10x Performance Improvements
31 likes3,907 views
This document describes 10 steps to achieve a 10x performance improvement for a MySQL database for a social media website. The steps include monitoring the database, identifying slow SQL queries, analyzing problem queries, improving indexes, offloading read load to slaves, improving SQL, using optimal storage engines, and implementing caching. Key actions include installing monitoring tools, using mk-query-digest to analyze SQL, adding indexes, configuring InnoDB, converting tables to InnoDB, and caching query results and content with Memcache. The goal is to optimize the database infrastructure and queries to handle the load of a growing web 2.0 application.
![3. Analyze Problem SQL
• Query Execution Plan (QEP)
• EXPLAIN [EXTENDED] SELECT ...
• Table/Index Structure
• SHOW CREATE TABLE <tablename>G
• Table Statistics
• SHOW TABLE STATUS LIKE ‘<tablename>’G](https://image.slidesharecdn.com/10x-improvement-mysql-100419105218-phpapp02/85/10x-Performance-Improvements-21-320.jpg)
10x Performance Improvements
- 1. 10x Performance Improvements in 10 steps A MySQL Case Study Ronald Bradford http://ronaldbradford.com MySQL Users Conference - 2010.04
- 2. Application Typical Web 2.0 social media site (Europe based) • Users - Visitors, Free Members, Paying Members • Friends • User Content - Video, Pictures • Forums, Chat, Email
- 3. Server Environment • 1 Master Database Server (MySQL 5.0.x) • 3 Slave Database Servers (MySQL 5.0.x) • 5 Web Servers (Apache/PHP) • 1 Static Content Server (Nginx) • 1 Mail Server
- 4. Step 1 Monitor, Monitor, Monitor
- 5. 1. Monitor, Monitor, Monitor • What’s happened? • What’s happening now? • What’s going to happen? Past, Present, Future
- 6. 1. Monitor, Monitor, Monitor Action 1 Monitoring Software • Installation of Cacti http://www.cacti.net/ - • Installation of MySQL Cacti Templates - http://code.google.com/p/mysql-cacti-templates/ • (Optional) Installation of MONyog - http://www.webyog.com/
- 7. 1. Monitor, Monitor, Monitor Action 2 Custom Dashboard • Most important - The state of NOW • Single Page Alerts - GREEN YELLOW RED
- 8. Screen print goes here Dashboard Example
- 9. 1. Monitor, Monitor, Monitor Action 3 Alerting Software • Installation of Nagios http://www.nagios.org/ - • MONyog also has some DB specific alerts
- 10. 1. Monitor, Monitor, Monitor Action 4 Application Metrics • Total page generation time
- 11. Step 2 Identify problem SQL
- 12. 2. Identify Problem SQL Identify SQL Statements • Slow Query Log • Processlist • Binary Log • Status Statistics
- 13. 2. Identify Problem SQL Problems • Sampling • Granularity Solution • tcpdump + mk-query-digest
- 14. 2. Identify Problem SQL Action 1 • Install maatkit - http://www.maatkit.org • Install OS tcpdump (if necessary) • Get sudo access to tcpdump http://ronaldbradford.com/blog/take-a-look-at-mk-query-digest-2009-10-08/
- 15. # Rank Query ID Response time Calls R/Call Item # ==== ================== ================ ======= ========== ==== # 1 0xB8CE56EEC1A2FBA0 14.0830 26.8% 78 0.180552 SELECT c u # 2 0x195A4D6CB65C4C53 6.7800 12.9% 257 0.026381 SELECT u # 3 0xCD107808735A693C 3.7355 7.1% 8 0.466943 SELECT c u # 4 0xED55DD72AB650884 3.6225 6.9% 77 0.047046 SELECT u # 5 0xE817EFFFF5F6FFFD 3.3616 6.4% 147 0.022868 SELECT UNION c # 6 0x15FD03E7DB5F1B75 2.8842 5.5% 2 1.442116 SELECT c u # 7 0x83027CD415FADB8B 2.8676 5.5% 70 0.040965 SELECT c u # 8 0x1577013C472FD0C6 1.8703 3.6% 61 0.030660 SELECT c # 9 0xE565A2ED3959DF4E 1.3962 2.7% 5 0.279241 SELECT c t u # 10 0xE15AE2542D98CE76 1.3638 2.6% 6 0.227306 SELECT c # 11 0x8A94BB83CB730494 1.2523 2.4% 148 0.008461 SELECT hv u # 12 0x959C3B3A967928A6 1.1663 2.2% 5 0.233261 SELECT c t u # 13 0xBC6E3F701328E95E 1.1122 2.1% 4 0.278044 SELECT c t u
