IE3: Immersion Event on High Availability and Disaster Recovery
Overview
This 5-day SQL Server training class is designed to give a solid grounding for database administrators in topics related to high availability and disaster recovery.
The class starts by explaining the methodology for creating a sound HA/DR plan and then covers backup and restore operations in depth. After that, the four major multi-instance technologies (clustering, database mirroring, replication, and availability groups) are covered. Log shipping will be discussed, but not in any great depth. We also cover consolidation and virtualization, both of which play an important role in availability and cost-cutting. Finally we examine database snapshots, and cover consistency checking and recovering from corruptions in depth.
Availability is an area that is critical in today’s environment and all DBAs should be able to design and appropriate availability strategy and coordinate a successful recovery after a disaster.
Instructors: Paul S. Randal, Kimberly L. Tripp, Jonathan Kehayias, Joe Sack
Please see our Immersion Events Schedule for class dates and our comprehensive Immersion Events FAQ for class costs and other frequently asked questions.
Curriculum
Module 1: Designing an Availability Strategy
This module describes a methodical approach to availability strategy planning, rather than the more common “let’s get a cluster” or “let’s try to use the incumbent technology” approaches. Topics covered include:
- Requirements gathering
- Non-technical imitations analysis
- Technical limitations analysis
- Technology evaluation
- Disaster recovery planning considerations
Module 2: Consolidation
In this module we cover the pros and cons of each of the popular methods to put more databases on less hardware. Topics covered include:
- Instance stacking
- Active/active clustering (or multi-instance clustering)
- Virtualization
- Plain old lots-of-databases-on-an-instance
- Planning
- Post-consolidation tools (Resource Governor, Performance Monitor, etc.)
Module 3: Backup Internals and Strategy
Backups are the cornerstone of any availability and disaster recovery strategy. In this module we cover the internals and usage of the three types of backups along with how to combine them into a valid strategy. Topics covered include:
- Full backups
- Log backups
- Differential backups
- File and filegroup strategies for VLDBs
- Backup consistency
- Backup compression
- Backup strategy
Module 4: Restore Internals and Scenarios
Backups are only taken because you want to be able to perform a restore, for whatever reason, but restoring backups is one of the least practiced DBA skills. This module covers the major RESTORE options you’re likely to need when recovering from a disaster. Topics covered include:
- Restore internals
- Completion options for restores
- Point-in-time restores
- Partial restores
- Piecemeal restores
- Restore sequences
- Advanced techniques
Module 5: Database Snapshots
Database snapshots provide a very easy way to protect against user error as well as providing a static copy of the database for reporting etc. This module explains how database snapshots work, as well as some of the problems that can arise from their use. Topics covered include:
- How database snapshots work
- Usage patterns and anti-patterns
- Monitoring
- Troubleshooting
Module 6: Failover Clustering
Clusters have a bad reputation, but since Windows 2008, they’ve turned the corner. They’ve gotten easier to deploy and manage, and we show you how to pull it off according to best practices. Topics covered include:
- Planning
- Active Directory requirements
- Storage and networking setup
- Quorums
- DTC
- Installation
- Performance tuning
- Troubleshooting
Module 7: Database Mirroring
Database mirroring has become immensely popular since its introduction in SQL Server 2005. This module covers configuring database mirroring, failure scenarios, and how to troubleshoot availability and performance. Topics covered include:
- Choosing a mirroring configuration
- Configuring mirroring
- Client redirection
- Failure scenarios
- Monitoring
- Troubleshooting
- Overview of SQL Server 2012 Availability Groups
Module 8: Availability Groups
Availability Groups are the flagship HA technology introduced in SQL Server 2012 as a major upgrade of the capabilities of database mirroring. Topics covered include:
- Availability Group overview
- Readable secondaries
- Application connection support
- Partial database containment
- Migrating from Database Mirroring
- Quorum model and node votes
- Asymmetric storage considerations
Module 9: Replication
Replication is often overlooked as an availability feature but it is useful as well as for providing scale-out. This module briefly examines the various replication types before delving more deeply into transactional and peer-to-peer replication. Topics covered include:
- Replication types and scenarios
- Transactional replication
- Performance considerations
- Combining transactional replication with other technologies
- Peer-to-peer replication
- Monitoring
- Troubleshooting
Module 10: Corruption Detection and Recovery
This module addresses one of the most important areas for a DBA to know: how to detect (inevitable) corruption and how to recover from it, both with and without backups. Topics covered include:
- How does corruption occur
- Detecting corruption
- DBCC CHECKDB and related commands
- Interpreting DBCC CHECKDB output
- Recovering from corruptions without using repair
- Using repair
- Advanced techniques
- Real-life horror stories
Module 11: Virtualization
Whether you’re using Hyper-V, VMware, or Xen, virtualization changes a lot about how you deploy and manage SQL Server. We show how things are different, how you need to compensate, and dive into the new HA/DR options that virtualization offers. Topics covered include:
- CPU
- Memory
- Storage
- Monitoring
- HA/DR options
Module 12: Appendix
This module is not taught. It presents a series of slides that compare the various HA technologies by various aspects.
Questions?
If you have any questions, please contact us.