Optimizing Amazon RDS – MySQLTuner Recommendations and Feedback

amazon-rdsdatabase-designMySQLperformancequery-performance

I'm currently trying to optimize a database I've inherited for speed.
The server is an Amazon RDS db.m3.large instance:

2 VCPUs
7.5 GB Memory
500 GB HDD

My current strategy is to do the following:

Remove all dev databases and push them to a different server
Add unique index on aggr_id column in the problematic tables below (How to do this properly without interrupting production?)
Switch MyISAM tables to InnoDB to prevent locking (What are the implications of doing this?)

I am making this post to get feedback on the current approach, and any additional ideas. I'm not a DBA by any means, just a web developer. Any and all feedback is welcomed, thanks in advance. I'll be sure to respond and approve answers! I will also edit this post with any insights gained and resources found.

There are a few problem tables all with the same format:

| xxxxx | CREATE TABLE `xxxxx` (
  `id` int(11) NOT NULL AUTO_INCREMENT,
  `aggr_id` int(11) NOT NULL,
  `craft_id` int(11) NOT NULL,
  `task_year` smallint(8) unsigned NOT NULL,
  `hc1` mediumint(8) unsigned DEFAULT NULL,
  ......
  `hc365` mediumint(8) unsigned DEFAULT NULL,
  PRIMARY KEY (`id`,`task_year`),
  KEY `IDX_AGGRLABORDAY` (`task_year`,`aggr_id`,`craft_id`)
) ENGINE=MyISAM AUTO_INCREMENT=50480037 DEFAULT CHARSET=latin1
/*!50500 PARTITION BY RANGE  COLUMNS(task_year)
(PARTITION p1 VALUES LESS THAN (2012) ENGINE = MyISAM,
 PARTITION p2 VALUES LESS THAN (2013) ENGINE = MyISAM,
 PARTITION p3 VALUES LESS THAN (2014) ENGINE = MyISAM,
 PARTITION p4 VALUES LESS THAN (2015) ENGINE = MyISAM,
 PARTITION p5 VALUES LESS THAN (2016) ENGINE = MyISAM,
 PARTITION p6 VALUES LESS THAN (2017) ENGINE = MyISAM,
 PARTITION p7 VALUES LESS THAN (9999) ENGINE = MyISAM) */

Here is an explain of a problematic query:

explain DELETE FROM xxxxx WHERE aggr_id = 3000010;
+----+-------------+---------------------------+------+---------------+------+---------+------+----------+-------------+
| id | select_type | table                     | type | possible_keys | key  | key_len | ref  | rows     | Extra       |
+----+-------------+---------------------------+------+---------------+------+---------+------+----------+-------------+
|  1 | SIMPLE      | xxxxx                     | ALL  | NULL          | NULL | NULL    | NULL | 46611048 | Using where |
+----+-------------+---------------------------+------+---------------+------+---------+------+----------+-------------+
1 row in set (0.05 sec)

Below is MySQLTuner results:

[--] Assuming 7680 MB of physical memory
[!!] Assuming 0 MB of swap space (use --forceswap to specify)
[OK] Currently running supported MySQL version 5.6.21-log

-------- Storage Engine Statistics -------------------------------------------
[--] Status: +ARCHIVE +BLACKHOLE +CSV -FEDERATED +InnoDB +MRG_MYISAM 
[--] Data in MyISAM tables: 274G (Tables: 75)
[--] Data in PERFORMANCE_SCHEMA tables: 0B (Tables: 52)
[--] Data in InnoDB tables: 7G (Tables: 227)
[!!] Total fragmented tables: 35

-------- Security Recommendations  -------------------------------------------
[OK] All database users have passwords assigned

-------- Performance Metrics -------------------------------------------------
[--] Up for: 107d 15h 56m 50s (110M q [11.901 qps], 3M conn, TX: 14149B, RX: 108B)
[--] Reads / Writes: 58% / 42%
[--] Total buffers: 5.4G global + 1.5M per thread (604 max threads)
[OK] Maximum possible memory usage: 6.2G (83% of installed RAM)
[OK] Slow queries: 0% (17K/110M)
[OK] Highest usage of available connections: 29% (177/604)
[OK] Key buffer size / total MyISAM indexes: 16.0M/49.6G
[OK] Key buffer hit rate: 99.4% (25B cached / 155M reads)
[!!] Query cache is disabled
[OK] Sorts requiring temporary tables: 2% (133K temp sorts / 6M sorts)
[!!] Temporary tables created on disk: 62% (7M on disk / 11M total)
[OK] Thread cache hit rate: 74% (945K created / 3M connections)
[!!] Table cache hit rate: 0% (1K open / 674K opened)
[OK] Open file limit used: 3% (2K/65K)
[!!] Table locks acquired immediately: 78%
[!!] InnoDB  buffer pool / data size: 5.3G/7.7G
[!!] InnoDB log waits: 1
-------- Recommendations -----------------------------------------------------
General recommendations:
    Run OPTIMIZE TABLE to defragment tables for better performance
    When making adjustments, make tmp_table_size/max_heap_table_size equal
    Reduce your SELECT DISTINCT queries without LIMIT clauses
    Increase table_open_cache gradually to avoid file descriptor limits
    Read this before increasing table_open_cache over 64: http://bit.ly/1mi7c4C
    Optimize queries and/or use InnoDB to reduce lock wait
Variables to adjust:
    query_cache_type (=1)
    tmp_table_size (> 16M)
    max_heap_table_size (> 16M)
    table_open_cache (> 2000)
    innodb_buffer_pool_size (>= 7G)
    innodb_log_buffer_size (>= 8M)

