Milestone 6.3: Basic Analysis Shipping Demonstration

Size: px
Start display at page:

Download "Milestone 6.3: Basic Analysis Shipping Demonstration"

Transcription

1 The HDF Group Milestone 6.3: Basic Analysis Shipping Demonstration Ruth Aydt, Mohamad Chaarawi, Ivo Jimenez, Quincey Koziol, Jerome Soumagne 12/17/2013 NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL UNDER ITS SUBCONTRACT WITH LAWRENCE LIVERMORE NATIONAL SECURITY, LLC WHO IS THE OPERATOR AND MANAGER OF LAWRENCE LIVERMORE NATIONAL LABORATORY UNDER CONTRACT NO. DE-AC52-07NA27344 WITH THE U.S. DEPARTMENT OF ENERGY. THE UNITED STATES GOVERNMENT RETAINS AND THE PUBLISHER, BY ACCEPTING THE ARTICLE OF PUBLICATION, ACKNOWLEDGES THAT THE UNITED STATES GOVERNMENT RETAINS A NON-EXCLUSIVE, PAID-UP, IRREVOCABLE, WORLD-WIDE LICENSE TO PUBLISH OR REPRODUCE THE PUBLISHED FORM OF THIS MANUSCRIPT, OR ALLOW OTHERS TO DO SO, FOR UNITED STATES GOVERNMENT PURPOSES. THE VIEWS AND OPINIONS OF AUTHORS EXPRESSED HEREIN DO NOT NECESSARILY REFLECT THOSE OF THE UNITED STATES GOVERNMENT OR LAWRENCE LIVERMORE NATIONAL SECURITY, LLC The HDF Group 1

2 Basic Analysis Shipping Objectives Locally enumerate/run analysis on objects that are local to the IO node or storage server node Perform operations as close as possible to the data to minimize data movement costs Easily modify analysis that locally runs on the node Gather results from local enumeration/analysis Main concepts Query: retrieves from an object specific elements / selection that match certain condition(s) Split: parallel and local operation on the data selected by the query Combine: gather data generated by the split operation 2

3 Basic Analysis Shipping Restrictions for this quarter: Queries only operate on dataset elements Ship analysis only to IO nodes (will do storage server nodes in next quarters) 3

4 Architecture Master Filename + Object name + Query + User-defined Python routines Analysis Shipping Client (Mercury client) Python Type "help", "copyright", "credits" or "license" for more information. >>> import h5py_ff >>> h5py_ff.h5asexecute(filename, obj_name, query, split, combine) 4

5 Architecture Master Retrieve storage layout Analysis Shipping Client (Mercury client) Python Type "help", "copyright", "credits" or "license" for more information. >>> import h5py_ff >>> h5py_ff.h5asexecute(filename, obj_name, query, split, combine) 5

6 Architecture Farm split to servers that own part of the object(s) Master Analysis Shipping Client (Mercury client) Python Type "help", "copyright", "credits" or "license" for more information. >>> import h5py_ff >>> h5py_ff.h5asexecute(filename, obj_name, query, split, combine) 6

7 Architecture Master Combine data Analysis Shipping Client (Mercury client) Python Type "help", "copyright", "credits" or "license" for more information. >>> import h5py_ff >>> h5py_ff.h5asexecute(filename, obj_name, query, split, combine) 7

8 Query API /* Query type */ typedef enum H5Q_type_t { H5Q_TYPE_DATA_ELEM, /* selects data elements */ H5Q_TYPE_ATTR_NAME, /* selects attributes */ H5Q_TYPE_LINK_NAME /* selects objects */ } H5Q_type_t; /* Query match conditions */ typedef enum H5Q_match_op_t { H5Q_MATCH_EQUAL, /* equal */ H5Q_MATCH_NOT_EQUAL, /* not equal */ H5Q_MATCH_LESS_THAN, /* less than */ H5Q_MATCH_GREATER_THAN /* greater than */ } H5Q_match_op_t; /* Query combine operators */ typedef enum H5Q_combine_op_t { H5Q_COMBINE_AND, H5Q_COMBINE_OR } H5Q_combine_op_t; 8

9 Query API /* Used on analysis shipping client */ H5_DLL hid_t H5Qcreate(H5Q_type_t query_type, H5Q_match_op_t match_op,...); H5_DLL herr_t H5Qclose(hid_t query_id); H5_DLL hid_t H5Qcombine(hid_t query_id1, H5Q_combine_op_t combine_op, hid_t query_id2); /* Used on server */ H5_DLL herr_t H5Qapply(hid_t query_id, hbool_t *result, hid_t type_id, const void *elem); /* Encode / decode */ H5_DLL herr_t H5Qencode(hid_t query_id, void *buf, size_t *nalloc); H5_DLL hid_t H5Qdecode(const void *buf); 9

10 Demo example - Lola Dataset D1 (512 x 3) CN 0 (lola-12) Process 1 ION 0 (lola-18) /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) /mnt/lustre/bb/lola19/.plfs shadow 10

11 Demo example - Lola D1 Write dataset using hyperslab selection filename: eff_analysis_file.h5 obj_name: D1 ION 0 (lola-18) CN 0 (lola-12) Process 1 /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) /mnt/lustre/bb/lola19/.plfs shadow 11

12 Demo example - Lola Send an analysis shipping request from filename: eff_analysis_file.h5 obj_name: D1 query: (39.1 < x < 42.1) OR (295 < x < 298) split: average combine: average CN 0 (lola-12) ION 0 (lola-18) /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) Process 1 /mnt/lustre/bb/lola19/.plfs shadow 12

