FastForward I/O and Storage: IOD M5 Demonstration (5.2, 5.3, 5.9, 5.10)
|
|
- Marian Owen
- 5 years ago
- Views:
Transcription
1 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 harrietz.qiu@emc.com Haiying Tang haiying.tang@emc.com Jerry Tirrell jerry.tirrell@emc.com Kelly Zhang kelly.zhang@emc.com Zhenhua Zhang zhenhua.zhang@emc.com
2 Agenda Brief introduction Source tree structure & build Demo objectives & environment Demonstrations Upcoming work 2
3 Introduction: IOD in FF stack CN 0 CN 1 CN n Rank 0, 1... Rank a, a+1... Rank x, x+1... Parallel application Process group Process group HDF5 library + VOL plugin CNs Function shipping client Fabrics IONs ION 0 ION m Function shipping server Function shipping server s process space Function shipping server s process space IOD (With POSIX BB) DAOS client Global storage DAOS 3
4 Introduction: IOD sub-modules overview API front-end Executing-Engine Task queue Event-Queues I/O scheduler Worker thread Pool Transaction manager Container manager KV object KV-store Object manager POSIX/DAOS storage access BLOB/ARRAY PLFS Inter-IODs communication 4
5 Source tree structure & build Source tree: Include - API header files Src - library source files Src/dep - dependent third party libraries such as plfs, mdhim. Test - testing/demonstration code Tools - simple binary tools Example - simple examples Build Just executes make in top directory (Will add autoconf/cmake script later) 5
6 Demo objectives M5 demonstration is specified to support Milestones: 5.2 IOD Object Store Demonstration 5.3 N ->M Stream Demonstration 5.9 Structure Storage and Retrieval between HDF and IO Dispatcher 5.10 Multi-format Replicas in the IO Dispatcher 6
7 In-scope demonstration elements Create, unlink, and listing contents of containers on IONs Create, unlink, read, write of kv, blob, and array objects on IONs Showing the different parameters for each object E.g. get/put for kv, byte offset for blobs, semantic offsets for arrays Migration of kv and blob objects between DAOS and IONs Will be shown in both directions Showing N-M stream transformation Querying layout of blob, kv, and array objects i.e. Returning a map of data chunks to storage locations Create, read, unlink of multi-format replicas of blob objects on IONs E.g. Create a replica of a blob with a different sharding 7
8 SOW Mapping 5.2. The Subcontractor shall demonstrate the initial object storage functionality of the I/O dispatcher. This will include an abstract storage layer with which PLFS can appropriately store data using a variety of storage interfaces such as the DAOS object store, possible key-value storage systems, POSIX, etc. [Since the SOW, the anticipated key-value storage interface to Flash memory did not materialize. Therefore, this milestone will demonstrate only the DAOS and POSIX interfaces by which IOD accesses physical storage devices.] 5.3. The Subcontractor shall demonstrate the initial N->M stream transformation functionality. This transformation is important to ensure that the IO dispatcher manages concurrency appropriately. If, for example, the dispatcher receives a stream of IO from every process but only has a more limited level of concurrency into either of the storage tiers, it will aggregate streams as necessary to reach an appropriate degree of concurrency. [Since the SOW, our thinking about this milestone has crystalized such that this demonstration will show how N streams between CN s and ION s are rearranged into M streams for the migration between ION s and DAOS. Changing N IO streams from HDF into M IO streams to the burst buffers will not be demo d here; rather N streams in burst buffers into M streams to DAOS.] The Subcontractor shall demonstrate the ability of HDF to store structural knowledge alongside the data using the IO Dispatcher API. This will allow the IO Dispatcher to make intelligent structure-aware sharding decisions. The demonstration will include a layout query by which an application (or job scheduler) can query the IO Dispatcher to discover on which burst buffer(s) a structurally-defined shard resides. [There are no clarifications or revisions to this milestone since the SOW.] The Subcontractor shall demonstrate the ability of the IO Dispatcher to store copies of data in different sharding organizations. Use of the existing plfs_map command will demonstrate that bytes ranges are replicated and will reveal in