Best Answer

(I'm writing this as I see your points; please read to the end before taking action. I 'develop' the best answer piece by piece -- hope you learn some things.)

DELETE FROM xxxxx WHERE aggr_id = 3000010;

must scan every row in every partition. That means it must do a lot of I/O, which will take a lot of time, regardless of the tuning.

If this is a common query, then add

INDEX(aggr_id)

Even better is change

KEY `IDX_AGGRLABORDAY` (`task_year`,`aggr_id`,`craft_id`)

Normally it is best to have the "partition key" (task_year) at the end of indexes, not the start. In this case, you may as well simply remove it. Any use of task_year will be for "pruning", then the KEY can take over. DROP that INDEX and ADD the following in a single ALTER:

KEY `IDX_AGGRLABORDAY` (`aggr_id`,`craft_id`)

Caution: It will lock the table for some time.

You ought to switch to InnoDB, which has online INDEX operations.

The tuner suggestions seem to be worse than usual:

Leave the query cache off
raising the tmp_table_size and max_heap_table_size is not necessarily useful
You seem to be using a mixture of InnoDB and MyISAM, so I recommend innodb_buffer_pool_size = 2500M and key_buffer_size = 500M. (The high buffer_pool setting was starving the data cache for MyISAM.)
As for the tmp tables; let's see your slowest query, plus the SHOW CREATE TABLE.
Ignore table fragmentation; don't bother OPTIMIZEing.
table_open_cache should be a few hundred. If open_files_limit is only 1K, then something in the OS is constraining the cache.

If aggr_id can be made UNIQUE, then perhaps it should be the PRIMARY KEY and eliminate id. Note that you cannot have a UNIQUEness constraint because of partitioning.

Note that there are two things intertwined here: Switching to InnoDB and making aggr_id the PK. If you do both, then do

PRIMARY KEY (aggr_id, craft_id, task_year)

and have no secondary keys. The rationale:

InnoDB really needs a PK.
task_year (the partition key) must be in every unique (that includes PK) key
It is usually best to put the partition key last
You apparently need craft_id for some other reason? Quite possibly it should be removed, especially since the PK coexists with the data in InnoDB. (Please show the SELECT(s) that wants craft_id; we can discuss this further.)

Switching to InnoDB is almost always "good". See my blog for the gotchas: http://mysql.rjweb.org/doc.php/myisam2innodb While you are at it, read about PARTITION issues here: http://mysql.rjweb.org/doc.php/partitionmaint It may conclude that partitioning is doing you no good.

If you go to all-InnoDB, then innodb_buffer_pool_size = 5G and key_buffer_size = 20M.

Related Solutions

Mysql – How to debug a db memory-leak causing thesql to go before it’s own limits

...even surpassing it's theorically maximum possible allocation.

[OK] Maximum possible memory usage: 7.3G (46% of installed RAM)

There is not actually a way to calculate maximum possible memory usage for MySQL, because there is no cap on the memory it can request from the system.

The calculation done by mysqltuner.pl is only an estimate, based on a formula that doesn't take into account all possible variables, because if all possible variables were taken into account, the answer would always be "infinite." It's unfortunate that it's labeled this way.

Here is my theory on what's contributing to your excessive memory usage:

thread_cache_size       = 128

Given that your max_connections is set to 200, the value of 128 for thread_cache_size seems far too high. Here's what makes me think this might be contributing to your problem:

When a thread is no longer needed, the memory allocated to it is released and returned to the system unless the thread goes back into the thread cache. In that case, the memory remains allocated.

^{http://dev.mysql.com/doc/refman/5.6/en/memory-use.html}

If your workload causes even an occasional client thread to require a large amount of memory, those threads may be holding onto that memory, then going back to the pool and sitting around, continuing to hold on to memory they don't technically "need" any more, on the premise that holding on to the memory is less costly than releasing it if you're likely to need it again.

I think it's worth a try to do the following, after first making a note of how much memory MySQL is using at the moment.