13 Demo example - Lola Send an analysis shipping request from filename: eff_analysis_file.h5 obj_name: D1 query: (39.1 < x < 42.1) OR (295 < x < 298) split: average combine: average CN 0 (lola-12) ION 0 (lola-18) iod_obj_query_map() /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) Process 1 /mnt/lustre/bb/lola19/.plfs shadow 13

14 Demo example - Lola Send an analysis shipping request from filename: eff_analysis_file.h5 obj_name: D1 query: (39.1 < x < 42.1) OR (295 < x < 298) split: average combine: average CN 0 (lola-12) Element 40 matches query Farm work and Execute query ION 0 (lola-18) /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) Process 1 Element 296 matches query /mnt/lustre/bb/lola19/.plfs shadow 14

15 Demo example - Lola Send an analysis shipping request from filename: eff_analysis_file.h5 obj_name: D1 query: (39.1 < x < 42.1) OR (295 < x < 298) split: average combine: average CN 0 (lola-12) Split average: 41.0 HDF5->NumPy Execute split ION 0 (lola-18) /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) Process 1 Split average: /mnt/lustre/bb/lola19/.plfs shadow 15

16 Demo example - Lola Send an analysis shipping request from filename: eff_analysis_file.h5 obj_name: D1 query: (39.1 < x < 42.1) OR (295 < x < 298) split: average combine: average CN 0 (lola-12) Combined average: NumPy->HDF5 Gather Data HDF5->NumPy Execute combine ION 0 (lola-18) /mnt/lustre/bb/lola18/.plfs shadow ION 1 (lola-19) Process 1 /mnt/lustre/bb/lola19/.plfs shadow 16

17 Demo Analysis Shipping $ mpirun np 2./h5ff_client_analysis Analysis Execute on file eff_analysis_file.h5 Object D1 (0) Calling iod_container_open on eff_analysis_file.h5 (0) coh 281 objoh 378 objid rtid 2 (0) Calling iod_obj_query_map (0) range: 0, start: 0 0, end: 255 2, n_cell: 768, loc: /mnt/lustre/bb/lola19/.plfs shadow (0) range: 1, start: 256 0, end: 511 2, n_cell: 768, loc: /mnt/lustre/bb/lola18/.plfs shadow (0) Server 1 owns this object (0) Server 0 owns this object Dim 0: start 256 stride 256 block 256 count 1 Dim 1: start 0 stride 3 block 3 count 1 Dim 0: start 0 stride 256 block 256 count 1 Dim 1: start 0 stride 3 block 3 count 1 (0) Element 296 matches query (0) Element 297 matches query (0) Applying split on data (1) Element 40 matches query (1) Element 42 matches query (1) Applying split on data Split sum: 1779 Split average: Split sum: 369 Split average: (1) Transferring split data back to master (0) Applying combine on data Combined sum: Combined average:

18 Conclusion Future Work Add python layer with h5py to call H5ASexecute from python interpreter Cython to automatically generate Python/C wrappers Support queries on attributes and links Run on both IO and storage server nodes Add support inside Mercury for usage of multiple network transports at the same time (external workstation with TCP, internal with IB or other specific interconnect protocol) 18

19 Questions 19

Milestone 8.1: HDF5 Index Demonstration

Milestone 8.1: HDF5 Index Demonstration The HDF Group Milestone 8.1: HDF5 Index Demonstration Ruth Aydt, Mohamad Chaarawi, Quincey Koziol, Aleksandar Jelenak, Jerome Soumagne 06/30/2014 NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY THE HDF GROUP

More information

EFF-IO M7.5 Demo. Semantic Migration of Multi-dimensional Arrays

EFF-IO M7.5 Demo. Semantic Migration of Multi-dimensional Arrays EFF-IO M7.5 Demo Semantic Migration of Multi-dimensional Arrays John Bent, Sorin Faibish, Xuezhao Liu, Harriet Qui, Haiying Tang, Jerry Tirrell, Jingwang Zhang, Kelly Zhang, Zhenhua Zhang NOTICE: THIS

More information

The HDF Group Q5 Demo

The HDF Group Q5 Demo The HDF Group The HDF Group Q5 Demo 5.6 HDF5 Transaction API 5.7 Full HDF5 Dynamic Data Structure NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL UNDER ITS SUBCONTRACT WITH LAWRENCE LIVERMORE NATIONAL

More information

Design Document (Historical) HDF5 Dynamic Data Structure Support FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

Design Document (Historical) HDF5 Dynamic Data Structure Support FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: July 24, 2013 Design Document (Historical) HDF5 Dynamic Data Structure Support FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor

More information

FastForward I/O and Storage: ACG 8.6 Demonstration

FastForward I/O and Storage: ACG 8.6 Demonstration FastForward I/O and Storage: ACG 8.6 Demonstration Kyle Ambert, Jaewook Yu, Arnab Paul Intel Labs June, 2014 NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL UNDER ITS SUBCONTRACT WITH LAWRENCE LIVERMORE

More information

High Level Design IOD KV Store FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

High Level Design IOD KV Store FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: January 10, 2013 High Level Design IOD KV Store FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor Address B599860

More information

FastForward I/O and Storage: ACG 5.8 Demonstration

FastForward I/O and Storage: ACG 5.8 Demonstration FastForward I/O and Storage: ACG 5.8 Demonstration Jaewook Yu, Arnab Paul, Kyle Ambert Intel Labs September, 2013 NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL UNDER ITS SUBCONTRACT WITH LAWRENCE

More information

5.4 - DAOS Demonstration and Benchmark Report

5.4 - DAOS Demonstration and Benchmark Report 5.4 - DAOS Demonstration and Benchmark Report Johann LOMBARDI on behalf of the DAOS team September 25 th, 2013 Livermore (CA) NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL UNDER ITS SUBCONTRACT WITH

More information

High Level Design Client Health and Global Eviction FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O MILESTONE: 4.