which shards, or on which burst buffers, each data chunk resides. This will be demonstrated at this time using simple byterange organizations (IOD blob objects) and the more complex demonstration where the replicas are structurally meaningful (IOD array objects) will be demonstrated in a later deliverable. [The functionality of this milestone is unchanged since the SOW; but the specific tool used to demonstrate this may not be the existing plfs_map command but an IOD version. This milestone is often referred to as multi-format replicas. In future deliverables, when this functionality is extended to array objects, it will be also referred to as semantic resharding.] 8
9 IOD Solution Architecture Mapping IOD SA 4.1: HDF / IOD API We will fully cover this by providing the full IOD API although not all functions will be fully implemented as described above. Of this SA, we will demonstrate the following: That the API for reading and writing all three IOD object types is fully implemented. IOD SA 4.2: Object storage We will partially cover this by demonstrating the following: Reads and writes to IOD objects perform correctly. IOD objects can be created and unlinked correctly. IOD containers can be read to retrieve the correct set of IOD objects existent in that container. IOD SA: 4.3 Data migration We will partially cover this by demonstrating the following: Migration in both directions of blob and kv objects. This will show the N-M transformation also as we ll provide a simple iod_obj_migrate command which will transform the N streams from the source into M streams in the target where M is aligned with the number of storage devices in the target. This may involve scatter-gather operations between IOD daemons in order to flatten plfs-style logs into a nicely striped set of objects. IOD SA 4.5. Structural awareness, KV stores, and data blobs We will partially cover this by demonstrating the following: Structural storage of distributed data structures into IOD array objects Reads and writes, creates and unlink of all object types 9
10 Out-of-scope demonstration elements Exascale test-bed will not be ready. Transactional I/O (6.2) End-to-end data integrity (7.3) Migration of array objects (7.5) Multi-format replicas of KV and array objects (8.3/8.4) Asynchronous IOD functionality (8.5) 10
11 Complete demo system VNX7500 with 1 15 drive tray, 1 Voyager (60 Drives) 1 Supermicro abba (4 IB ports, GPGPU) 1 Megatron abba (4 IB ports) 4 Lustre servers (1MDS, 3OSS) (1 IB each) 8 Compute Nodes (1 IB each) 1 Cisco 3560 Gigabit switch for managemement 1 Mellanox QDR switch 11
12 Demo configuration MPI Test Program BB 0 BB 1 DAOS POSIX 12
13 And away we go! 13
14 14
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 informationHigh 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 informationEnd-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 information8.5 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 informationlibhio: 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 informationLustre* - Fast Forward to Exascale High Performance Data Division. Eric Barton 18th April, 2013
Lustre* - Fast Forward to Exascale High Performance Data Division Eric Barton 18th April, 2013 DOE Fast Forward IO and Storage Exascale R&D sponsored by 7 leading US national labs Solutions to currently
More informationFast Forward I/O & Storage
Fast Forward I/O & Storage Eric Barton Lead Architect 1 Department of Energy - Fast Forward Challenge FastForward RFP provided US Government funding for exascale research and development Sponsored by 7
More informationUK LUG 10 th July Lustre at Exascale. Eric Barton. CTO Whamcloud, Inc Whamcloud, Inc.
UK LUG 10 th July 2012 Lustre at Exascale Eric Barton CTO Whamcloud, Inc. eeb@whamcloud.com Agenda Exascale I/O requirements Exascale I/O model 3 Lustre at Exascale - UK LUG 10th July 2012 Exascale I/O
More informationDesign 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 informationAPI 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 information5.4 - 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 informationFastForward 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 informationEMC Lustre Contributions
EMC Lustre Contributions It s all about speed. Tao Peng Xuezhao Liu as presented by John Bent Fast Data Group Office of the CTO 1 EMC Lustre activities Support Lustre bug fixes (LU-1126, LU-1322, etc.)
