查询的MySQL性能问题|优化最佳实践？

我们得到了一个4GB内存和MySQL 5.1.41的Linux服务器。我们在MySQL上有默认设置。我们运行类似于Twitter的实时社交网络服务。数据库由1,4百万条推文条目组成。我们的网站和桌面客户端每5秒获取一次新条目发送查询。我们现在正在测试少数客户端的系统，但它挂断了SQL - 每个请求都会持续15秒。如果删除“order by” - 时间减少一半，但仍然太慢，我们期望有几百个客户端，这可能会挂起实例。来自Mysql性能专家的任何建议？查询的MySQL性能问题|优化最佳实践？

`SELECT 
            A.ID id, 
            A.msg, 
            A.pic, 
            A.time_added, 
            A.public, 
            A.twitter_status_id, 
            A.twitter_update_type, 
            A.USERID userid, 
            A.pic_host, 
            B.ppic_host, 
            B.username, 
            B.profilepicture 
            FROM posts A, members B WHERE 

(A.channel_id IN (489, 109, 490, 4, 493, 494, 495, 496, 497, 452, 80, 53, 198, 834, 498, 500, 501, 852, 505, 506, 507, 508, 509, 510, 511, 778, 515, 132, 517, 519, 931, 932, 934, 521, 523, 524, 780, 569, 527, 12, 529, 530, 532, 533, 534, 936, 938, 536, 537, 708, 940, 704) 

AND A.USERID=B.USERID 
AND A.type='update' 
AND A.public=1 
AND A.TIME_ADDED>1305543735) 

order by A.TIME_ADDED 
desc limit 50; 


mysql> EXPLAIN 
    -> 
    -> SELECT 
    ->         A.ID id, 
    ->         A.msg, 
    ->         A.pic, 
    ->         A.time_added, 
    ->         A.public, 
    ->         A.twitter_status_id, 
    ->         A.twitter_update_type, 
    ->         A.USERID userid, 
    ->         A.pic_host, 
    ->         B.ppic_host, 
    ->         B.username, 
    ->         B.profilepicture 
    ->         FROM posts A, members B WHERE 
    -> 
    -> (A.channel_id IN (489, 109, 490, 4, 493, 494, 495, 496, 497, 452, 80, 53, 198, 834, 498, 500, 501, 852, 505, 506, 507, 508, 509, 510, 511, 778, 515, 132, 517, 519, 931, 932, 934, 521, 523, 524, 780, 569, 527, 12, 529, 530, 532, 533, 534, 936, 938, 536, 537, 708, 940, 704) 
    -> 
    -> AND A.USERID=B.USERID 
    -> AND A.type='update' 
    -> AND A.public=1 
    -> AND A.TIME_ADDED>1305543735) 
    -> 
    -> order by A.TIME_ADDED 
    -> desc limit 50; 
+----+-------------+-------+--------+-----------------------------------------------+------------+---------+------------------------+------+-------------+ 
| id | select_type | table | type | possible_keys         | key  | key_len | ref     | rows | Extra  | 
+----+-------------+-------+--------+-----------------------------------------------+------------+---------+------------------------+------+-------------+ 
| 1 | SIMPLE  | A  | range | USERID,type,time_added,userid+type,channel_id | time_added | 5  | NULL     | 18 | Using where | 
| 1 | SIMPLE  | B  | eq_ref | PRIMARY,USERID        | PRIMARY | 8  | twitter_devel.A.USERID | 1 |    | 
+----+-------------+-------+--------+-----------------------------------------------+------------+---------+------------------------+------+-------------+ 
2 rows in set (0.00 sec) 


    DDL for table posts: 

CREATE TABLE `posts` (
    `ID` bigint(20) NOT NULL AUTO_INCREMENT, 
    `type` varchar(10) NOT NULL, 
    `USERID` bigint(20) NOT NULL, 
    `PID` bigint(20) NOT NULL, 
    `UID` bigint(20) NOT NULL, 
    `UIDO` bigint(20) NOT NULL, 
    `channel_id` bigint(20) DEFAULT NULL, 
    `msg` text NOT NULL, 
    `pic` varchar(20) NOT NULL, 
    `pic_host` varchar(40) DEFAULT NULL, 
    `pdf_host` varchar(40) DEFAULT NULL, 
    `time_added` int(10) DEFAULT NULL, 
    `views` bigint(20) NOT NULL, 
    `reply` bigint(20) NOT NULL, 
    `edited` varchar(20) NOT NULL, 
    `public` int(11) NOT NULL DEFAULT '1', 
    `twitter_status_id` bigint(50) DEFAULT NULL, 
    `twitter_update_type` varchar(10) DEFAULT NULL, 
    PRIMARY KEY (`ID`), 
    UNIQUE KEY `ID` (`ID`), 
    KEY `USERID` (`USERID`), 
    KEY `type` (`type`), 
    KEY `UID` (`UID`), 
    KEY `time_added` (`time_added`), 
    KEY `userid+type` (`type`,`USERID`) USING BTREE, 
    KEY `pic` (`pic`), 
    KEY `channel_id` (`channel_id`), 
    FULLTEXT KEY `msg` (`msg`) 
) ENGINE=MyISAM AUTO_INCREMENT=1591683 DEFAULT CHARSET=utf8; 