High Level Design Client Health and Global Eviction FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O MILESTONE: 4. Date: 2013-06-01 High Level Design Client Health and Global Eviction FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O MILESTONE: 4.1 LLNS Subcontract No. Subcontractor

More information

RFC: HDF5 File Space Management: Paged Aggregation

RFC: HDF5 File Space Management: Paged Aggregation RFC: HDF5 File Space Management: Paged Aggregation Vailin Choi Quincey Koziol John Mainzer The current HDF5 file space allocation accumulates small pieces of metadata and raw data in aggregator blocks.

More information

8.5 End-to-End Demonstration Exascale Fast Forward Storage Team June 30 th, 2014

8.5 End-to-End Demonstration Exascale Fast Forward Storage Team June 30 th, 2014 8.5 End-to-End Demonstration Exascale Fast Forward Storage Team June 30 th, 2014 NOTICE: THIS MANUSCRIPT HAS BEEN AUTHORED BY INTEL, THE HDF GROUP, AND EMC UNDER INTEL S SUBCONTRACT WITH LAWRENCE LIVERMORE

More information

Parallel I/O and Portable Data Formats

Parallel I/O and Portable Data Formats Parallel I/O and Portable Data Formats Sebastian Lührs s.luehrs@fz-juelich.de Jülich Supercomputing Centre Forschungszentrum Jülich GmbH Reykjavík, August 25 th, 2017 Overview I/O can be the main bottleneck

More information

Parallel I/O and Portable Data Formats HDF5

Parallel I/O and Portable Data Formats HDF5 Parallel I/O and Portable Data Formats HDF5 Sebastian Lührs s.luehrs@fz-juelich.de Jülich Supercomputing Centre Forschungszentrum Jülich GmbH Jülich, March 13th, 2018 Outline Introduction Structure of

More information

Automated Characterization of Parallel Application Communication Patterns

Automated Characterization of Parallel Application Communication Patterns Automated Characterization of Parallel Application Communication Patterns Philip C. Roth Jeremy S. Meredith Jeffrey S. Vetter Oak Ridge National Laboratory 17 June 2015 ORNL is managed by UT-Battelle for

More information

HDF5 User s Guide. HDF5 Release November

HDF5 User s Guide. HDF5 Release November HDF5 User s Guide HDF5 Release 1.8.8 November 2011 http://www.hdfgroup.org Copyright Notice and License Terms for HDF5 (Hierarchical Data Format 5) Software Library and Utilities HDF5 (Hierarchical Data

More information

State of OpenMP & Outlook on OpenMP 4.1

State of OpenMP & Outlook on OpenMP 4.1 State of OpenMP & Outlook on OpenMP 4.1 Thursday, October 11, 2015 Bronis R. de Supinski Chair, OpenMP Language Committee This work has been authored by Lawrence Livermore National Security, LLC under

More information

Milestone Burst Buffer & Data Integrity Demonstra>on Milestone End- to- End Epoch Recovery Demonstra>on

Milestone Burst Buffer & Data Integrity Demonstra>on Milestone End- to- End Epoch Recovery Demonstra>on he HF Group ilestone 7.2 - Burst Buffer & ata Integrity emonstra>on ilestone 7.3 - End- to- End Epoch Recovery emonstra>on NOICE: HIS ANUSCRIP HAS BEEN AUHORE BY HE HF GROUP UNER HE INEL SUBCONRAC WIH

More information

Progress on Efficient Integration of Lustre* and Hadoop/YARN

Progress on Efficient Integration of Lustre* and Hadoop/YARN Progress on Efficient Integration of Lustre* and Hadoop/YARN Weikuan Yu Robin Goldstone Omkar Kulkarni Bryon Neitzel * Some name and brands may be claimed as the property of others. MapReduce l l l l A

More information

API and Usage of libhio on XC-40 Systems

API and Usage of libhio on XC-40 Systems API and Usage of libhio on XC-40 Systems May 24, 2018 Nathan Hjelm Cray Users Group May 24, 2018 Los Alamos National Laboratory LA-UR-18-24513 5/24/2018 1 Outline Background HIO Design HIO API HIO Configuration

More information

An exceedingly high-level overview of ambient noise processing with Spark and Hadoop

An exceedingly high-level overview of ambient noise processing with Spark and Hadoop IRIS: USArray Short Course in Bloomington, Indian Special focus: Oklahoma Wavefields An exceedingly high-level overview of ambient noise processing with Spark and Hadoop Presented by Rob Mellors but based

More information

DRAFT. HDF5 Data Flow Pipeline for H5Dread. 1 Introduction. 2 Examples

DRAFT. HDF5 Data Flow Pipeline for H5Dread. 1 Introduction. 2 Examples This document describes the HDF5 library s data movement and processing activities when H5Dread is called for a dataset with chunked storage. The document provides an overview of how memory management,

More information

Power Bounds and Large Scale Computing

Power Bounds and Large Scale Computing 1 Power Bounds and Large Scale Computing Friday, March 1, 2013 Bronis R. de Supinski 1 Tapasya Patki 2, David K. Lowenthal 2, Barry L. Rountree 1 and Martin Schulz 1 2 University of Arizona This work has

More information

Common Persistent Memory POSIX* Runtime (CPPR) API Reference (MS21) API Reference High Performance Data Division