More informationStructuring PLFS for Extensibility
Structuring PLFS for Extensibility Chuck Cranor, Milo Polte, Garth Gibson PARALLEL DATA LABORATORY Carnegie Mellon University What is PLFS? Parallel Log Structured File System Interposed filesystem b/w
More informationMS 52 Distributed Persistent Memory Class Storage Model (DAOS-M) Solution Architecture Revision 1.2 September 28, 2015
MS 52 Distributed Persistent Memory Class Storage Model (DAOS-M) Solution Architecture Revision 1.2 September 28, 2015 Intel Federal, LLC Proprietary i Solution Architecture Generated under Argonne Contract
More informationExtreme I/O Scaling with HDF5
Extreme I/O Scaling with HDF5 Quincey Koziol Director of Core Software Development and HPC The HDF Group koziol@hdfgroup.org July 15, 2012 XSEDE 12 - Extreme Scaling Workshop 1 Outline Brief overview of
More informationLustre overview and roadmap to Exascale computing
HPC Advisory Council China Workshop Jinan China, October 26th 2011 Lustre overview and roadmap to Exascale computing Liang Zhen Whamcloud, Inc liang@whamcloud.com Agenda Lustre technology overview Lustre
More informationExascale Process Management Interface
Exascale Process Management Interface Ralph Castain Intel Corporation rhc@open-mpi.org Joshua S. Ladd Mellanox Technologies Inc. joshual@mellanox.com Artem Y. Polyakov Mellanox Technologies Inc. artemp@mellanox.com
More informationAn Exploration into Object Storage for Exascale Supercomputers. Raghu Chandrasekar
An Exploration into Object Storage for Exascale Supercomputers Raghu Chandrasekar Agenda Introduction Trends and Challenges Design and Implementation of SAROJA Preliminary evaluations Summary and Conclusion
More informationFastForward 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 informationLustre and PLFS Parallel I/O Performance on a Cray XE6
Lustre and PLFS Parallel I/O Performance on a Cray XE6 Cray User Group 2014 Lugano, Switzerland May 4-8, 2014 April 2014 1 Many currently contributing to PLFS LANL: David Bonnie, Aaron Caldwell, Gary Grider,
More informationIntroduction to High Performance Parallel I/O
Introduction to High Performance Parallel I/O Richard Gerber Deputy Group Lead NERSC User Services August 30, 2013-1- Some slides from Katie Antypas I/O Needs Getting Bigger All the Time I/O needs growing
More informationDAOS and Friends: A Proposal for an Exascale Storage System
DAOS and Friends: A Proposal for an Exascale Storage System Jay Lofstead, Ivo Jimenez, Carlos Maltzahn, Quincey Koziol, John Bent, Eric Barton Sandia National Laboratories gflofst@sandia.gov University
More informationCRFS: A Lightweight User-Level Filesystem for Generic Checkpoint/Restart
CRFS: A Lightweight User-Level Filesystem for Generic Checkpoint/Restart Xiangyong Ouyang, Raghunath Rajachandrasekar, Xavier Besseron, Hao Wang, Jian Huang, Dhabaleswar K. Panda Department of Computer
More informationFile Open, Close, and Flush Performance Issues in HDF5 Scot Breitenfeld John Mainzer Richard Warren 02/19/18
File Open, Close, and Flush Performance Issues in HDF5 Scot Breitenfeld John Mainzer Richard Warren 02/19/18 1 Introduction Historically, the parallel version of the HDF5 library has suffered from performance
More informationSDS: A Framework for Scientific Data Services
SDS: A Framework for Scientific Data Services Bin Dong, Suren Byna*, John Wu Scientific Data Management Group Lawrence Berkeley National Laboratory Finding Newspaper Articles of Interest Finding news articles
More informationStore Process Analyze Collaborate Archive Cloud The HPC Storage Leader Invent Discover Compete
Store Process Analyze Collaborate Archive Cloud The HPC Storage Leader Invent Discover Compete 1 DDN Who We Are 2 We Design, Deploy and Optimize Storage Systems Which Solve HPC, Big Data and Cloud Business
More informationA Plugin for HDF5 using PLFS for Improved I/O Performance and Semantic Analysis
2012 SC Companion: High Performance Computing, Networking Storage and Analysis A for HDF5 using PLFS for Improved I/O Performance and Semantic Analysis Kshitij Mehta, John Bent, Aaron Torres, Gary Grider,