- 16. # Query 2: 4.94 QPS, 0.13x concurrency, ID 0x195A4D6CB65C4C53 at byte 4851683 # This item is included in the report because it matches --limit. # pct total min max avg 95% stddev median # Count 3 257 # Exec time 10 7s 35us 492ms 26ms 189ms 78ms 332us # Time range 2009-10-16 11:48:55.896978 to 2009-10-16 11:49:47.760802 # bytes 2 10.75k 41 43 42.85 42.48 0.67 42.48 # Errors 1 none # Rows affe 0 0 0 0 0 0 0 0 # Warning c 0 0 0 0 0 0 0 0 # Query_time distribution # 1us # 10us # # 100us ################################################################ # 1ms #### # 10ms ### # 100ms ######## # 1s # 10s+ # Tables # SHOW TABLE STATUS LIKE 'u'G # SHOW CREATE TABLE `u`G # EXPLAIN SELECT ... FROM u ...G
- 17. 2. Identify Problem SQL Action 2 • Wrappers to capture SQL • Re-run on single/multiple servers • e.g. Different slave configurations
- 18. 2. Identify Problem SQL Tip • Enable General Query Log in Development/Testing • Great for testing Batch Jobs
- 19. 2. Identify Problem SQL Action 3 Application Logic • Show total master/slave SQL statements executed • Show all SQL with execution time (admin user only) Tip • Have abstracted class/method to execute ALL SQL
- 20. Step 3 Analyze problem SQL
- 21. 3. Analyze Problem SQL • Query Execution Plan (QEP) • EXPLAIN [EXTENDED] SELECT ... • Table/Index Structure • SHOW CREATE TABLE <tablename>G • Table Statistics • SHOW TABLE STATUS LIKE ‘<tablename>’G
- 22. 3. Analyze Problem SQL Good mysql> EXPLAIN SELECT id FROM example_table WHERE id=1G *************************** 1. row *************************** id: 1 select_type: SIMPLE table: example_table type: const possible_keys: PRIMARY key: PRIMARY key_len: 4 ref: const rows: 1 Extra: Using index
- 23. 3. Analyze Problem SQL Bad mysql> EXPLAIN SELECT * FROM example_tableG *************************** 1. row *************************** id: 1 select_type: SIMPLE table: example_table type: ALL possible_keys: NULL key: NULL key_len: NULL ref: NULL rows: 659 Extra:
- 24. 3. Analyze Problem SQL Tip • SQL Commenting • Identify batch statement SQL • Identify cached SQL SELECT /* Cache: 10m */ .... SELECT /* Batch: EOD report */ ... SELECT /* Func: 123 */ ....
- 25. Step 4 The Art of Indexes
- 26. 4. The Art of Indexes • Different Types • Column • Concatenated • Covering • Partial http://ronaldbradford.com/blog/understanding-different-mysql-index-implementations-2009-07-22/
- 27. 4. The Art of Indexes Action 1 • EXPLAIN Output • Possible keys • Key used • Key length • Using Index
- 28. 4. The Art of Indexes Tip • Generally only 1 index used per table • Make column NOT NULL when possible • Statistics affect index choices • Storage engines affect operations
- 29. Before (7.88 seconds) After (0.04 seconds) *************************** 2. row ** *************************** 2. row *** id: 2 id: 2 select_type: DEPENDENT SUBQUERY select_type: DEPENDENT SUBQUERY table: h_p table: h_p type: ALL type: index_subquery possible_keys: NULL possible_keys: UId key: NULL key: UId key_len: NULL key_len: 4 ref: NULL ref: func rows: 33789 rows: 2 Extra: Using where Extra: Using index ALTER TABLE h_p ADD INDEX (UId);
- 30. mysql> explain SELECT UID, FUID, COUNT(*) AS Count FROM f GROUP BY UID, FUID ORDER BY Count DESC LIMIT 2000G *************************** 1. row *************************** id: 1 select_type: SIMPLE table: f type: index possible_keys: NULL key: UID key_len: 8 ref: NULL rows: 2151326 Extra: Using index; Using temporary; Using filesort ALTER TABLE f DROP INDEX UID, ADD INDEX (UID,FUID)
- 31. 4. The Art of Indexes Indexes can hurt performance
- 32. Step 5 Offloading Master Load
- 33. 5. Offloading Master Load • Identify statements for READ ONLY slave(s) • e.g. Long running batch statements Single point v scalable solution
- 34. Step 6 Improving SQL
- 35. 6. Improving SQL • Poor SQL Examples • ORDER BY RAND() • SELECT * • Lookup joins • ORDER BY The database is best for storing and retrieving data not logic
- 36. Step 7 Storage Engines
