You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
You'll learn how to use a measurement-driven approach to tune MySQL on Arm-based platforms.
34
+
You'll explore system factors — storage technology and file systems,
35
+
disk scheduling, kernel memory management, compiler, and library versions — that you can adjust. In addition, you'll focus on MySQL parameters related to connection
36
+
handling, memory usage, disk flush behavior, and concurrency, and learn how to enable and
37
+
size huge pages based on the InnoDB buffer pool. By the end, you'll know what parameters to update for running controlled experiments,
38
+
and be able to make persistent configuration choices aligned with your workload.
39
+
faqs:
40
+
- question: How do I know a MySQL tuning change actually helped?
41
+
answer: >-
42
+
Run the same repeatable workload before and after the change and compare throughput, latency,
43
+
and profiles. Change one parameter at a time or use a designed experiment so results are
44
+
attributable to specific settings.
45
+
- question: Should I set MySQL parameters in an option file or on the mysqld command line?
46
+
answer: >-
47
+
Use an option file for persistent tuning so changes are reviewable, version controlled,
48
+
and applied on restart. The examples target the [mysqld] group; command-line flags are suitable
49
+
for temporary tests.
50
+
- question: Which storage option should I use when testing performance?
51
+
answer: >-
52
+
In general, locally attached SSD storage performs best, but network-based storage can also
53
+
perform well. Test the storage technologies and file systems you have, and review disk scheduling
54
+
behavior with your workload.
55
+
- question: When should I look at kernel, compiler, or library choices instead of MySQL settings?
56
+
answer: >-
57
+
Evaluate them before or alongside MySQL tuning because operating system settings, kernel
58
+
memory management, compiler choices, and library versions can affect throughput and latency.
59
+
Treat them as part of the same performance experiment set.
60
+
- question: How should I size huge pages for MySQL?
61
+
answer: >-
62
+
Enable huge pages for MySQL and size them based on the InnoDB buffer pool. This aligns page
63
+
allocation with the primary memory consumer in typical MySQL deployments.
0 commit comments