More informationReal-time monitoring Slurm jobs with InfluxDB September Carlos Fenoy García
Real-time monitoring Slurm jobs with InfluxDB September 2016 Carlos Fenoy García Agenda Problem description Current Slurm profiling Our solution Conclusions Problem description Monitoring of jobs is becoming
More informationData Management. Parallel Filesystems. Dr David Henty HPC Training and Support
Data Management Dr David Henty HPC Training and Support d.henty@epcc.ed.ac.uk +44 131 650 5960 Overview Lecture will cover Why is IO difficult Why is parallel IO even worse Lustre GPFS Performance on ARCHER
More informationIME (Infinite Memory Engine) Extreme Application Acceleration & Highly Efficient I/O Provisioning
IME (Infinite Memory Engine) Extreme Application Acceleration & Highly Efficient I/O Provisioning September 22 nd 2015 Tommaso Cecchi 2 What is IME? This breakthrough, software defined storage application
More informationECMWF's Next Generation IO for the IFS Model and Product Generation
ECMWF's Next Generation IO for the IFS Model and Product Generation Future workflow adaptations Tiago Quintino, B. Raoult, S. Smart, A. Bonanni, F. Rathgeber, P. Bauer ECMWF tiago.quintino@ecmwf.int ECMWF
More informationThe Exascale Architecture
The Exascale Architecture Richard Graham HPC Advisory Council China 2013 Overview Programming-model challenges for Exascale Challenges for scaling MPI to Exascale InfiniBand enhancements Dynamically Connected
More information<Insert Picture Here> End-to-end Data Integrity for NFS
End-to-end Data Integrity for NFS Chuck Lever Consulting Member of Technical Staff Today s Discussion What is end-to-end data integrity? T10 PI overview Adapting
More informationFast Forward Storage & I/O. Jeff Layton (Eric Barton)
Fast Forward & I/O Jeff Layton (Eric Barton) DOE Fast Forward IO and Exascale R&D sponsored by 7 leading US national labs Solutions to currently intractable problems of Exascale required to meet the 2020
More informationParallel File Systems. John White Lawrence Berkeley National Lab
Parallel File Systems John White Lawrence Berkeley National Lab Topics Defining a File System Our Specific Case for File Systems Parallel File Systems A Survey of Current Parallel File Systems Implementation
More informationTest On Line: reusing SAS code in WEB applications Author: Carlo Ramella TXT e-solutions
Test On Line: reusing SAS code in WEB applications Author: Carlo Ramella TXT e-solutions Chapter 1: Abstract The Proway System is a powerful complete system for Process and Testing Data Analysis in IC
More informationMEAN Stack. 1. Introduction. 2. Foundation a. The Node.js framework b. Installing Node.js c. Using Node.js to execute scripts
MEAN Stack 1. Introduction 2. Foundation a. The Node.js framework b. Installing Node.js c. Using Node.js to execute scripts 3. Node Projects a. The Node Package Manager b. Creating a project c. The package.json
More informationCrossing the Chasm: Sneaking a parallel file system into Hadoop
Crossing the Chasm: Sneaking a parallel file system into Hadoop Wittawat Tantisiriroj Swapnil Patil, Garth Gibson PARALLEL DATA LABORATORY Carnegie Mellon University In this work Compare and contrast large
More informationCSE544 Database Architecture
CSE544 Database Architecture Tuesday, February 1 st, 2011 Slides courtesy of Magda Balazinska 1 Where We Are What we have already seen Overview of the relational model Motivation and where model came from
More informationWhy Actors Rock: Designing a Distributed Database with libcppa
Why Actors Rock: Designing a Distributed Database with libcppa Matthias Vallentin matthias@bro.org University of California, Berkeley C ++ Now May 15, 2014 Outline 1. System Overview: VAST 2. Architecture:
More informationRobinHood Project Status
FROM RESEARCH TO INDUSTRY RobinHood Project Status Robinhood User Group 2015 Thomas Leibovici 9/18/15 SEPTEMBER, 21 st 2015 Project history... 1999: simple purge tool for HPC
More informationCA485 Ray Walshe Google File System