Common Persistent Memory POSIX* Runtime (CPPR) API Reference (MS21) API Reference High Performance Data Division Common Persistent Memory POSIX* Runtime (CPPR) API Reference High Performance Data Division INTEL FEDERAL, LLC PROPRIETARY December 2017 Generated under Argonne Contract number: B609815 DISTRIBUTION STATEMENT:

More information

Implementing HDF5 in MATLAB

Implementing HDF5 in MATLAB Implementing HDF5 in MATLAB Jeff Mather & Alec Rogers The MathWorks, Inc. 2006 The MathWorks, Inc. 29 November 2006 HDF4 1-1 mapping of C API first. (1998) Customer requests for high-level functions. HDFREAD,

More information

NIF ICCS Test Controller for Automated & Manual Testing

NIF ICCS Test Controller for Automated & Manual Testing UCRL-CONF-235325 NIF ICCS Test Controller for Automated & Manual Testing J. S. Zielinski October 5, 2007 International Conference on Accelerator and Large Experimental Physics Control Systems Knoxville,

More information

Hierarchical Data Format 5:

Hierarchical Data Format 5: Hierarchical Data Format 5: Giusy Muscianisi g.muscianisi@cineca.it SuperComputing Applications and Innovation Department May 17th, 2013 Outline What is HDF5? Overview to HDF5 Data Model and File Structure

More information

Reduction Network Discovery Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

Reduction Network Discovery Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: May 01, 2014 Reduction Network Discovery Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor

More information

High Scalability Resource Management with SLURM Supercomputing 2008 November 2008

High Scalability Resource Management with SLURM Supercomputing 2008 November 2008 High Scalability Resource Management with SLURM Supercomputing 2008 November 2008 Morris Jette (jette1@llnl.gov) LLNL-PRES-408498 Lawrence Livermore National Laboratory What is SLURM Simple Linux Utility

More information

Use of a new I/O stack for extreme-scale systems in scientific applications

Use of a new I/O stack for extreme-scale systems in scientific applications 1 Use of a new I/O stack for extreme-scale systems in scientific applications M. Scot Breitenfeld a, Quincey Koziol b, Neil Fortner a, Jerome Soumagne a, Mohamad Chaarawi a a The HDF Group, b Lawrence

More information

and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. '4 L NMAS CORE: UPDATE AND CURRENT DRECTONS DSCLAMER This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any

More information

Adding a System Call to Plan 9

Adding a System Call to Plan 9 Adding a System Call to Plan 9 John Floren (john@csplan9.rit.edu) Sandia National Laboratories Livermore, CA 94551 DOE/NNSA Funding Statement Sandia is a multiprogram laboratory operated by Sandia Corporation,

More information

Mixed language programming with NumPy arrays

Mixed language programming with NumPy arrays Mixed language programming with NumPy arrays Simon Funke 1,2 Ola Skavhaug 3 Joakim Sundnes 1,2 Hans Petter Langtangen 1,2 Center for Biomedical Computing, Simula Research Laboratory 1 Dept. of Informatics,

More information

RFC: HDF5 Virtual Dataset

RFC: HDF5 Virtual Dataset RFC: HDF5 Virtual Dataset Quincey Koziol (koziol@hdfgroup.org) Elena Pourmal (epourmal@hdfgroup.org) Neil Fortner (nfortne2@hdfgroup.org) This document introduces Virtual Datasets (VDS) for HDF5 and summarizes

More information

Introduction to HDF5

Introduction to HDF5 The HDF Group Introduction to HDF5 Quincey Koziol Director of Core Software & HPC The HDF Group October 15, 2014 Blue Waters Advanced User Workshop 1 Why HDF5? Have you ever asked yourself: How will I

More information

Caching and Buffering in HDF5

Caching and Buffering in HDF5 Caching and Buffering in HDF5 September 9, 2008 SPEEDUP Workshop - HDF5 Tutorial 1 Software stack Life cycle: What happens to data when it is transferred from application buffer to HDF5 file and from HDF5

More information

Dynamic Languages for HPC at LLNL VEESC Workshop September 3-4, 2010

Dynamic Languages for HPC at LLNL VEESC Workshop September 3-4, 2010 Dynamic Languages for HPC at LLNL VEESC Workshop September 3-4, 2010 T.J. Alumbaugh Thanks to: Doug Miller, Mike Owen, Tom Brunner, Patrick Brantley (LLNL) Forrest Iandola (UIUC) This work performed under

More information

Accelerated Machine Learning Algorithms in Python

Accelerated Machine Learning Algorithms in Python Accelerated Machine Learning Algorithms in Python Patrick Reilly, Leiming Yu, David Kaeli reilly.pa@husky.neu.edu Northeastern University Computer Architecture Research Lab Outline Motivation and Goals

More information

RFC: Reading Bit field Values from NPOESS Product File

RFC: Reading Bit field Values from NPOESS Product File April7,2009 RFCTHG2009 04 07.v3 RFC:ReadingBit fieldvaluesfromnpoessproductfile ElenaPourmal M.ScotBreitenfeld ThisRFCdescribesahelperAPIthatextractsbit fieldvaluesfromadatasetstoredinan NPOESSproductfile.