    auto_increment_increment 2 
auto_increment_offset 2 
autocommit ON 
automatic_sp_privileges ON 
back_log 50 
basedir /usr/ 
big_tables OFF 
binlog_cache_size 32768 
binlog_format MIXED 
bulk_insert_buffer_size 8388608 
character_set_client utf8 
character_set_connection utf8 
character_set_database utf8 
character_set_filesystem binary 
character_set_results utf8 
character_set_server utf8 
character_set_system utf8 
character_sets_dir /usr/share/mysql/charsets/ 
collation_connection utf8_general_ci 
collation_database utf8_general_ci 
collation_server utf8_unicode_ci 
completion_type 0 
concurrent_insert 1 
connect_timeout 10 
datadir /var/lib/mysql/ 
date_format %Y-%m-%d 
datetime_format %Y-%m-%d %H:%i:%s 
default_week_format 0 
delay_key_write ON 
delayed_insert_limit 100 
delayed_insert_timeout 300 
delayed_queue_size 1000 
div_precision_increment 4 
engine_condition_pushdown ON 
error_count 0 
event_scheduler OFF 
example_enum_var e1 
example_ulong_var 8 
expire_logs_days 10 
flush OFF 
flush_time 0 
foreign_key_checks ON 
ft_boolean_syntax + -><()~*:""&| 
ft_max_word_len 84 
ft_min_word_len 4 
ft_query_expansion_limit 20 
ft_stopword_file (built-in) 
general_log OFF 
general_log_file /var/run/mysql/mysqld.log 
group_concat_max_len 1024 
have_community_features YES 
have_compress YES 
have_crypt YES 
have_csv YES 
have_dynamic_loading YES 
have_geometry YES 
have_innodb YES 
have_ndbcluster DISABLED 
have_openssl DISABLED 
have_partitioning YES 
have_query_cache YES 
have_rtree_keys YES 
have_ssl DISABLED 
have_symlink YES 
hostname vteu02.veritweet.com 
identity 0 
ignore_builtin_innodb OFF 
init_connect  
init_file 
init_slave 
innodb_adaptive_hash_index ON 
innodb_additional_mem_pool_size 1048576 
innodb_autoextend_increment 8 
innodb_autoinc_lock_mode 1 
innodb_buffer_pool_size 8388608 
innodb_checksums ON 
innodb_commit_concurrency 0 
innodb_concurrency_tickets 500 
innodb_data_file_path ibdata1:10M:autoextend 
innodb_data_home_dir  
innodb_doublewrite ON 
innodb_fast_shutdown 1 
innodb_file_io_threads 4 
innodb_file_per_table OFF 
innodb_flush_log_at_trx_commit 1 
innodb_flush_method 
innodb_force_recovery 0 
innodb_lock_wait_timeout 50 
innodb_locks_unsafe_for_binlog OFF 
innodb_log_buffer_size 1048576 
innodb_log_file_size 5242880 
innodb_log_files_in_group 2 
innodb_log_group_home_dir ./ 
innodb_max_dirty_pages_pct 90 
innodb_max_purge_lag 0 
innodb_mirrored_log_groups 1 
innodb_open_files 300 
innodb_rollback_on_timeout OFF 
innodb_stats_on_metadata ON 
innodb_support_xa ON 
innodb_sync_spin_loops 20 
innodb_table_locks ON 
innodb_thread_concurrency 8 
innodb_thread_sleep_delay 10000 
innodb_use_legacy_cardinality_algorithm ON 
insert_id 0 
interactive_timeout 28800 
join_buffer_size 131072 
keep_files_on_create OFF 
key_buffer_size 16777216 
key_cache_age_threshold 300 
key_cache_block_size 1024 
key_cache_division_limit 100 
language /usr/share/mysql/english/ 
large_files_support ON 
large_page_size 0 
large_pages OFF 
last_insert_id 0 
lc_time_names en_US 
license GPL 
local_infile ON 
locked_in_memory OFF 
log OFF 
log_bin ON 
log_bin_trust_function_creators OFF 
log_bin_trust_routine_creators OFF 
log_error /var/log/mysql/mysqld.log 
log_output FILE 
log_queries_not_using_indexes ON 
log_slave_updates OFF 
log_slow_queries ON 
log_warnings 1 
long_query_time 10.000000 
low_priority_updates OFF 
lower_case_file_system OFF 
lower_case_table_names 0 
max_allowed_packet 1048576 
max_binlog_cache_size 18446744073709547520 
max_binlog_size 524288000 
max_connect_errors 10 
max_connections 151 
max_delayed_threads 20 
max_error_count 64 
max_heap_table_size 16777216 
max_insert_delayed_threads 20 
max_join_size 18446744073709551615 
max_length_for_sort_data 1024 
max_prepared_stmt_count 16382 
max_relay_log_size 0 
max_seeks_for_key 18446744073709551615 