- 37. 7. Storage Engines • MyISAM is default • Table level locking • Concurrent SELECT statements • INSERT/UPDATE/DELETE blocked by long running SELECT • All SELECT’s blocked by INSERT/UPDATE/DELETE • Supports FULLTEXT
- 38. 7. Storage Engines • InnoDB supports transactions • Row level locking with MVCC • Does not support FULLTEXT • Different memory management • Different system variables
- 39. 7. Storage Engines • There are other storage engines • Memory • Archive • Blackhole • Third party
- 40. 7. Storage Engines Using Multiple Engines • Different memory management • Different system variables • Different monitoring • Affects backup strategy
- 41. 7. Storage Engines Action 1 • Configure InnoDB correctly • innodb_buffer_pool_size • innodb_log_file_size • innodb_flush_log_at_trx_commit
- 42. 7. Storage Engines Action 2 • Converted the two primary tables • Users • Content Locking eliminated
- 43. Step 8 Caching
- 44. 8. Caching Action 1 • Memcache is your friend http://memcached.org/ - • Cache query results • Cache lookup data (eliminate joins) • Cache aggregated per user information • Caching Page Content • Top rated (e.g. for 5 minutes)
- 45. 8. Caching Action 2 • MySQL has a Query Cache • Determine the real benefit • Turn on or off dynamically • SET GLOBAL query_cache_size = 1024*1024*32;
- 46. 8. Caching Tip The best performance improvement for an SQL statement is to eliminate it.
- 47. Step 9 Sharding
- 48. 9. Sharding • Application level horizontal and vertical partitioning • Vertical Partitioning • Grouping like structures together (e.g. logging, forums) • Horizontal Partitioning • Affecting a smaller set of users (i.e. not 100%)
- 49. 9. Sharding Action 1 • Separate Logging • Reduced replication load on primary server
- 50. Step 10 Database Management
- 51. 10. Database Management Database Maintenance • Adding indexes (e.g. ALTER) • OPTIMIZE TABLE • Archive/purging data (e.g DELETE) Blocking Operations
- 52. 10. Database Maintenance Action 1 • Automate slave inclusion/exclusion • Ability to apply DB changes to slaves • Master still a problem
- 53. 10. Database Maintenance Action 2 • Install Fail-Over Master Server • Slave + Master features • Master extra configuration • Scripts to switch slaves • Scripts to enable/disable Master(s) • Scripts to change application connection
- 54. 10. Database Maintenance Higher Availability & Testing Disaster Recovery
- 56. 11. Front End Improvements • Know your total website load time http://getfirebug.com/ - • How much time is actually database related? • Reduce HTML page size - 15% improvement • Remove full URL’s, inline css styles • Reduce/combine css & js files • Identify blocking elements (e.g. js)
- 57. 11. Front End Improvements • Split static content to different ServerName • Spread static content over multiple ServerNames (e.g. 3) • Sprites - Combining lightweight images http://spriteme.org/ - • Cookie-less domain name for static content
- 58. Conclusion
- 59. Before • Users experienced slow or unreliable load times • Management could observe, but no quantifiable details • Concern over load for increased growth • Release of some new features on hold
- 60. Now • Users experienced consistent load times (~60ms) • Quantifiable and visible real-time results • Far greater load now supported (Clients + DB) • Better testability and verification for scaling • New features can be deployed
- 61. Consulting Available Now http://ronaldbradford.com
Editor's Notes
- #8: Server (Ping, Uptime) Database Processlist (Total/Active/Idle/Locked) Database Replication (Up,Lag) Web Server (Page Load x 3, Page Size, http processes)
- #13: Traditional Existing Methods
- #29: InnoDb, the PK value is stored
- #32: Too many indexes Disk volume Slows DML Index with unused columns
- #36: Text fields network bandwidth temporary tables filesort
- #52: Site is effectively unavailable
- #53: DB Changes * Add indexes * Change Storage Engines * Change Table structure