More information

Java Based Open Architecture Controller

Java Based Open Architecture Controller Preprint UCRL-JC- 137092 Java Based Open Architecture Controller G. Weinet? This article was submitted to World Automation Conference, Maui, HI, June 1 I- 16,200O U.S. Department of Energy January 13,200O

More information

Common Persistent Memory POSIX Runtime (CPPR) API Reference Manual. Reference Manual High Performance Data Division

Common Persistent Memory POSIX Runtime (CPPR) API Reference Manual. Reference Manual High Performance Data Division Common Persistent Memory POSIX Runtime (CPPR) Reference Manual High Performance Data Division INTEL FEDERAL, LLC PROPRIETARY October 2016 Generated under Argonne Contract number: B609815 DISTRIBUTION STATEMENT:

More information

COSC 6374 Parallel Computation. Scientific Data Libraries. Edgar Gabriel Fall Motivation

COSC 6374 Parallel Computation. Scientific Data Libraries. Edgar Gabriel Fall Motivation COSC 6374 Parallel Computation Scientific Data Libraries Edgar Gabriel Fall 2013 Motivation MPI I/O is good It knows about data types (=> data conversion) It can optimize various access patterns in applications

More information

ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis

ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis ECP Alpine: Algorithms and Infrastructure for In Situ Visualization and Analysis Presented By: Matt Larsen LLNL-PRES-731545 This work was performed under the auspices of the U.S. Department of Energy by

More information

LA-UR Approved for public release; distribution is unlimited.

LA-UR Approved for public release; distribution is unlimited. LA-UR-15-27727 Approved for public release; distribution is unlimited. Title: Survey and Analysis of Multiresolution Methods for Turbulence Data Author(s): Pulido, Jesus J. Livescu, Daniel Woodring, Jonathan

More information

Taming Parallel I/O Complexity with Auto-Tuning

Taming Parallel I/O Complexity with Auto-Tuning Taming Parallel I/O Complexity with Auto-Tuning Babak Behzad 1, Huong Vu Thanh Luu 1, Joseph Huchette 2, Surendra Byna 3, Prabhat 3, Ruth Aydt 4, Quincey Koziol 4, Marc Snir 1,5 1 University of Illinois

More information

Optimizing Bandwidth Utilization in Packet Based Telemetry Systems. Jeffrey R Kalibjian

Optimizing Bandwidth Utilization in Packet Based Telemetry Systems. Jeffrey R Kalibjian UCRL-JC-122361 PREPRINT Optimizing Bandwidth Utilization in Packet Based Telemetry Systems Jeffrey R Kalibjian RECEIVED NOV 17 1995 This paper was prepared for submittal to the 1995 International Telemetry

More information

DAOS API and DAOS POSIX DESIGN DOCUMENT FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

DAOS API and DAOS POSIX DESIGN DOCUMENT FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: December 13, 2012 DAOS API and DAOS POSIX DESIGN DOCUMENT FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor

More information

FastForward I/O and Storage: IOD M5 Demonstration (5.2, 5.3, 5.9, 5.10)

FastForward I/O and Storage: IOD M5 Demonstration (5.2, 5.3, 5.9, 5.10) FastForward I/O and Storage: IOD M5 Demonstration (5.2, 5.3, 5.9, 5.10) 1 EMC September, 2013 John Bent john.bent@emc.com Sorin Faibish faibish_sorin@emc.com Xuezhao Liu xuezhao.liu@emc.com Harriet Qiu

More information

DAOS Server Collectives Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

DAOS Server Collectives Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: June 05, 2013 DAOS Server Collectives Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor Address B599860

More information

zorder-lib: Library API for Z-Order Memory Layout

zorder-lib: Library API for Z-Order Memory Layout zorder-lib: Library API for Z-Order Memory Layout E. Wes Bethel Lawrence Berkeley National Laboratory Berkeley, CA, USA, 94720 April, 2015 i Acknowledgment This work was supported by the Director, Office

More information

Mellanox Scalable Hierarchical Aggregation and Reduction Protocol (SHARP) API Guide. Version 1.0

Mellanox Scalable Hierarchical Aggregation and Reduction Protocol (SHARP) API Guide. Version 1.0 Mellanox Scalable Hierarchical Aggregation and Reduction Protocol (SHARP) API Guide Version 1.0 Table of Contents Copyright... 3 Introduction... 4 Class Index... 5 File Index... 6 Class Documentation...

More information

Progress on OpenMP Specifications

Progress on OpenMP Specifications Progress on OpenMP Specifications Wednesday, November 13, 2012 Bronis R. de Supinski Chair, OpenMP Language Committee This work has been authored by Lawrence Livermore National Security, LLC under contract

More information

The State and Needs of IO Performance Tools

The State and Needs of IO Performance Tools The State and Needs of IO Performance Tools Scalable Tools Workshop Lake Tahoe, CA August 6 12, 2017 This work was performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National

More information

DAOS Epoch Recovery Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

DAOS Epoch Recovery Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: June 4, 2014 DAOS Epoch Recovery Design FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor Address B599860 Intel

More information

h5perf_serial, a Serial File System Benchmarking Tool

h5perf_serial, a Serial File System Benchmarking Tool h5perf_serial, a Serial File System Benchmarking Tool The HDF Group April, 2009 HDF5 users have reported the need to perform serial benchmarking on systems without an MPI environment. The parallel benchmarking

More information

A new perspective on processing-in-memory architecture design

A new perspective on processing-in-memory architecture design A new perspective on processing-in-memory architecture design Dong Ping Zhang, Nuwan Jayasena, Alexander Lyashevsky, Joe Greathouse, Mitesh Meswani, Mark Nutter, Mike Ignatowski AMD Research These data

More information

OPC UA Protocol Stack

OPC UA Protocol Stack Enhanced Universal Realtime Operating System OPC UA Protocol Stack Programming Guide and Reference Document version: 05/2018 EUROS Embedded Systems GmbH Campestraße 12 D-90419 Nuremberg Germany Phone:

More information

PDF Document structure, that need for managing of PDF file. It uses in all functions from EMF2PDF SDK.

