Task-based programming models have enabled the optimized execution of the computation workloads of applications. These programming models can take advantage of large-scale distributed infrastructures by allowing the parallel and distributed execution of applications in high-level work components called tasks. Nevertheless, in the era of Big Data and Exascale, the amount of data produced by modern scientific applications has already surpassed terabytes and is rapidly increasing. Hence, I/O performance became the bottleneck to overcome in order to achieve more total performance improvement. New storage technologies offer higher bandwidth and faster solutions than traditional Parallel File Systems (PFS). Such storage devices are deployed in modern day infrastructures to boost I/O performance by offering a fast layer that absorbs the generated data. Therefore, it is necessary for any programming model targeting more performance to manage this heterogeneity and take advantage of it to improve the I/O performance of applications. Towards this goal, we propose in this paper a set of programming model capabilities that we refer to as Storage-Heterogeneity Awareness. Such capabilities include: (i) abstracting the heterogeneity of storage systems, and (ii) optimizing I/O performance by supporting dedicated I/O schedulers and an automatic data flushing technique. The evaluation section of this paper presents the performance results of different applications on the MareNostrum CTE-Power heterogeneous storage cluster. Our experiments demonstrate that a storage-heterogeneity aware programming model can achieve up to almost 5x I/O performance speedup and 48% total time improvement compared to the reference PFS-based usage of the execution infrastructure. ; This work is partially supported by the European Union through the Horizon 2020 research and innovation programme under contracts 721865 (EXPERTISE Project) by the Spanish Government (PID2019-107255GB) and the Generalitat de Catalunya (contract 2014-SGR-1051). ; Peer ...
Task-based programming models offer a flexible way to express the unstructured parallelism patterns of nowadays complex applications. This expressive capability is required to achieve maximum possible performance for applications that are executed in distributed execution platforms. In current task-based workflows, tasks are launched for execution when their data dependencies are satisfied. However, even though the data dependencies of a certain task might have been already produced, the execution of this task will be delayed until its predecessor tasks completely finish their execution. As a consequence of this approach of releasing dependencies, the amount of parallelism inherent in applications is limited and performance improvement opportunities are wasted. To mitigate this limitation, we propose an eager approach for releasing data dependencies. Following this approach, the execution of tasks will not be delayed until their predecessor tasks completely finish their execution, instead, tasks will be launched for execution as soon as their data requirements are available. Hence, more parallelism is exposed and applications can achieve higher levels of performance by overlapping the execution of tasks. Towards achieving this goal, in this paper we propose applying two changes to task-based workflow systems. First, modifying the dependency relationships of tasks to be specified not only in terms of predecessor and successor tasks but also in terms of the data that caused these dependencies. Second, triggering the release of dependencies as soon as a predecessor task generates the output data instead of having to wait until the end of the predecessor execution to release all of its dependencies. We realize this proposal using PyCOMPSs: a task-based programming model for parallelizing Python applications. Our experiments show that using an eager approach for releasing dependencies achieves more than 50% performance improvement in the total execution time as compared to the default approach of releasing dependencies. ; This work is partially supported by the European Union through the Horizon 2020 research and innovation programme under contracts 721865 (EXPERTISE Project) by the Spanish Government (SEV2015-0493,TIN2015-65316-P) and the Generalitat de Catalunya (contract 2014-SGR1051). ; Peer Reviewed ; Postprint (author's final draft)
The last improvements in programming languages and models have focused on simplicity and abstraction; leading Python to the top of the list of the programming languages. However, there is still room for improvement when preventing users from dealing directly with distributed and parallel computing issues. This paper proposes and evaluates AutoParallel, a Python module to automatically find an appropriate task-based parallelisation of affine loop nests and execute them in parallel in a distributed computing infrastructure. It is based on sequential programming and contains one single annotation (in the form of a Python decorator) so that anyone with intermediate-level programming skills can scale up an application to hundreds of cores. The evaluation demonstrates that AutoParallel goes one step further in easing the development of distributed applications. On the one hand, the programmability evaluation highlights the benefits of using a single Python decorator instead of manually annotating each task and its parameters or, even worse, having