Note how many threads are currently cached:

mysql> show status like 'Threads_cached';
+----------------+-------+
| Variable_name  | Value |
+----------------+-------+
| Threads_cached | 9     |
+----------------+-------+
1 row in set (0.00 sec)

Next, disable the thread cache.

mysql> SET GLOBAL thread_cache_size = 0;

This disables the thread cache, but the cached threads will stay in the pool until they're used one more time. Disconnect from the server, then reconnect and repeat.

mysql> show status like 'Threads_cached';

Continue disconnecting, reconnecting, and checking until the counter reaches 0.

Then, see how much memory MySQL is holding.

You may see a decrease, possibly significant, and then again you may not. I tested this on one of my systems, which had 9 threads in the cache. Once those threads had all been cleared out of the cache, the total memory held by MySQL did decrease... not by much, but it does illustrate that threads in the cache do release at least some memory when they are destroyed.

If you see a significant decrease, you may have found your problem. If you don't, then there's one more thing that needs to happen, and how quickly it can happen depends on your environment.

If the theory holds that the other threads -- the ones currently servicing active client connections -- have significant memory allocated to them, either because of recent work in their current client session or because of work requiring a lot of memory that was done by another connection prior to them languishing in the pool, then you won't see all of the potential reduction in memory consumption until those threads are allowed to die and be destroyed. Presumably your application doesn't hold them forever, but how long it will take to know for sure whether there's a difference will depend on whether you have the option of cycling your application (dropping and reconnecting the client threads) or if you'll have to just wait for them to be dropped and reconnected over time on their own.

But... it seems like a worthwhile test. You should not see a substantial performance penalty by setting thread_cache_size to 0. Fortunately, thread_cache_size is a dynamic variable, so you can freely change it with the server running.

MySQL Performance – Why Simple SELECTs on InnoDB Are 100x Slower Than on MyISAM

YOUR QUERY

SELECT post.postid, post.attach FROM newbb_innopost AS post WHERE post.threadid = 51506;

At first glance, that query should only touches 1.1597% (62510 out of 5390146) of the table. It should be fast given the key distribution of threadid 51506.

REALITY CHECK

No matter which version of MySQL (Oracle, Percona, MariaDB) you use, none of them can fight to one enemy they all have in common : The InnoDB Architecture.

InnoDB Architecture

CLUSTERED INDEX

Please keep in mind that the each threadid entry has a primary key attached. This means that when you read from the index, it must do a primary key lookup within the ClusteredIndex (internally named gen_clust_index). In the ClusteredIndex, each InnoDB page contains both data and PRIMARY KEY index info. See my post Best of MyISAM and InnoDB for more info.

REDUNDANT INDEXES

You have a lot of clutter in the table because some indexes have the same leading columns. MySQL and InnoDB has to navigate through the index clutter to get to needed BTREE nodes. You should reduced that clutter by running the following:

ALTER TABLE newbb_innopost
    DROP INDEX threadid,
    DROP INDEX threadid_2,
    DROP INDEX threadid_visible_dateline,
    ADD INDEX threadid_visible_dateline_index (`threadid`,`visible`,`dateline`,`userid`)
;

Why strip down these indexes ?

The first three indexes start with threadid
threadid_2 and threadid_visible_dateline start with the same three columns
threadid_visible_dateline does not need postid since it's the PRIMARY KEY and it's embedded

BUFFER CACHING

The InnoDB Buffer Pool caches data and index pages. MyISAM only caches index pages.

Just in this area alone, MyISAM does not waste time caching data. That's because it's not designed to cache data. InnoDB caches every data page and index page (and its grandmother) it touches. If your InnoDB Buffer Pool is too small, you could be caching pages, invalidating pages, and removing pages all in one query.

TABLE LAYOUT

You could shave of some space from the row by considering importthreadid and importpostid. You have them as BIGINTs. They take up 16 bytes in the ClusteredIndex per row.

You should run this

SELECT importthreadid,importpostid FROM newbb_innopost PROCEDURE ANALYSE();

This will recommend what data types these columns should be for the given dataset.

CONCLUSION

MyISAM has a lot less to contend with than InnoDB, especially in the area of caching.

While you revealed the amount of RAM (32GB) and the version of MySQL (Server version: 10.0.12-MariaDB-1~trusty-wsrep-log mariadb.org binary distribution, wsrep_25.10.r4002), there are still other pieces to this puzzle you have not revealed

The InnoDB settings
The Number of Cores
Other settings from my.cnf

If you can add these things to the question, I can further elaborate.

UPDATE 2014-08-28 11:27 EDT

You should increase threading

innodb_read_io_threads = 64
innodb_write_io_threads = 16
innodb_log_buffer_size = 256M

I would consider disabling the query cache (See my recent post Why query_cache_type is disabled by default start from MySQL 5.6?)

query_cache_size = 0

I would preserve the Buffer Pool

innodb_buffer_pool_dump_at_shutdown=1
innodb_buffer_pool_load_at_startup=1

Increase purge threads (if you do DML on multiple tables)

innodb_purge_threads = 4