IBM Spectrum Computing Solutions
Total Page:16
File Type:pdf, Size:1020Kb
Front cover IBM Spectrum Computing Solutions Dino Quintero Daniel de Souza Casali Eduardo Luis Cerdas Moya Federico Fros Maciej Olejniczak Redbooks International Technical Support Organization IBM Spectrum Computing Solutions May 2017 SG24-8373-00 Note: Before using this information and the product it supports, read the information in “Notices” on page vii. First Edition (May 2017) This edition applies to: Red Hat Linux ppc64 Little Endian version 7.2 IBM Spectrum Scale version 4.2.1 IBM Cluster Foundation version v4.2.2 IBM Spectrum Conductor with Spark version 2.2 IBM Spectrum MPI version 10 © Copyright International Business Machines Corporation 2017. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp. Contents Notices . vii Trademarks . viii Preface . ix Authors. ix Now you can become a published author, too . .x Comments welcome. xi Stay connected to IBM Redbooks . xi Chapter 1. Introduction to IBM Spectrum Computing . 1 1.1 Overview . 2 1.2 Big data and resource management . 2 1.3 The new era for high-performance computing (HPC) . 2 1.4 Hybrid cloud bursting . 3 1.5 The big data challenge . 4 1.5.1 Hadoop . 4 1.5.2 Apache Spark . 5 1.5.3 Hadoop Distributed File System (HDFS) . 5 1.5.4 Multi-tenancy. 5 1.6 IBM Spectrum Cluster Foundation . 6 1.7 IBM Spectrum Computing . 6 1.7.1 IBM Spectrum Conductor with Spark . 7 1.7.2 IBM Spectrum LSF . 7 1.7.3 IBM Spectrum Symphony . 7 Chapter 2. IBM Spectrum Computing family . 9 2.1 IBM Software-Defined Infrastructure. 10 2.2 IBM Spectrum Computing . 11 2.2.1 IBM Spectrum LSF . 13 2.2.2 IBM Spectrum Symphony . 14 2.2.3 IBM Spectrum Conductor . 16 2.2.4 IBM Cluster Foundation . 16 2.3 IBM Spectrum Storage . 17 Chapter 3. IBM Spectrum Computing requirements . 19 3.1 IBM Spectrum LSF system requirements . 20 3.1.1 Operating system support. 20 3.1.2 Hardware requirements for the master host . 22 3.1.3 Server host compatibility. 22 3.2 IBM Spectrum Symphony system requirements . 22 3.2.1 The Minimum hardware requirements for IBM Spectrum Symphony Developer Edition V7.2 . 22 3.2.2 Software requirements . 23 3.3 IBM Spectrum Conductor with Spark requirements . 24 3.3.1 Hardware requirements. 24 3.3.2 Software requirements . 26 3.4 Our lab test environment. 27 Chapter 4. IBM Spectrum LSF . 29 © Copyright IBM Corp. 2017. All rights reserved. iii 4.1 IBM Spectrum LSF family overview . 30 4.1.1 IBM Spectrum LSF family offerings. 31 4.1.2 IBM Spectrum LSF optional add-ons . 33 4.2 IBM Spectrum LSF integration with Docker . 35 4.2.1 IBM Spectrum LSF and Docker integration. 36 4.2.2 IBM Spectrum LSF 10.1 solution Docker job support . 37 4.3 IBM Spectrum LSF Data Manager . 39 4.3.1 Concepts and terminology . 40 4.3.2 How IBM Spectrum LSF Data Manager works . 41 4.3.3 Single cluster implementation . 42 4.3.4 MultiCluster implementation . 43 4.4 IBM Spectrum Symphony MapReduce Accelerator for IBM Spectrum LSF . 45 4.4.1 Configure the Apache Hadoop integration . 46 4.4.2 Run a Hadoop application on IBM Spectrum LSF . 47 4.5 IBM Spectrum LSF MultiCluster capability . 48 4.6 Resource connector for IBM Spectrum LSF . 50 Chapter 5. IBM Spectrum Symphony . 53 5.1 IBM Spectrum Symphony overview . 54 5.1.1 Key highlights of IBM Spectrum Symphony . 54 5.1.2 Scalability . 55 5.1.3 IBM Spectrum Symphony MapReduce, YARN, and Docker integration. 56 5.2 IBM Spectrum Symphony editions . 56 5.3 IBM Spectrum Symphony MultiCluster . 58 5.3.1 Cluster management. 58 5.3.2 MultiCluster master cluster . 59 5.3.3 MultiCluster silo clusters . 61 5.3.4 MultiCluster roles . 61 5.4 Multidimensional schedule . 62 5.5 Multitenancy infractructure . 63 5.5.1 The narrow view of multitenancy. 63 5.5.2 Advantages and challenges . 64 5.5.3 Multitenant designs ..