Google File System Overview Google File System is scalable, distributed file system on inexpensive commodity hardware that provides: Fault Tolerance File system runs on hundreds or thousands of storage
More informationUse 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 informationParallel I/O Libraries and Techniques
Parallel I/O Libraries and Techniques Mark Howison User Services & Support I/O for scientifc data I/O is commonly used by scientific applications to: Store numerical output from simulations Load initial
More informationPersistent Memory over Fabric (PMoF) Adding RDMA to Persistent Memory Pawel Szymanski Intel Corporation
Persistent Memory over Fabric (PMoF) Adding RDMA to Persistent Memory Pawel Szymanski Intel Corporation 1 Adding RDMA to Persisteny memory Agenda PMoF Overview Comparison with other remote replication
More informationI/O in scientific applications
COSC 4397 Parallel I/O (II) Access patterns Spring 2010 I/O in scientific applications Different classes of I/O operations Required I/O: reading input data and writing final results Checkpointing: data
More informationWrite a technical report Present your results Write a workshop/conference paper (optional) Could be a real system, simulation and/or theoretical
Identify a problem Review approaches to the problem Propose a novel approach to the problem Define, design, prototype an implementation to evaluate your approach Could be a real system, simulation and/or
More informationGridNFS: Scaling to Petabyte Grid File Systems. Andy Adamson Center For Information Technology Integration University of Michigan
GridNFS: Scaling to Petabyte Grid File Systems Andy Adamson Center For Information Technology Integration University of Michigan What is GridNFS? GridNFS is a collection of NFS version 4 features and minor
More informationAndreas Dilger High Performance Data Division RUG 2016, Paris
Andreas Dilger High Performance Data Division RUG 2016, Paris Multi-Tiered Storage and File Level Redundancy Full direct data access from clients to all storage classes Management Target (MGT) Metadata
More informationCrossing the Chasm: Sneaking a parallel file system into Hadoop
Crossing the Chasm: Sneaking a parallel file system into Hadoop Wittawat Tantisiriroj Swapnil Patil, Garth Gibson PARALLEL DATA LABORATORY Carnegie Mellon University In this work Compare and contrast large
More informationMilestone 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 informationChallenges in HPC I/O
Challenges in HPC I/O Universität Basel Julian M. Kunkel German Climate Computing Center / Universität Hamburg 10. October 2014 Outline 1 High-Performance Computing 2 Parallel File Systems and Challenges
More informationAdvanced Software for the Supercomputer PRIMEHPC FX10. Copyright 2011 FUJITSU LIMITED
Advanced Software for the Supercomputer PRIMEHPC FX10 System Configuration of PRIMEHPC FX10 nodes Login Compilation Job submission 6D mesh/torus Interconnect Local file system (Temporary area occupied
More informationLustre Lockahead: Early Experience and Performance using Optimized Locking. Michael Moore
Lustre Lockahead: Early Experience and Performance using Optimized Locking Michael Moore Agenda Purpose Investigate performance of a new Lustre and MPI-IO feature called Lustre Lockahead (LLA) Discuss
More informationNew Storage Architectures
New Storage Architectures OpenFabrics Software User Group Workshop Replacing LNET routers with IB routers #OFSUserGroup Lustre Basics Lustre is a clustered file-system for supercomputing Architecture consists
More informationLecture 7: February 10
CMPSCI 677 Operating Systems Spring 2016 Lecture 7: February 10 Lecturer: Prashant Shenoy Scribe: Tao Sun 7.1 Server Design Issues 7.1.1 Server Design There are two types of server design choices: Iterative
More informationUsing DDN IME for Harmonie
Irish Centre for High-End Computing Using DDN IME for Harmonie Gilles Civario, Marco Grossi, Alastair McKinstry, Ruairi Short, Nix McDonnell April 2016 DDN IME: Infinite Memory Engine IME: Major Features
More informationLustre A Platform for Intelligent Scale-Out Storage