to develop the parallel code explicitly (e.g., using OpenMP, MPI). On the other hand, the performance evaluation demonstrates that AutoParallel is capable of automatically generating task-based workflows from sequential Python code while achieving the same performances than manually taskified versions of established state-of-the-art algorithms (i.e., Cholesky, LU, and QR decompositions). Finally, AutoParallel is also capable of automatically building data blocks to increase the tasks' granularity; freeing the user from creating the data chunks, and re-designing the algorithm. For advanced users, we believe that this feature can be useful as a baseline to design blocked algorithms. ; This work has been supported by the Spanish Government through contracts SEV2015-0493 and TIN2015-65316-P, and by Generalitat de Catalunya through contract 2014-SGR-1051. Cristian Ramon-Cortes predoctoral contract is financed by the Ministry of Economy and Competitiveness under the contract BES-2016-076791. ; Peer Reviewed ; Postprint (author's final draft)
Workflow systems promise scientists an automated end-to-end path from hypothesis to discovery. However, expecting any single workflow system to deliver such a wide range of capabilities is impractical. A more practical solution is to compose the end-to-end workflow from more than one system. With this goal in mind, the integration of task-based and in situ workflows is explored, where the result is a hierarchical heterogeneous workflow composed of subworkflows, with different levels of the hierarchy using different programming, execution, and data models. Materials science use cases demonstrate the advantages of such heterogeneous hierarchical workflow composition. ; This work is a collaboration between Argonne National Laboratory and the Barcelona Supercomputing Center within the Joint Laboratory for Extreme-Scale Computing. This research is supported by the U.S. Department of Energy, Office of Science, Office of Advanced Scientific Computing Research, under contract number DE-AC02- 06CH11357, program manager Laura Biven, and by the Spanish Government (SEV2015-0493), by the Spanish Ministry of Science and Innovation (contract TIN2015-65316-P), by Generalitat de Catalunya (contract 2014-SGR-1051). ; Peer Reviewed ; Postprint (author's final draft)
With the advent of distributed computing, the need for frameworks that facilitate its programming and management has also appeared. These tools have typically been used to support the research on application areas that require them. This poses good initial conditions for translational computer science (TCS), although this does not always occur. This article describes our experience with the PyCOMPSs project, a programming model for distributed computing. While it is a research instrument for our team, it has also been applied in multiple real use cases under the umbrella of European Funded projects, or as part of internal projects between various departments at the Barcelona Supercomputing Center. This article illustrates how the authors have engaged in TCS as an underlying research methodology, collecting experiences from three European projects. ; This work was supported in part by Spanish Government under Contract TIN2015-65316-P, in part by the Generalitat de Catalunya under Contract 2014-SGR-1051, and in part by the European Commission's Horizon 2020 Framework program through BioExcel Center of Excellence under Contract 823830 and Contract 675728, in part by the ExaQUte Project under Contract 800898, in part by the European High-Performance Computing Joint Undertaking (JU) under Grant 955558, in part by the MCIN/AEI/10.13039/501100011033, and in part by the European Union NextGenerationEU/PRTR. ; Peer Reviewed ; Postprint (author's final draft)
Genome-wide association studies (GWAS) are not fully comprehensive, as current strategies typically test only the additive model, exclude the X chromosome, and use only one reference panel for genotype imputation. We implement an extensive GWAS strategy, GUIDANCE, which improves genotype imputation by using multiple reference panels and includes the analysis of the X chromosome and non-additive models to test for association. We apply this methodology to 62,281 subjects across 22 age-related diseases and identify 94 genome-wide associated loci, including 26 previously unreported. Moreover, we observe that 27.7% of the 94 loci are missed if we use standard imputation strategies with a single reference panel, such as HRC, and only test the additive model. Among the new findings, we identify three novel low-frequency recessive variants with odds ratios larger than 4, which need at least a three-fold larger sample size to be detected under the additive model. This study highlights the benefits of applying innovative strategies to better uncover the genetic architecture of complex diseases. ; This work has been sponsored by the grant SEV-2011-00067 and SEV2015-0493 of Severo Ochoa Program, awarded by the Spanish Government, by the grant TIN2015- 65316-P, awarded by the Spanish Ministry of Science and Innovation, and by the Generalitat de Catalunya (contract 2014-SGR-1051). This work was supported by an EFSD/Lilly research fellowship. Josep M. Mercader was supported by a Sara Borrell Fellowship from the Instituto Carlos III, Beatriu de Pinós fellowship from the Agency for Management of University and Research Grants (AGAUR) and by the American Diabetes Association