Capacity Planning

Scale Enterprise Workloads

Value Proposition Achieve reliable multi-agent concurrency through optimized connection pooling, efficient schema caching, and tailored payloads that support high throughput. Read the full value proposition.

1. 🛜 Connection Pooling

• Connection lifecycle: mysql2 keeps idle connections alive. There is no idle timeout in the pool. Connections persist until the server process exits or MySQL closes them via wait_timeout.

Tip

Set --pool-size to 2× the expected concurrent AI tool calls. For a single-agent setup, the default of 10 provides ample headroom.

2. ⚡ Schema Caching

The server caches schema metadata in memory. This reduces redundant INFORMATION_SCHEMA queries.

• Default TTL: 30000 ms (30 seconds), controlled via METADATA_CACHE_TTL_MS.
• Footprint: Moderate schema metadata consumes minimal memory (e.g., a few megabytes).
• Invalidation: DDL tools automatically invalidate the cache upon execution.

Recommended TTLs

Environment	TTL	Rationale
Production (stable schema)	300000 (5 min) or higher	Eliminates introspection overhead during AI reasoning
Active development	5000–30000 (5–30 s)	Keeps AI in sync with frequent schema changes
Migration runs	0 (disabled)	Guarantees fresh metadata after every DDL statement

export METADATA_CACHE_TTL_MS=300000

3. 🏥 Database Maintenance

As AI agents modify data, InnoDB tables may accumulate fragmentation. Regular maintenance workflows help preserve throughput.

OPTIMIZE TABLE

Because InnoDB does not automatically reclaim disk space from deleted rows, optimize your storage footprint by using OPTIMIZE TABLE to seamlessly rebuild tables and indexes, effectively defragmenting your data files.

• When to use: After large bulk deletes, archival operations, or significant churn.
• How: Use the mysql_write_query tool to execute OPTIMIZE TABLE.
• Note: OPTIMIZE TABLE locks the table. It remains I/O-intensive. Schedule this during low-traffic windows.

ANALYZE TABLE

• When to use: After bulk loads that change data distribution significantly. Stale statistics cause the query optimizer to choose suboptimal indexes.
• How: Use the mysql_write_query tool to execute ANALYZE TABLE.
• Note: For MySQL, consider using histogram statistics. Use ANALYZE TABLE ... UPDATE HISTOGRAM ON .... Do this for columns with skewed distributions.

Partitioned Tables

Consider range partitioning for exceptionally large tables or time-series segments. Use the mysql_write_query tool to manage them. Benefits:

• Partition pruning reduces scan scope for time-bounded queries.
• ALTER TABLE ... DROP PARTITION is instant. This is compared to DELETE FROM ... WHERE date < X.

4. 🚀 Buffer Pool Optimization

The InnoDB buffer pool is MySQL's primary memory cache. Its size directly impacts query performance and supports enterprise-scale AI data operations.

• Monitoring: Use mysql_read_query to query performance metrics. Inspect hit rates, dirty page ratios, and free buffers.
• Sizing rule of thumb: Set innodb_buffer_pool_size to 70–80% of available RAM on a dedicated MySQL server.
• Hit rate target: A hit rate below 99% indicates a pool too small for the working set. Use mysql_read_query to review this metric.

Note

Administrators should use the pre-configured Grafana dashboard. This aids ongoing capacity planning. It visualizes buffer pool stats. This reduces constant manual polling.

5. 🪙 Token Efficiency

• Cost Optimization: Transmitting raw query results to LLMs incurs scaling limits and cost bottlenecks. Code Mode addresses this by aggregating data server-side to reduce context size and token consumption.
• Default limit: Granular tool-level arguments efficiently enforce row limits on read queries, delivering precise control over context windows without relying on restrictive global server defaults. Code Mode strictly enforces a global payload boundary (CODE_MODE_MAX_RESULT_SIZE).
• Cursor pagination: Leverage keyset-based cursor pagination for scanning large tables instead of traditional OFFSET, which inefficiently forces MySQL to scan and discard rows. Cursor pagination executes in optimal O(1) time on indexed columns, ensuring consistent performance regardless of table size.
• Token-saving flags: Many tools support compact, summary, and limit flags. These provide a significant reduction in token overhead. Truncating tools return limited and totalAvailable flags. This informs agents about capped results.

---

See also: Performance-Tuning · Configuration · Tool Filtering