Sponsored by:

Table of Contents:

Introduction

1. Defining Availability

2. Redundancy: The Key to Carrier-Grade Availability

3. Preserving Application State: Two Techniques

4. Highly Available Database Managers

5. A Sidebar on How an RDBMS Works

6. Understanding Safeness Levels

7. Data Loss Considerations

8. MTTR Considerations

9. Application Throughput Considerations

10. Three-Way Tradeoff Summary

Self-Test

Test Answers

Glossary

PDF of this tutorial

Differences in safeness levels are defined by the moments that the log records of the active and standby databases are written to disk.

In a 1-Safe system, only the log of the active database is written before the application sees that the transaction has committed. This is shown in Figure 8.

Figure 8: 1-Safe Timeline

In a 2-Safe Received system, the log of the active database is written to disk, and the active database receives notification that the log record has been received by the standby database before the application sees that the transaction has committed. This is shown in Figure 9.

Figure 9: 2-Safe Received Timeline

In a 2-Safe Durable system, the log records of both databases are written to disk before the application sees that the transaction has committed. This is shown in Figure 10.

Figure 10: 2-Safe Durable Timeline

The table in Figure 11 shows for each safeness level the condition of the two databases after the application sees that a transaction has committed.

These three techniques offer different data safeness (because data is stored more or less reliably), different MTTR (because the amount of work that the standby database must perform to be ready for business varies), and different application speed (because more or less activity tales place asynchronously).