Performance Exploration: Server Options Impacting OLTP Workloads

This ticket tracks the set of MariaDB server options to be tested for performance impact across OLTP workloads (HammerDB TPROC-C).

The goal is to establish a baseline and measure overhead introduced by durability, replication, charset, flushing, and concurrency-related settings.

All tests will be compared to a baseline using:

• MariaDB default/recommended options
• crash-safe configuration
• log-bin=0
• innodb_flush_log_at_trx_commit=1
• large buffer pool so all data fits in memory

Test workload:

• Suite: HammerDB TPROC-C
• Load: Cached
• Threads: 10, 50, 100, 400
• Warmup: 4 minutes
• Duration: 15 minutes
• Iterations: 3

===============================================================
OPTIONS TO TEST (WITH RATIONALE)
===============================================================

1. innodb_doublewrite
WHY: Crash-safety vs performance.
EXPECTED: OFF is faster, ON is safer.

2. character_set_server (latin1 | utf8 | utf8mb4)
WHY: Charset affects CPU and index width.

3. performance_schema (ON | OFF)
WHY: Instrumentation overhead.

4. binlog_row_image (FULL | MINIMAL | NOBLOB) (requires log-bin=1)
WHY: Controls binlog volume.

5. sync_binlog (0 | 1) (requires log-bin=1)
WHY: Binlog durability cost.

6. log-bin (0 | 1)
WHY: Baseline cost of enabling replication/PITR.

7. thread_handling (one-thread-per-connection | pool-of-threads)
WHY: Thread model scalability.

8. innodb_flush_log_at_trx_commit (0 | 1 | 2 )
WHY: Durability vs performance.

9. innodb_adaptive_hash_index (0 | 1)
WHY: Lookup speed vs contention.

10. query_cache_type (ON | OFF)
WHY: QC mutex contention vs repeated SELECT speed.

11. rpl_semi_sync_master_enable (0 | 1)
WHY: Commit latency vs replication safety.

12. rpl_semi_sync_master_wait_point (AFTER_SYNC | AFTER_COMMIT)
WHY: Different semi-sync wait semantics.

13. transaction_isolation (RU | RC | RR | SERIALIZABLE)
WHY: Locking/MVCC overhead.
NOTE: Must test with log-bin=0 and log-bin=1.

===============================================================
Purpose:
Provide a comprehensive performance map of MariaDB under different durability, replication, charset, flushing, and concurrency configurations. Others may add additional options or combinations here.
===============================================================