Lustre A Platform for Intelligent Scale-Out Storage Rumi Zahir, rumi. May 2003 rumi.zahir@intel.com Agenda Problem Statement Trends & Current Data Center Storage Architectures The Lustre File System Project
More informationXtreemFS a case for object-based storage in Grid data management. Jan Stender, Zuse Institute Berlin
XtreemFS a case for object-based storage in Grid data management Jan Stender, Zuse Institute Berlin In this talk... Traditional Grid Data Management Object-based file systems XtreemFS Grid use cases for
More informationMetadata Performance Evaluation LUG Sorin Faibish, EMC Branislav Radovanovic, NetApp and MD BWG April 8-10, 2014
Metadata Performance Evaluation Effort @ LUG 2014 Sorin Faibish, EMC Branislav Radovanovic, NetApp and MD BWG April 8-10, 2014 OpenBenchmark Metadata Performance Evaluation Effort (MPEE) Team Leader: Sorin
More informationNational Aeronautics and Space and Administration Space Administration. cfe Release 6.6
National Aeronautics and Space and Administration Space Administration cfe Release 6.6 1 1 A Summary of cfe 6.6 All qualification testing and documentation is now complete and the release has been tagged
More informationGot Burst Buffer. Now What? Early experiences, exciting future possibilities, and what we need from the system to make it work
Got Burst Buffer. Now What? Early experiences, exciting future possibilities, and what we need from the system to make it work The Salishan Conference on High-Speed Computing April 26, 2016 Adam Moody
More informationDistributed Filesystem
Distributed Filesystem 1 How do we get data to the workers? NAS Compute Nodes SAN 2 Distributing Code! Don t move data to workers move workers to the data! - Store data on the local disks of nodes in the
More informationLustre* is designed to achieve the maximum performance and scalability for POSIX applications that need outstanding streamed I/O.
Reference Architecture Designing High-Performance Storage Tiers Designing High-Performance Storage Tiers Intel Enterprise Edition for Lustre* software and Intel Non-Volatile Memory Express (NVMe) Storage
More informationCentera Fixed Content Device configuration troubleshooting, and tuning
Centera Fixed Content Device configuration troubleshooting, and tuning Storage Management Ruben Salazar and Patricia Gatewood - CE FileNET P8 support September 30 th. 2015 Agenda Introduction to Centera
More informationLoosely coupled: asynchronous processing, decoupling of tiers/components Fan-out the application tiers to support the workload Use cache for data and content Reduce number of requests if possible Batch
More informationHPC Storage Use Cases & Future Trends
Oct, 2014 HPC Storage Use Cases & Future Trends Massively-Scalable Platforms and Solutions Engineered for the Big Data and Cloud Era Atul Vidwansa Email: atul@ DDN About Us DDN is a Leader in Massively
More informationDemonstration Milestone for Parallel Directory Operations
Demonstration Milestone for Parallel Directory Operations This milestone was submitted to the PAC for review on 2012-03-23. This document was signed off on 2012-04-06. Overview This document describes
More informationA memcached implementation in Java. Bela Ban JBoss 2340
A memcached implementation in Java Bela Ban JBoss 2340 AGENDA 2 > Introduction > memcached > memcached in Java > Improving memcached > Infinispan > Demo Introduction 3 > We want to store all of our data
More informationEIOW Exa-scale I/O workgroup (exascale10)
EIOW Exa-scale I/O workgroup (exascale10) Meghan McClelland Peter Braam Lug 2013 Large scale data management is fundamentally broken but functions somewhat successfully as an awkward patchwork Current
More informationThe State of Samba (June 2011) Jeremy Allison Samba Team/Google Open Source Programs Office
The State of Samba (June 2011) Jeremy Allison Samba Team/Google Open Source Programs Office jra@samba.org jra@google.com What is Samba? Provides File/Print/Authentication services to Windows clients from
More informationPLFS and Lustre Performance Comparison
