior as well as to capture entry and exit of functions in Python to provide the according application context. We evaluate our implementation in Extrae by analyzing the well-established electronic structure simulation package GPAW and demonstrate that the recorded traces provide equivalent information as for traditional C or Fortran applications and, therefore, offering the same profound analysis capabilities now for Python, as well.
2022 International Joint Conference on Neural Networks (IJCNN)
The majority of density-based clustering algorithms can not perform properly when data expose ver... more The majority of density-based clustering algorithms can not perform properly when data expose very different density through the feature space. These algorithms implicitly presume that all clusters almost have the same density, therefore, they normally use global parameters. Consequently, they are often biased towards finding dense clusters in front of sparse ones. In this paper, we propose a parametric multilinear transformation method to homogenize cluster densities while preserving the topological structure of the dataset. The transformed clusters have approximately the same density while all inter-cluster regions become globally low-density. In our method, the feature space is locally bent by dense data point concentrations the same way as stars bend the space-time dimensions in Theory of Relativity. We present a new Gravitational Self-organization Map to model the feature space curvature by plugging the concepts of gravity and fabric of space into the Self-organization Map algorithm to mathematically describe the density structure of the data. To homogenize the cluster density, we introduce a novel mapping mechanism to project the data from a non-Euclidean curved space to a new Euclidean flat space. Specifically, this mechanism transfers the basis vectors instead of the feature vectors to guarantee the continuity of the mapping function and optimize the computation cost of the algorithm. As a result, our method can efficiently and explicitly homogenize the density of any dataset globally to then apply existing clustering algorithms without modification. Our experimental results over both realworld and synthetic datasets show that our approach outperforms the current statistical-based methods.
Heterogeneous systems are gaining more importance in supercomputing, yet they are challenging to ... more Heterogeneous systems are gaining more importance in supercomputing, yet they are challenging to program and developers require support tools to understand how well their accelerated codes perform and how they can be improved. The OpenMP Tools Interface (OMPT) is a new performance monitoring interface that is being considered for integration into the OpenMP standard. OMPT allows monitoring the execution of heterogeneous OpenMP applications by revealing the activity of the runtime through a standardized API as well as facilitating the exchange of performance information between devices with accelerated codes, and the analysis tool. In this paper we describe our efforts implementing parts of the OMPT specification necessary to monitor accelerators. In particular, the integration of the OMPT features to our parallel runtime system and instrumentation framework helps to obtain detailed performance information about the execution of the accelerated tasks issued to the devices to allow an insightful analysis. As a result of this analysis, the parallel runtime of the programming model has been improved. We focus on the evaluation of monitoring FPGA devices studying the performance of a common kernel in scientific algorithms: matrix multiplication. Nonetheless, this development is as well applicable to monitor GPU accelerators and Intel R Xeon Phi TM co-processors operating under the OmpSs programming model. 1 Introduction High performance computing (HPC) systems provide tremendous computational resources, however, nowadays, the gain in performance comes with a gain in complexity as well. One of the contributing factors is the growing heterogeneity
To maximise the scientific output of a high-performance computing system, different stakeholders ... more To maximise the scientific output of a high-performance computing system, different stakeholders pursue different strategies. While individual application developers are trying to shorten the time to solution by optimising their codes, system administrators are tuning the configuration of the overall system to increase its throughput. Yet, the complexity of today's machines with their strong interrelationship between application and system performance presents serious challenges to achieving these goals. The HOPSA project (HOlistic Performance System Analysis) therefore sets out to create an integrated diagnostic infrastructure for combined application and system-level tuning-with the former provided by the EU and the latter by the Russian project partners. Starting from system-wide basic performance screening of individual jobs, an automated workflow routes findings on potential bottlenecks either to application developers or system administrators with recommendations on how to identify their root cause using more powerful diagnostic tools. Developers can choose from a variety of mature performance-analysis
2020 19th International Symposium on Parallel and Distributed Computing (ISPDC)
Hybrid parallelization may be the only path for most codes to use HPC systems on a very large sca... more Hybrid parallelization may be the only path for most codes to use HPC systems on a very large scale. Even within a small scale, with an increasing number of cores per node, combining MPI with some shared memory thread-based library allows to reduce the application network requirements. Despite the benefits of a hybrid approach, it is not easy to achieve an efficient hybrid execution. This is not only because of the added complexity of combining two different programming models, but also because in many cases the code was initially designed with just one level of parallelization and later extended to a hybrid mode. This paper presents our model to diagnose the efficiency of hybrid applications, distinguishing the contribution of each parallel programming paradigm. The flexibility of the proposed methodology allows us to use it for different paradigms and scenarios, like comparing the MPI+OpenMP and MPI+CUDA versions of the same code.
All in-text references underlined in blue are linked to publications on ResearchGate, letting you... more All in-text references underlined in blue are linked to publications on ResearchGate, letting you access and read them immediately.