Innovative and Clinical Translational Award 1-19-ICTS-068. Sílvia Bonàs was supported by FI-DGR Fellowship from FIDGR 2013 from Agència de Gestió d'Ajuts Universitaris i de Recerca (AGAUR, Generalitat de Catalunya), and a 'Juan de la Cierva' postdoctoral fellowship (MINECO;FJCI-2017-32090). Cecilia Salvoro received funding from the European Union's Horizon 2020 research and innovation program under the Marie Skłodowska-Curie grant agreement H2020-MSCA-COFUND-2016- 754433. Cristian Ramon-Cortes pre-doctoral contract is financed by the Spanish Ministry of Science, Innovation, and Universities under contract BES-2016-076791. Elizabeth G. Atkinson was supported by the National Institutes of Mental Health (grants K01MH121659 and T32MH017119). Jose Florez was supported by NIH/NIDDK award K24 DK110550. This study made use of data generated by the UK10K Consortium, derived from samples from UK10K COHORT IMPUTATION (EGAS00001000713). A full list of the investigators who contributed to the generation of the data is available at www.UK10K.org. Funding for UK10K was provided by the Wellcome Trust under award WT091310. This study made use of data generated by the 'Genome of the Netherlands' project, which is funded by the Netherlands Organization for Scientific Research (grant no. 184021007). The data were made available as a Rainbow Project of BBMRI-NL. Samples were contributed by LifeLines (http://lifelines.nl/lifelines-research/general), the Leiden Longevity Study (http://www.healthy-ageing.nl; http://www.langleven.net), the Netherlands Twin Registry (NTR: http://www.tweelingenregister.org), the Rotterdam studies (http://www.erasmus-epidemiology.nl/rotterdamstudy) and the Genetic Research in Isolated Populations program (http://www.epib.nl/research/geneticepi/research. html#gip). The sequencing was carried out in collaboration with the Beijing Institute for Genomics (BGI). This study also made use of data generated by The Haplotype Reference Consortium (HRC) accessed through The European Genome-phenome Archive at the European Bioinformatics Institute with the accession numbers EGAD00001002729, after a form agreed by the Barcelona Supercomputing Center (BSC) with WTSI. This research has been conducted using also the UK Biobank Resource (application number 31063 and 27892). The Genotype-Tissue Expression (GTEx) Project was supported by the Common Fund of the Office of the Director of the National Institutes of Health, and by NCI, NHGRI, NHLBI, NIDA, NIMH, and NINDS. The data used for the analyses described in this manuscript were obtained from the GTEx Portal on 07/16/2019. We acknowledge PRACE for awarding us access to both MareNostrum supercomputer from the Barcelona Supercomputing Center, based in Spain at Barcelona, and the SuperMUC supercomputer of the Leibniz Supercomputing Center (LRZ), based in Garching at Germany (proposals numbers 2016143358 and 2016163985). The technical support group from the Barcelona Supercomputing Center is gratefully acknowledged. Finally, we thank all the Computational Genomics group at the BSC for their helpful discussions and valuable comments on the manuscript. We also acknowledge Elias Rodriguez Fos for designing the GUIDANCE logo. ; Peer Reviewed ; Article signat per 22 autors/autores: Marta Guindo-Martínez 1,18; Ramon Amela 1,18; Silvia Bonàs-Guarch 1,2,3; Montserrat Puiggròs 1; Cecilia Salvoro 1; Irene Miguel-Escalada 1,2,3; Caitlin E. Carey 4,5; Joanne B. Cole 6,7,8,9; Sina Rüeger 10; Elizabeth Atkinson 4,5,11; Aaron Leong 8,12; Friman Sanchez 1; Cristian Ramon-Cortes 1; Jorge Ejarque 1; Duncan S. Palmer 4,5,17; Mitja Kurki 10; FinnGen Consortium*, Krishna Aragam 11,13,14; Jose C. Florez 6,7,15; Rosa M. Badia 1; Josep M. Mercader 1,6,7,15,19✉ & David Torrents 1,16,19✉ *A full list of members and their affiliations appears in the Supplementary Information 1 Barcelona Supercomputing Center (BSC), Barcelona, Spain. 2 Regulatory Genomics and Diabetes, Centre for Genomic Regulation, The Barcelona Institute of Science and Technology, Barcelona, Spain. 3 CIBER de Diabetes y Enfermedades Metabólicas Asociadas, Madrid, Spain. 4 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard, Cambridge, MA, USA. 5 Analytic and Translational Genetics Unit, Department of Medicine, Massachusetts General Hospital, Boston, MA, USA. 6 Programs in Metabolism and Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA. 7 Diabetes Unit and Center for Genomic Medicine, Massachusetts General Hospital, Boston, MA, USA. 8 Harvard Medical School, Boston, MA, USA. 9 Division of Endocrinology and Center for Basic and Translational Obesity Research, Boston Children's Hospital, Boston, MA, USA. 10 Institute for Molecular Medicine Finland, FIMM, HiLIFE, University of Helsinki, Helsinki, Finland. 11 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard, Cambridge, MA, USA. 12 Department of Medicine, Massachusetts General Hospital, Boston, MA, USA. 13 Cardiology Division, Massachusetts General Hospital, Boston, MA, USA. 14 Cardiovascular Research Center, Massachusetts General Hospital, Boston, MA, USA. 15 Department of Medicine, Harvard Medical School, Boston, MA, USA. 16 Institució Catalana de Recerca i Estudis Avançats (ICREA), Barcelona, Spain. 17 Present address: GENOMICS plc, Oxford, UK. 18 These authors contributed equally: Marta Guindo-Martínez, Ramon Amela. 19 These authors jointly supervised this work: Josep M. Mercader, David Torrents. ; Postprint (published version)