PLFS and Lustre Performance Comparison Lustre User Group 2014 Miami, FL April 8-10, 2014 April 2014 1 Many currently contributing to PLFS LANL: David Bonnie, Aaron Caldwell, Gary Grider, Brett Kettering,
More informationMotivation. Threads. Multithreaded Server Architecture. Thread of execution. Chapter 4
Motivation Threads Chapter 4 Most modern applications are multithreaded Threads run within application Multiple tasks with the application can be implemented by separate Update display Fetch data Spell
More informationAN OVERVIEW OF DISTRIBUTED FILE SYSTEM Aditi Khazanchi, Akshay Kanwar, Lovenish Saluja
www.ijecs.in International Journal Of Engineering And Computer Science ISSN:2319-7242 Volume 2 Issue 10 October, 2013 Page No. 2958-2965 Abstract AN OVERVIEW OF DISTRIBUTED FILE SYSTEM Aditi Khazanchi,
More informationNVMf based Integration of Non-volatile Memory in a Distributed System - Lessons learned
14th ANNUAL WORKSHOP 2018 NVMf based Integration of Non-volatile Memory in a Distributed System - Lessons learned Jonas Pfefferle, Bernard Metzler, Patrick Stuedi, Animesh Trivedi and Adrian Schuepbach
More informationFederated Array of Bricks Y Saito et al HP Labs. CS 6464 Presented by Avinash Kulkarni
Federated Array of Bricks Y Saito et al HP Labs CS 6464 Presented by Avinash Kulkarni Agenda Motivation Current Approaches FAB Design Protocols, Implementation, Optimizations Evaluation SSDs in enterprise
More informationANSE: Advanced Network Services for [LHC] Experiments
ANSE: Advanced Network Services for [LHC] Experiments Artur Barczyk California Institute of Technology Joint Techs 2013 Honolulu, January 16, 2013 Introduction ANSE is a project funded by NSF s CC-NIE
More informationUnderstanding StoRM: from introduction to internals
Understanding StoRM: from introduction to internals 13 November 2007 Outline Storage Resource Manager The StoRM service StoRM components and internals Deployment configuration Authorization and ACLs Conclusions.
More informationAn Evolutionary Path to Object Storage Access
An Evolutionary Path to Object Storage Access David Goodell +, Seong Jo (Shawn) Kim*, Robert Latham +, Mahmut Kandemir*, and Robert Ross + *Pennsylvania State University + Argonne National Laboratory Outline
More informationImproving Application Performance and Predictability using Multiple Virtual Lanes in Modern Multi-Core InfiniBand Clusters
Improving Application Performance and Predictability using Multiple Virtual Lanes in Modern Multi-Core InfiniBand Clusters Hari Subramoni, Ping Lai, Sayantan Sur and Dhabhaleswar. K. Panda Department of
More informationCSE 124: Networked Services Fall 2009 Lecture-19
CSE 124: Networked Services Fall 2009 Lecture-19 Instructor: B. S. Manoj, Ph.D http://cseweb.ucsd.edu/classes/fa09/cse124 Some of these slides are adapted from various sources/individuals including but
More informationTHOUGHTS ABOUT THE FUTURE OF I/O
THOUGHTS ABOUT THE FUTURE OF I/O Dagstuhl Seminar Challenges and Opportunities of User-Level File Systems for HPC Franz-Josef Pfreundt, May 2017 Deep Learning I/O Challenges Memory Centric Computing :
More information- 1 perforce instance for artists, to keep history of source asset files - 1 other perforce instance, in which they export in a platform agnostic
1 2 - 1 perforce instance for artists, to keep history of source asset files - 1 other perforce instance, in which they export in a platform agnostic format, using plugins we provide - Design data also
More information6.5 Collective Open/Close & Epoch Distribution Demonstration
6.5 Collective Open/Close & Epoch Distribution Demonstration Johann LOMBARDI on behalf of the DAOS team December 17 th, 2013 Fast Forward Project - DAOS DAOS Development Update Major accomplishments of
More informationPerformance and Optimization Issues in Multicore Computing
Performance and Optimization Issues in Multicore Computing Minsoo Ryu Department of Computer Science and Engineering 2 Multicore Computing Challenges It is not easy to develop an efficient multicore program