PDF Document structure, that need for managing of PDF file. It uses in all functions from EMF2PDF SDK. EMF2PDF SDK Pilot Structures struct pdf_document { PDFDocument4 *pdfdoc; }; PDF Document structure, that need for managing of PDF file. It uses in all functions from EMF2PDF SDK. typedef enum { conone

More information

Introduction to serial HDF5

Introduction to serial HDF5 Introduction to serial HDF Matthieu Haefele Saclay, - March 201, Parallel filesystems and parallel IO libraries PATC@MdS Matthieu Haefele Training outline Day 1: AM: Serial HDF (M. Haefele) PM: Parallel

More information

HDF5 I/O Performance. HDF and HDF-EOS Workshop VI December 5, 2002

HDF5 I/O Performance. HDF and HDF-EOS Workshop VI December 5, 2002 HDF5 I/O Performance HDF and HDF-EOS Workshop VI December 5, 2002 1 Goal of this talk Give an overview of the HDF5 Library tuning knobs for sequential and parallel performance 2 Challenging task HDF5 Library

More information

Data Placement Optimization in GPU Memory Hierarchy Using Predictive Modeling

Data Placement Optimization in GPU Memory Hierarchy Using Predictive Modeling Data Placement Optimization in GPU Memory Hierarchy Using Predictive Modeling Larisa Stoltzfus*, Murali Emani, Pei-Hung Lin, Chunhua Liao *University of Edinburgh (UK), Lawrence Livermore National Laboratory

More information

METADATA REGISTRY, ISO/IEC 11179

METADATA REGISTRY, ISO/IEC 11179 LLNL-JRNL-400269 METADATA REGISTRY, ISO/IEC 11179 R. K. Pon, D. J. Buttler January 7, 2008 Encyclopedia of Database Systems Disclaimer This document was prepared as an account of work sponsored by an agency

More information

RID IETF Draft Update

RID IETF Draft Update RID IETF Draft Update Kathleen M. Moriarty INCH Working Group 29 March 2005 This work was sponsored by the Air Force under Air Force Contract Number F19628-00-C-0002. "Opinions, interpretations, conclusions,

More information

The HDF Group. Parallel HDF5. Extreme Scale Computing Argonne.

The HDF Group. Parallel HDF5. Extreme Scale Computing Argonne. The HDF Group Parallel HDF5 Advantage of Parallel HDF5 Recent success story Trillion particle simulation on hopper @ NERSC 120,000 cores 30TB file 23GB/sec average speed with 35GB/sec peaks (out of 40GB/sec

More information

Remote Visualization, Analysis and other things

Remote Visualization, Analysis and other things Remote Visualization, Analysis and other things Frank Schlünzen DESY-IT The Problems > Remote analysis (access to data, compute or controls) Simple & secure access to resources Experiment / User specific

More information

Parallel I/O and Portable Data Formats I/O strategies

Parallel I/O and Portable Data Formats I/O strategies Parallel I/O and Portable Data Formats I/O strategies Sebastian Lührs s.luehrs@fz-juelich.de Jülich Supercomputing Centre Forschungszentrum Jülich GmbH Jülich, March 13 th, 2017 Outline Common I/O strategies

More information

Keras: Handwritten Digit Recognition using MNIST Dataset

Keras: Handwritten Digit Recognition using MNIST Dataset Keras: Handwritten Digit Recognition using MNIST Dataset IIT PATNA February 9, 2017 1 / 24 OUTLINE 1 Introduction Keras: Deep Learning library for Theano and TensorFlow 2 Installing Keras Installation

More information

Overview of ATLAS PanDA Workload Management

Overview of ATLAS PanDA Workload Management Overview of ATLAS PanDA Workload Management T. Maeno 1, K. De 2, T. Wenaus 1, P. Nilsson 2, G. A. Stewart 3, R. Walker 4, A. Stradling 2, J. Caballero 1, M. Potekhin 1, D. Smith 5, for The ATLAS Collaboration

More information

An introduction to scientific programming with. Session 5: Extreme Python

An introduction to scientific programming with. Session 5: Extreme Python An introduction to scientific programming with Session 5: Extreme Python PyTables For creating, storing and analysing datasets from simple, small tables to complex, huge datasets standard HDF5 file format

More information

Parallel Execution of Functional Mock-up Units in Buildings Modeling

Parallel Execution of Functional Mock-up Units in Buildings Modeling ORNL/TM-2016/173 Parallel Execution of Functional Mock-up Units in Buildings Modeling Ozgur Ozmen James J. Nutaro Joshua R. New Approved for public release. Distribution is unlimited. June 30, 2016 DOCUMENT

More information

Versioning Object Storage Device (VOSD) Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O

Versioning Object Storage Device (VOSD) Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O Date: June 4, 2014 Versioning Object Storage Device (VOSD) Design Document FOR EXTREME-SCALE COMPUTING RESEARCH AND DEVELOPMENT (FAST FORWARD) STORAGE AND I/O LLNS Subcontract No. Subcontractor Name Subcontractor

More information

Getting Insider Information via the New MPI Tools Information Interface

Getting Insider Information via the New MPI Tools Information Interface Getting Insider Information via the New MPI Tools Information Interface EuroMPI 2016 September 26, 2016 Kathryn Mohror This work was performed under the auspices of the U.S. Department of Energy by Lawrence

More information

Parallel I/O and Portable Data Formats PnetCDF and NetCDF 4

Parallel I/O and Portable Data Formats PnetCDF and NetCDF 4 Parallel I/O and Portable Data Formats PnetDF and NetDF 4 Sebastian Lührs s.luehrs@fz-juelich.de Jülich Supercomputing entre Forschungszentrum Jülich GmbH Jülich, March 13 th, 2017 Outline Introduction