Performance analysis tools are essential in the process of understanding application behavior, id... more Performance analysis tools are essential in the process of understanding application behavior, identifying critical performance issues and adapting applications to new architectures and increasingly scaling HPC systems. State-of-the-art tools provide extensive functionality and a plenitude of specialized analysis capabilities. At the same time, the complexity of the potential performance issues and sometimes the tools themselves remains a challenging task, especially for non-experts. In particular, identifying the main issues in the overwhelming amount of data and tool opportunities as well as quantifying their impact and potential for improvement can be tedious and time consuming. In this paper we present a structured approach to performance analysis used within the EU Centre of Excellence for Performance Optimization and Productivity (POP). The structured approach features a method to get a general overview, determine the focus of the analysis, and identify the main issues and areas for potential improvement with a statistical performance model that leads to starting points for a subsequent in-depth analysis. All steps of the structured approach are accompanied with according tools from the BSC tool suite and underlined with an exemplary performance analysis.
Feature selection techniques are enormously applied in a variety of data analysis tasks in order ... more Feature selection techniques are enormously applied in a variety of data analysis tasks in order to reduce the dimensionality. According to the type of learning, feature selection algorithms are categorized to: supervised or unsupervised. In unsupervised learning scenarios, selecting features is a much harder problem, due to the lack of class labels that would facilitate the search for relevant features. The selecting feature difficulty is amplified when the data is corrupted by different noises. Almost all traditional unsupervised feature selection methods are not robust against the noise in samples. These approaches do not have any explicit mechanism for detaching and isolating the noise thus they can not produce an optimal feature subset. In this article, we propose an unsupervised approach for feature selection on noisy data, called Robust Independent Feature Selection (RIFS). Specifically, we choose feature subset that contains most of the underlying information, using the same criteria as the Independent component analysis (ICA). Simultaneously, the noise is separated as an independent component. The isolation of representative noise samples is achieved using factor oblique rotation whereas noise identification is performed using factor pattern loadings. Extensive experimental results over divers real-life data sets have showed the efficiency and advantage of the proposed algorithm.
2017 46th International Conference on Parallel Processing Workshops (ICPPW), Aug 1, 2017
In this paper, we address the decreasing performance of the FFTXlib, the Fast Fourier Transformat... more In this paper, we address the decreasing performance of the FFTXlib, the Fast Fourier Transformation (FFT) kernel of Quantum ESPRESSO, when scaling to a full KNL node. An increased performance in the FFTXlib will likewise increase the performance of the entire Quantum ESPRESSO code one of the most used plane-wave DFT codes in the community of material science. Our approach focuses on, first, overlapping computation and communication and, second, decreasing resource contention for higher compute efficiency. In order to achieve this we use the OmpSs programming model based on task dependencies. We allow overlapping of computation and communication by converting all steps of the FFT into tasks following a flow dependency. In the same way, we decrease resource contention by converting each FFT into an individual task that can be scheduled asynchronously. In both cases, multiple FFTs can be computed in parallel. The task-based optimizations are implemented in the FFTXlib and show up to 10 % runtime reduction on the already highly optimized version. Since the task scheduling is done dynamically during execution by the parallel runtime, not statically by the user, it also frees the user from finding the ideal parallel configuration himself.
Abstract. Knowing the performance of an application in a Grid environment is an important issue i... more Abstract. Knowing the performance of an application in a Grid environment is an important issue in application development and for scheduling decisions. In this paper we describe the analysis and optimisation of a computation- and communication-intensive application from the field of bioinformatics, which was demonstrated at the HPC-Challenge of Supercomputing 2002 at Baltimore. This application has been adapted to be run on an heterogeneous computational Grid by means of PACX-MPI. The analysis and optimisation is based on trace driven tools, mainly Dimemas and Vampir. All these methodologies and tools are being extended in the frame of the DAMIEN IST project. 1
Life Science is developing into one of the largest e- Infrastructure users in Europe, in part due... more Life Science is developing into one of the largest e- Infrastructure users in Europe, in part due to the ever-growing amount of biological data. Modern drug design typically includes both sequence bioinformatics, in silico virtual screening, and free energy calculations, e.g. of drug binding. This development will accelerate tremendously, and puts high demands on simulation software and support services. e-Infrastructure projects such as PRACE/DEISA have made important advances on hardware and scalability, but have largely been focused on theoretical scalability for large systems, while typical life science applications rather concern small-to-medium size molecules. Here, we propose to address this with by implementing new techniques for efficient small-system parallelization combined with throughput and ensemble computing to enable the life science community to exploit the largest next-generation e-Infrastructures. We will also build a new cross-disciplinary Competence Network for ...
The growing gap between processor and memory speeds results in complex memory hierarchies as proc... more 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 s...
Uploads
Papers by Judit Gimenez