An Update on Database Optimization and Statistics Aggregation

Excerpt

Following up on our recent announcement, we are sharing real-world results from our new database optimization tool, Condense. Testing on actual production databases achieved structural volume reductions between 79% and 99%. Continue reading to learn how this automated aggregation improves reporting speeds and backup efficiency across our hosted Revive Adserver installations.

Introduction

A few weeks ago, we announced a new process to optimize database sizes and reporting performance by aggregating long term statistics by month or by day. We collaborated with the core developers of the Revive Adserver Download edition and the team at Revive Adserver Hosted edition to develop a long-term statistics aggregation tool. This new feature was officially released on June 3rd, 2026, as part of Revive Adserver v6.0.7.

How the aggregation works

The tool allows us to define specific aggregation profiles. A typical profile looks like this:

• Older than 12 months: Statistics are combined (aggregated) by month.
• Between 3 and 12 months old: Statistics are combined by day.
• Less than 3 months old: Statistics are left as they are, maintaining full hourly detail.

The tool fully automates the process to ensure it runs quickly and safely, avoiding the risks of manual database manipulation.

Real world results

During testing with actual customer databases, the tool delivered impressive results that exceeded our expectations:

• Test 1: We applied the tool to a database using the profile described above (monthly aggregation for data older than 12 months, daily for data between 3 and 12 months). After completion, the database shrank by approximately 99%.
• Test 2: We applied a less aggressive profile to a database (monthly aggregation for data older than 24 months, daily for data between 12 and 24 months). The database size was reduced by about 79%.
• Test 3: We applied the same less aggressive profile to a much larger database, with statistics going back over a decade. The database size was reduced by about 89%.

Expected benefits

Shrinking the database sizes provides several clear advantages for your ad server platform:

• Faster Reporting: Generating reports will be faster. Because database queries finish more quickly, system resources are freed up faster, improving overall performance for all users.
• Efficient Backups: Smaller databases mean quicker backup cycles. This reduces the load on our infrastructure and may allow us to back up your data even more frequently than we do now.
• Optimized Background Processing: During our testing of the tool (code-named Condense), we found that it runs remarkably fast. It uses “batch processing” to ensure the database server is never overloaded during the clean-up process.

Implementation plan

Given these benefits, and following positive feedback from our customers, we will begin optimizing all hosted Revive Adserver databases over the coming weeks.

Initially, our team will run the aggregation process manually to monitor execution times and database performance. Once we confirm everything operates smoothly, we will schedule the process to run automatically once a month.

Because we manage a large number of installations and some databases are exceptionally large, this initial “catch-up” phase will be spread out over a few weeks. After this initial optimization is complete, the tool will run automatically in the background once a month.