max_sort_length 1024 
max_sp_recursion_depth 0 
max_tmp_tables 32 
max_user_connections 0 
max_write_lock_count 18446744073709551615 
min_examined_row_limit 0 
multi_range_count 256 
myisam_data_pointer_size 6 
myisam_max_sort_file_size 9223372036853727232 
myisam_recover_options OFF 
myisam_repair_threads 1 
myisam_sort_buffer_size 8388608 
myisam_stats_method nulls_unequal 
myisam_use_mmap OFF 
ndb_autoincrement_prefetch_sz 1 
ndb_batch_size 32768 
ndb_cache_check_time 0 
ndb_connectstring 
ndb_extra_logging 1 
ndb_force_send ON 
ndb_index_stat_cache_entries 32 
ndb_index_stat_enable OFF 
ndb_index_stat_update_freq 20 
ndb_log_binlog_index ON 
ndb_log_update_as_write ON 
ndb_log_updated_only ON 
ndb_optimization_delay 10 
ndb_optimized_node_selection 3 
ndb_report_thresh_binlog_epoch_slip 3 
ndb_report_thresh_binlog_mem_usage 10 
ndb_table_no_logging OFF 
ndb_table_temporary OFF 
ndb_use_copying_alter_table OFF 
ndb_use_exact_count OFF 
ndb_use_transactions ON 
ndbinfo_database ndbinfo 
ndbinfo_max_bytes 0 
ndbinfo_max_rows 10 
ndbinfo_show_hidden OFF 
ndbinfo_table_prefix ndb$ 
ndbinfo_version 458765 
net_buffer_length 8192 
net_read_timeout 30 
net_retry_count 10 
net_write_timeout 60 
new OFF 
old OFF 
old_alter_table OFF 
old_passwords OFF 
open_files_limit 1024 
optimizer_prune_level 1 
optimizer_search_depth 62 
optimizer_switch index_merge=on,index_merge_union=on,index_merge_sort_union=on,index_merge_intersection=on 
pid_file /var/run/mysql/mysqld.pid 
plugin_dir /usr/lib64/mysql/plugin 
port 3306 
preload_buffer_size 32768 
profiling ON 
profiling_history_size 15 
protocol_version 10 
pseudo_thread_id 304885 
query_alloc_block_size 8192 
query_cache_limit 1048576 
query_cache_min_res_unit 4096 
query_cache_size 0 
query_cache_type ON 
query_cache_wlock_invalidate OFF 
query_prealloc_size 8192 
rand_seed1 
rand_seed2 
range_alloc_block_size 4096 
read_buffer_size 262144 
read_only OFF 
read_rnd_buffer_size 524288 
relay_log /var/lib/mysql/slave-relay.log 
relay_log_index /var/lib/mysql/slave-relay-log.index 
relay_log_info_file relay-log.info 
relay_log_purge ON 
relay_log_space_limit 0 
report_host 
report_password 
report_port 3306 
report_user 
rpl_recovery_rank 0 
secure_auth OFF 
secure_file_priv  
server_id 22 
skip_external_locking ON 
skip_networking OFF 
skip_show_database OFF 
slave_allow_batching OFF 
slave_compressed_protocol OFF 
slave_exec_mode STRICT 
slave_load_tmpdir /var/tmp/mysql.XeaVJ2 
slave_net_timeout 3600 
slave_skip_errors OFF 
slave_transaction_retries 10 
slow_launch_time 2 
slow_query_log ON 
slow_query_log_file /var/run/mysql/mysqld-slow.log 
socket /var/run/mysql/mysql.sock 
sort_buffer_size 524288 
sql_auto_is_null ON 
sql_big_selects ON 
sql_big_tables OFF 
sql_buffer_result OFF 
sql_log_bin ON 
sql_log_off OFF 
sql_log_update ON 
sql_low_priority_updates OFF 
sql_max_join_size 18446744073709551615 
sql_mode  
sql_notes ON 
sql_quote_show_create ON 
sql_safe_updates OFF 
sql_select_limit 18446744073709551615 
sql_slave_skip_counter 
sql_warnings OFF 
ssl_ca 
ssl_capath 
ssl_cert  
ssl_cipher 
ssl_key 
storage_engine MyISAM 
sync_binlog 0 
sync_frm ON 
system_time_zone EEST 
table_definition_cache 256 
table_lock_wait_timeout 50 
table_open_cache 64 
table_type MyISAM 
thread_cache_size 0 
thread_handling one-thread-per-connection 
thread_stack 262144 
time_format %H:%i:%s 
time_zone SYSTEM 
timed_mutexes OFF 
timestamp 1305623046 
tmp_table_size 16777216 
tmpdir /var/tmp/mysql.XeaVJ2 
transaction_alloc_block_size 8192 
transaction_allow_batching ON 
transaction_prealloc_size 4096 
tx_isolation REPEATABLE-READ 
unique_checks ON 
updatable_views_with_limit YES 
version 5.1.41-ndb-7.0.13-log 
version_comment SUSE MySQL Cluster RPM 
version_compile_machine x86_64 
version_compile_os suse-linux-gnu 
wait_timeout 28800 
warning_count 0