More information

Steven Carter. Network Lead, NCCS Oak Ridge National Laboratory OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 1

Steven Carter. Network Lead, NCCS Oak Ridge National Laboratory OAK RIDGE NATIONAL LABORATORY U. S. DEPARTMENT OF ENERGY 1 Networking the National Leadership Computing Facility Steven Carter Network Lead, NCCS Oak Ridge National Laboratory scarter@ornl.gov 1 Outline Introduction NCCS Network Infrastructure Cray Architecture

More information

CTF: State-of-the-Art and Building the Next Generation ASE 2017

CTF: State-of-the-Art and Building the Next Generation ASE 2017 CTF: State-of-the-Art and Building the Next Generation ASE 2017 August 15, 2017 Clark Taylor Lawrence Livermore National Laboratory University of Arizona 737334 LLNL-PRES-XXXXXX This work was performed

More information

cstore_fdw Columnar store for analytic workloads Hadi Moshayedi & Ben Redman

cstore_fdw Columnar store for analytic workloads Hadi Moshayedi & Ben Redman cstore_fdw Columnar store for analytic workloads Hadi Moshayedi & Ben Redman What is CitusDB? CitusDB is a scalable analytics database that extends PostgreSQL Citus shards your data and automa/cally parallelizes

More information

LA-UR Approved for public release; distribution is unlimited.

LA-UR Approved for public release; distribution is unlimited. LA-UR-15-27727 Approved for public release; distribution is unlimited. Title: Survey and Analysis of Multiresolution Methods for Turbulence Data Author(s): Pulido, Jesus J. Livescu, Daniel Woodring, Jonathan

More information

A new approach to interoperability using HDF5

A new approach to interoperability using HDF5 A new approach to interoperability using HDF5 Second International Workshop on Software Solutions for Integrated Computational Materials Engineering ICME 2016 14 th April 2016, Barcelona, Spain Anshuman

More information

SSC14-V-9 UNCLASSIFIED UNCLASSIFIED LLNL-PRES

SSC14-V-9 UNCLASSIFIED UNCLASSIFIED LLNL-PRES Government-owned CubeSat Next Generation Bus Reference Architecture Vincent Riot, Darrell Carter, Todd Decker, Lance Simms (LLNL) Jim Newman, Lara Magallanes, Jim Horning, David Rigmaiden (NPS) Meagan

More information

HDF5: theory & practice

HDF5: theory & practice HDF5: theory & practice Giorgio Amati SCAI Dept. 15/16 May 2014 Agenda HDF5: main issues Using the API (serial) Using the API (parallel) Tools Some comments PHDF5 Initial Target Support for MPI programming

More information

The HDF Group. Parallel HDF5. Quincey Koziol Director of Core Software & HPC The HDF Group.

The HDF Group. Parallel HDF5. Quincey Koziol Director of Core Software & HPC The HDF Group. The HDF Group Parallel HDF5 Quincey Koziol Director of Core Software & HPC The HDF Group Parallel HDF5 Success Story Recent success story Trillion particle simulation on hopper @ NERSC 120,000 cores 30TB

More information

Noise Injection Techniques to Expose Subtle and Unintended Message Races

Noise Injection Techniques to Expose Subtle and Unintended Message Races Noise Injection Techniques to Expose Subtle and Unintended Message Races PPoPP2017 February 6th, 2017 Kento Sato, Dong H. Ahn, Ignacio Laguna, Gregory L. Lee, Martin Schulz and Christopher M. Chambreau

More information

Performance Tools and Holistic HPC Workflows

Performance Tools and Holistic HPC Workflows Performance Tools and Holistic HPC Workflows Karen L. Karavanic Portland State University Work Performed with: Holistic HPC Workflows: David Montoya (LANL) PSU Drought Project: Yasodha Suriyakumar (CS),

More information

Nail. A Practical Tool for Parsing and Generating Data Formats. Julian Bangert, Nickolai Zeldovich MIT CSAIL. OSDI 14 October 2014

Nail. A Practical Tool for Parsing and Generating Data Formats. Julian Bangert, Nickolai Zeldovich MIT CSAIL. OSDI 14 October 2014 Nail A Practical Tool for Parsing and Generating Data Formats Julian Bangert, Nickolai Zeldovich MIT CSAIL OSDI 14 October 2014 1 / 12 Motivation Parsing Vulnerabilities hand-crafted input parsing and

More information

End-to-End Data Integrity in the Intel/EMC/HDF Group Exascale IO DOE Fast Forward Project

End-to-End Data Integrity in the Intel/EMC/HDF Group Exascale IO DOE Fast Forward Project End-to-End Data Integrity in the Intel/EMC/HDF Group Exascale IO DOE Fast Forward Project As presented by John Bent, EMC and Quincey Koziol, The HDF Group Truly End-to-End App provides checksum buffer

More information

APPLICATION NOTE. Atmel AT03261: SAM D20 System Interrupt Driver (SYSTEM INTERRUPT) SAM D20 System Interrupt Driver (SYSTEM INTERRUPT)

APPLICATION NOTE. Atmel AT03261: SAM D20 System Interrupt Driver (SYSTEM INTERRUPT) SAM D20 System Interrupt Driver (SYSTEM INTERRUPT) APPLICATION NOTE Atmel AT03261: SAM D20 System Interrupt Driver (SYSTEM INTERRUPT) ASF PROGRAMMERS MANUAL SAM D20 System Interrupt Driver (SYSTEM INTERRUPT) This driver for SAM D20 devices provides an

More information

HDF5 Single Writer/Multiple Reader Feature Design and Semantics

HDF5 Single Writer/Multiple Reader Feature Design and Semantics HDF5 Single Writer/Multiple Reader Feature Design and Semantics The HDF Group Document Version 5.2 This document describes the design and semantics of HDF5 s single writer/multiplereader (SWMR) feature.

