0%

Book Description

This IBM® Redpaper introduces the IBM Spectrum® Scale Erasure Code Edition (ECE) as a scalable, high-performance data and file management solution. ECE is designed to run on any commodity server that meets the ECE minimum hardware requirements. ECE provides all the functionality, reliability, scalability, and performance of IBM Spectrum Scale with the added benefit of network-dispersed IBM Spectrum Scale RAID, which provides data protection, storage efficiency, and the ability to manage storage in hyperscale environments that are composed from commodity hardware.

In this publication, we explain the benefits of ECE and the use cases where we believe it fits best. We also provide a technical introduction to IBM Spectrum Scale RAID. Next, we explain the key aspects of planning an installation, provide an example of an installation scenario, and describe the key aspects of day-to-day management and a process for problem determination. We conclude with an overview of possible enhancements that are being considered for future versions of IBM Spectrum Scale Erasure Code Edition.

Overall knowledge of IBM Spectrum Scale Erasure Code Edition is critical to planning a successful storage system deployment. This paper is targeted toward technical professionals (consultants, technical support staff, IT Architects, and IT Specialists) who are responsible for delivering cost effective storage solutions. The goal of this paper is to describe the benefits of using IBM Spectrum Scale Erasure Code Edition for the creation of high performing storage systems.

Table of Contents

  1. Front cover
  2. Notices
    1. Trademarks
  3. Preface
    1. Authors
    2. Now you can become a published author, too!
    3. Comments welcome
    4. Stay connected to IBM Redbooks
  4. Chapter 1. Introduction to IBM Spectrum Scale Erasure Code Edition
    1. 1.1 Overview
    2. 1.2 Value proposition
    3. 1.3 Advantages and key features
      1. 1.3.1 High-performance erasure coding
      2. 1.3.2 Declustered erasure coding
      3. 1.3.3 End-to-end checksum for comprehensive data integrity
      4. 1.3.4 Extreme scalability
      5. 1.3.5 Enterprise storage features and manageability
    4. 1.4 Configuration options
    5. 1.5 Example ECE use cases
      1. 1.5.1 High-performance file serving
      2. 1.5.2 High-performance compute tier
      3. 1.5.3 High capacity data storage
    6. 1.6 Example configuration
    7. 1.7 Summary
  5. Chapter 2. IBM Spectrum Scale Erasure Code Edition use cases
    1. 2.1 High-performance tier for ML/DL and analytics
    2. 2.2 High-performance file serving with CES protocol nodes
    3. 2.3 High-capacity data storage
  6. Chapter 3. IBM Spectrum Scale RAID technical overview
    1. 3.1 Definitions of IBM Spectrum Scale RAID
    2. 3.2 Software RAID
      1. 3.2.1 RAID codes
      2. 3.2.2 Declustered RAID
      3. 3.2.3 Fault-tolerance
    3. 3.3 End-to-end checksum and data versions
    4. 3.4 Integrity Manager
    5. 3.5 Disk hospital
    6. 3.6 Storage hardware software interface
    7. 3.7 IBM Spectrum Scale RAID software component layout
    8. 3.8 Start up sequence for recovery group and log groups
    9. 3.9 Recovery of recovery group and log groups
    10. 3.10 ECE read and write strategies
      1. 3.10.1 Reads
      2. 3.10.2 Full track writes
      3. 3.10.3 Promoted full track writes
      4. 3.10.4 Medium writes
      5. 3.10.5 Small writes
      6. 3.10.6 Deferred writes and stale strips
  7. Chapter 4. Planning an ECE installation
    1. 4.1 Sizing considerations
    2. 4.2 Precheck tools
      1. 4.2.1 SpectrumScale_ECE_OS_READINESS helper tool
      2. 4.2.2 SpectrumScale_ECE_OS_OVERVIEW helper tool
      3. 4.2.3 SpectrumScale_NETWORK_READINESS tool
    3. 4.3 Erasure code selection
    4. 4.4 Spare space allocation
    5. 4.5 Network planning
    6. 4.6 IBM Spectrum Scale node roles
    7. 4.7 Cluster Export Services
    8. 4.8 System management and monitoring
    9. 4.9 Other IBM Spectrum Scale components
    10. 4.10 Running applications
    11. 4.11 File and Object Solution Design Studio tool
  8. Chapter 5. ECE installation procedures
    1. 5.1 Installation overview
    2. 5.2 IBM Spectrum Scale ECE installation prerequisites
      1. 5.2.1 Minimum requirements for ECE
      2. 5.2.2 SSH and network setup
      3. 5.2.3 Repository setup
    3. 5.3 IBM Spectrum Scale ECE installation background
    4. 5.4 IBM Spectrum Scale ECE installation and configuration
  9. Chapter 6. Daily management of ECE storage
    1. 6.1 Drive replacement
      1. 6.1.1 Drive replacement cancellation
    2. 6.2 Replacing nodes
      1. 6.2.1 Node replacement with new drives
      2. 6.2.2 Replacing nodes and preserving drives from the old node
    3. 6.3 Adding nodes
    4. 6.4 Upgrading to a new IBM Spectrum Scale release
    5. 6.5 Upgrading operating system, firmware, driver, and patch
  10. Chapter 7. Problem determination and debugging an ECE system
    1. 7.1 Check whether the ECE nodes are active in the cluster
    2. 7.2 Check whether the recovery groups are active
    3. 7.3 Check for pdisks that are ready for replacement
    4. 7.4 Check for pdisks that are not in OK state
    5. 7.5 Pdisk states
    6. 7.6 Check each recovery group’s event log for messages
    7. 7.7 Using the mmhealth command with ECE
    8. 7.8 System health monitoring use cases
    9. 7.9 Collecting data for problem determination
    10. 7.10 Network tools
  11. Chapter 8. Summary
    1. 8.1 New Deployment Models
      1. 8.1.1 ECE on cloud
      2. 8.1.2 Building a containerized ECE solution
      3. 8.1.3 New erasure codes
    2. 8.2 Conclusion
  12. Related publications
    1. IBM Redbooks
    2. Other publication
    3. Online resources
    4. Help from IBM
  13. Back cover
18.191.213.215