来源

2011-05-16 richard

这是一串长长的数字，它们是不变的，它们是用户的输入还是它是另一个查询的结果？ – 2011-05-16 11:46:04

你有没有索引？特别是关于'TIME_ADDED'？ – 2011-05-16 11:46:31

你的查询计划是怎样的？ – 2011-05-16 11:48:18

你应该在channel_id，userid，type，public和time_added上有索引。

来源

2011-05-16 11:45:57 ic3b3rg

+1不知道类型和公共，虽然。 – 2011-05-16 12:05:50

经验法则是在连接中出现的每个字段上都有一个索引，其中，分组依据，排序顺序和子顺序排列。在这种情况下，如果索引是为channel_id，userid和time_added创建的，但不是type和public，则数据库将使用索引来查找与channel_id，userid和time_added过滤器匹配的记录子集，然后扫描所有记录子集满足类型和公共过滤器。 – ic3b3rg 2011-05-16 12:25:03

数据库通常对每个搜索只使用一个索引 - 所以在这种情况下，它需要是一个复合索引。是否包含公共和类型取决于它将对需要扫描的数据集有多大的影响。 – 2011-05-16 14:11:50

我们对MySQL的

默认设置这是不好的，因为这些都是非常保守的值。如果你的存储引擎是InnoDB（如果你的存储引擎不是，那么切换到它），那么默认的缓存大小是8MB - 所以你的其他4GB的RAM是相当无用的。

将innodb_buffer_pool_size设置为3GB（如果这是一个DB专用服务器） - 否则在那里放一些体面的价值。

另外，请告诉我们关于该表的索引。

来源

2011-05-16 11:47:25

我们真的需要从MyISAM切换到InnoDb吗？ InnoDb在我们的案例中能带来什么收益？ – richard 2011-05-17 09:26:43

它带来了行级锁定，而不是表级锁定，在读取（新版本的MySQL）时没有损失。另外，对于MyISAM，你必须微调许多参数，在InnoDB中，你基本上有一个参数。 – 2011-05-17 11:02:04

当问MySQL查询的性能问题，你应该总是张贴如下：

输出的 SHOW CREATE TABLE tablename

输出的

EXPLAIN SELECT query_goes_here输出的SHOW VARIABLES

发生了什么事是，你的表可能没有被优化，因为它应该是MySQL的默认设置运行 - 它调整了慢机器，并没有使用任何adva您可能拥有的硬件支持。

现在，在选择索引时 - 我们必须考虑几个因素，我读过一个人提到你应该索引你加入的所有东西 - 在所有情况下都不是这样，基数在选择索引时起作用。

您应该编辑您的问题并向我们提供更多数据。

来源

2011-05-16 15:07:03

我们在上面添加了mor einfo。 – richard 2011-05-17 09:27:11

查询的MySQL性能问题|优化最佳实践？

回答

相关问题