More information

Nimsoft Monitor. ntp_response Guide. v1.2 series

Nimsoft Monitor. ntp_response Guide. v1.2 series Nimsoft Monitor ntp_response Guide v1.2 series Legal Notices Copyright 2012, CA. All rights reserved. Warranty The material contained in this document is provided "as is," and is subject to being changed,

More information

HDF5: An Introduction. Adam Carter EPCC, The University of Edinburgh

HDF5: An Introduction. Adam Carter EPCC, The University of Edinburgh HDF5: An Introduction Adam Carter EPCC, The University of Edinburgh What is HDF5? Hierarchical Data Format (version 5) From www.hdfgroup.org: HDF5 is a unique technology suite that makes possible the management

More information

FY97 ICCS Prototype Specification

FY97 ICCS Prototype Specification FY97 ICCS Prototype Specification John Woodruff 02/20/97 DISCLAIMER This document was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government

More information

Seagate Kinetic Open Storage Platform. Mayur Shetty - Senior Solutions Architect

Seagate Kinetic Open Storage Platform. Mayur Shetty - Senior Solutions Architect Seagate Kinetic Open Storage Platform Mayur Shetty - Senior Solutions Architect 2 Application Clustering Management Interconnect App App LibKinetic App A D App No. 77103, LibKinetic effective Jan. 18,

More information

Memory management. Johan Montelius KTH

Memory management. Johan Montelius KTH Memory management Johan Montelius KTH 2017 1 / 22 C program # include int global = 42; int main ( int argc, char * argv []) { if( argc < 2) return -1; int n = atoi ( argv [1]); int on_stack

More information

Improved Versioning, Building, and Distribution of Lustre

Improved Versioning, Building, and Distribution of Lustre Improved Versioning, Building, and Distribution of Lustre LUG 2016 Christopher J. Morrone, Giuseppe Di Natale April 5, 2016 This work was performed under the auspices of the U.S. Department of Energy by

More information

ENDF/B-VII.1 versus ENDFB/-VII.0: What s Different?

ENDF/B-VII.1 versus ENDFB/-VII.0: What s Different? LLNL-TR-548633 ENDF/B-VII.1 versus ENDFB/-VII.0: What s Different? by Dermott E. Cullen Lawrence Livermore National Laboratory P.O. Box 808/L-198 Livermore, CA 94550 March 17, 2012 Approved for public

More information

libhio: Optimizing IO on Cray XC Systems With DataWarp

libhio: Optimizing IO on Cray XC Systems With DataWarp libhio: Optimizing IO on Cray XC Systems With DataWarp May 9, 2017 Nathan Hjelm Cray Users Group May 9, 2017 Los Alamos National Laboratory LA-UR-17-23841 5/8/2017 1 Outline Background HIO Design Functionality

More information

Let s Rust in Samba. Trying to use Samba with Rust libraries. Kai Blin Samba Team. SambaXP

Let s Rust in Samba. Trying to use Samba with Rust libraries. Kai Blin Samba Team. SambaXP Let s Rust in Samba Trying to use Samba with Rust libraries Kai Blin Samba Team SambaXP 2018 2017-06-06 Intro M.Sc. in Computational Biology Ph.D. in Microbiology Samba Team member 2/42 Overview Rust Intro

More information

Mercury: Enabling Remote Procedure Call for High-Performance Computing

Mercury: Enabling Remote Procedure Call for High-Performance Computing Mercury: Enabling Remote Procedure Call for High-Performance Computing Jerome Soumagne, Dries Kimpe, Judicael Zounmevo, Mohamad Chaarawi, Quincey Koziol, Ahmad Afsahi, Robert Ross The HDF Group Champaign,

More information

Fast, Interactive, Language-Integrated Cluster Computing

Fast, Interactive, Language-Integrated Cluster Computing Spark Fast, Interactive, Language-Integrated Cluster Computing Matei Zaharia, Mosharaf Chowdhury, Tathagata Das, Ankur Dave, Justin Ma, Murphy McCauley, Michael Franklin, Scott Shenker, Ion Stoica www.spark-project.org

More information

Parallel I/O CPS343. Spring Parallel and High Performance Computing. CPS343 (Parallel and HPC) Parallel I/O Spring / 22

Parallel I/O CPS343. Spring Parallel and High Performance Computing. CPS343 (Parallel and HPC) Parallel I/O Spring / 22 Parallel I/O CPS343 Parallel and High Performance Computing Spring 2018 CPS343 (Parallel and HPC) Parallel I/O Spring 2018 1 / 22 Outline 1 Overview of parallel I/O I/O strategies 2 MPI I/O 3 Parallel

More information

Opening up OpenSM with the Subnet Monitoring Tools

Opening up OpenSM with the Subnet Monitoring Tools Opening up OpenSM with the Subnet Monitoring Tools OFS User Group Workshop March 19, 2015 Timothy Meier tameier@llnl.gov This work was performed under the auspices of the U.S. Department of Energy by under

More information

A performance portable implementation of HOMME via the Kokkos programming model

A performance portable implementation of HOMME via the Kokkos programming model E x c e p t i o n a l s e r v i c e i n t h e n a t i o n a l i n t e re s t A performance portable implementation of HOMME via the Kokkos programming model L.Bertagna, M.Deakin, O.Guba, D.Sunderland,

More information