More informationToward portable I/O performance by leveraging system abstractions of deep memory and interconnect hierarchies
Toward portable I/O performance by leveraging system abstractions of deep memory and interconnect hierarchies François Tessier, Venkatram Vishwanath, Paul Gressier Argonne National Laboratory, USA Wednesday
More informationApplication Performance on IME
Application Performance on IME Toine Beckers, DDN Marco Grossi, ICHEC Burst Buffer Designs Introduce fast buffer layer Layer between memory and persistent storage Pre-stage application data Buffer writes
More informationNiraj Kumar Lead Azure Architect, MCT( Microsoft Certified Trainer)
Niraj Kumar Lead Azure Architect, MCT( Microsoft Certified Trainer) Azure Storage Azure Storage Overview Types of Storage Account and performance tiers Storage Replication Scope Types of Storage Managed
More informationEvaluation of Parallel I/O Performance and Energy with Frequency Scaling on Cray XC30 Suren Byna and Brian Austin
Evaluation of Parallel I/O Performance and Energy with Frequency Scaling on Cray XC30 Suren Byna and Brian Austin Lawrence Berkeley National Laboratory Energy efficiency at Exascale A design goal for future
More informationDell EMC Unity: Performance Analysis Deep Dive. Keith Snell Performance Engineering Midrange & Entry Solutions Group
Dell EMC Unity: Performance Analysis Deep Dive Keith Snell Performance Engineering Midrange & Entry Solutions Group Agenda Introduction Sample Period Unisphere Performance Dashboard Unisphere uemcli command
More informationPage 1. Analogy: Problems: Operating Systems Lecture 7. Operating Systems Lecture 7
Os-slide#1 /*Sequential Producer & Consumer*/ int i=0; repeat forever Gather material for item i; Produce item i; Use item i; Discard item i; I=I+1; end repeat Analogy: Manufacturing and distribution Print
More informationI/O at JSC. I/O Infrastructure Workloads, Use Case I/O System Usage and Performance SIONlib: Task-Local I/O. Wolfgang Frings
Mitglied der Helmholtz-Gemeinschaft I/O at JSC I/O Infrastructure Workloads, Use Case I/O System Usage and Performance SIONlib: Task-Local I/O Wolfgang Frings W.Frings@fz-juelich.de Jülich Supercomputing
More informationCampaign Storage. Peter Braam Co-founder & CEO Campaign Storage
Campaign Storage Peter Braam 2017-04 Co-founder & CEO Campaign Storage Contents Memory class storage & Campaign storage Object Storage Campaign Storage Search and Policy Management Data Movers & Servers
More informationDeveloping Microsoft Azure Solutions
Course 20532C: Developing Microsoft Azure Solutions Course details Course Outline Module 1: OVERVIEW OF THE MICROSOFT AZURE PLATFORM This module reviews the services available in the Azure platform and
More informationArchitecting a High Performance Storage System
WHITE PAPER Intel Enterprise Edition for Lustre* Software High Performance Data Division Architecting a High Performance Storage System January 2014 Contents Introduction... 1 A Systematic Approach to
More informationCSE 544: Principles of Database Systems
CSE 544: Principles of Database Systems Anatomy of a DBMS, Parallel Databases 1 Announcements Lecture on Thursday, May 2nd: Moved to 9am-10:30am, CSE 403 Paper reviews: Anatomy paper was due yesterday;
More informationChapter 1: Distributed Information Systems
Chapter 1: Distributed Information Systems Contents - Chapter 1 Design of an information system Layers and tiers Bottom up design Top down design Architecture of an information system One tier Two tier
More informationLustre on ZFS. At The University of Wisconsin Space Science and Engineering Center. Scott Nolin September 17, 2013
Lustre on ZFS At The University of Wisconsin Space Science and Engineering Center Scott Nolin September 17, 2013 Why use ZFS for Lustre? The University of Wisconsin Space Science and Engineering Center
More informationRobinHood Project Update
FROM RESEARCH TO INDUSTRY RobinHood Project Update Robinhood User Group 2016 Thomas Leibovici SEPTEMBER, 19 th 2016 Project update Latest Releases Robinhood 2.5.6 (july 2016)
More information