I have multiple big tables for business data with smallest one having 38million rows(24G data, 26G index size). I have indexes setup to speed up the lookups and buffer pool set to 80% of total RAM(116G). Even after these settings, over time we have started observing performance issues. I have constraints with the disk size(1T) and sharding is not an option currently. The data growth has increased to 0.5M rows per day. This is leading to frequent optimisation and master switch exercises. Table schemas and indexes have already been optimised. Hence, I have started looking at partitioning the table to improve performance. My primary partitioning use case is to delete data on monthly basis by dropping partitions so that optimisations are not required and read/write latencies are improved. Following is the structure for one of the big tables(column names have been changed for legal reasons - assume that the columns where indexes are defined have lookup use cases):
CREATE TABLE `table_name` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`data_1` int(11) NOT NULL,
`data_2` varchar(40) COLLATE utf8_unicode_ci NOT NULL,
`data_3` varchar(50) COLLATE utf8_unicode_ci DEFAULT NULL,
`data_4` varchar(20) COLLATE utf8_unicode_ci DEFAULT NULL,
`created_at` datetime DEFAULT NULL,
`updated_at` datetime DEFAULT NULL,
PRIMARY KEY (`id`),
KEY `index_data1` (`data_1`),
KEY `index_data2` (`data_2`)
) ENGINE=InnoDB AUTO_INCREMENT=100572 DEFAULT CHARSET=utf8 COLLATE=utf8_unicode_ci;
I am planning to partition on the created_at column. However, the problem is that the partitioning column has to be part of all the unique keys. I can add the created_at column to the primary key but that would lead to increase in index size which in turn has its own side effects. Is there some workaround or any better solution?
Apart from solving this problem, there are few more questions whose answers couldn't be found in any documentation or articles present.
1. Why does mysql warrant partitioning column to be part of unique key?
2. The queries from the ORM don't have created_at clause present that means pruning is not possible with reads which we were okay with provided inserts are always pruned. However, doesn't look like this is the case. Why does mysql open all the partitions for inserts?
Mysql Version - 5.6.33-79.0-log Percona Server (GPL), Release 79.0, Revision 2084bdb
PRIMARY KEY(id, created_at) will take only an tiny bit more space than PRIMARY KEY(id). I estimate it at much less than 1% for your data. I can't tell about the index space -- can you show us the non-primary index(es)?
Explanation: The leaf nodes of the data (which is a BTree organized by the PK), will not change in size. The non-leaf nodes will have created_at added to each 'row'. As a rule of thumb in InnoDB, non-leaf nodes take up about 1% of the space for the BTree.
For the INDEX BTrees, the leaf nodes need an extra 4 bytes/row for created_at unless created_at is already in the index.
Let's say you currently have INDEX(foo) where foo is INT and id is also INT. That's a total of 8 bytes (plus overhead). Adding created_at (a 4-byte TIMESTAMP) expands each leaf 'row' to 12+overhead. So, that index may double in size.
A guess: Your 24G+26G might grow to 25G+33G.
It sounds like you have several indexes. You do understand that INDEX(a) is not useful if you also have INDEX(a,b)? And that INDEX(x,y) is a lot better than INDEX(x), INDEX(y) in some situations? Let's discuss your indexes.
The main benefit for PARTITIONing is your use case -- DROP PARTITION is a lot faster than DELETE. My blog on such.
Don't be lulled by partitioning. You are hoping for "read/write latencies are improved"; such is not likely to happen. If you would like further explanation please provide a SELECT where you think it might happen.
How many "months" will you partition on? I recommend not more than 50. PARTITIONing has some inefficiencies when there are lots of partitions.
Because of the need for the partition key to be in UNIQUE keys, the uniqueness constraint is almost totally useless. Having it on the end of an AUTO_INCREMENT id is not an issue.
Consider whether something other than id can be the PK.
Question 1: When INSERTing a row, all UNIQUE keys are immediately checked for "dup key". Without the partition key being part of the unique key, this would mean probing every partition. This is too costly to contemplate; so it was not done. (In the future, a 'global-to-the-table' UNIQUE key may be implemented. Version 8.0 has some hooks for such.)
Question 2a: Yes, if the SELECT's WHERE does not adequately specify the partition key, all partitions will be opened and looked at. This is another reason to minimize the number of partitions. Hmmm... If you do a SELECT on the 31st of the month and do the same SELECT the next day, you could get fewer rows (even without any deletes, just the DROP PARTITION); this seems "wrong".
Question 2b: "Why does mysql open all the partitions for inserts?" -- What makes you think it does? There is an odd case where the "first" partition is 'unnecessarily' opened -- the partition key is DATETIME.
Related
I have a table called customer_type with fields (id, customer_type). It has 5 rows, each describing customer type.
I also have a table called quote that uses customer_type_id as one of the foreign key columns:
CREATE TABLE `quote` (
`id` int NOT NULL AUTO_INCREMENT,
`number` int NOT NULL,
`customer_type_id` tinyint(4) DEFAULT NULL,
`comments` text,
PRIMARY KEY (`id`),
KEY `fk_customer_type` (`customer_type_id`),
CONSTRAINT `fk_customer_type`
FOREIGN KEY (`customer_type_id`)
REFERENCES `customer_type` (`id`),
);
There are other columns and indices in the quote table, total of 10 indices. Lately INSERTs to the database started being slow and one possible reason could be having too many indices.
And so I want to remove some, including for example, fk_customer_type. Cardinality of that index is 5, while cardinality of some other indices is much higher (i.e. 5000, or 20000).
I cannot simply drop the index because of foreign key constraint.
Question
Does my situation warrant removing this foreign key and corresponding foreign key constraint?
Where my reasons for removing are:
reduce the number of indices in hopes of improving INSERT performance
cardinality of 'customer_type_id' is very low, to where performance will not likely be affected
My reasons against removing could be:
I will lose the referential integrity (foreign key constraint)
Are there any specific downsides that will happen if I do remove the index? Is it worth keeping the index just to keep the index constraint?
The slowlog is an excellent way to identify the slowest queries. More: mysql.rjweb.org/doc.php/mysql_analysis#slow_queries_and_slowlog
I started with that comment because I suspect that your question about cardinality and FKs and dropping indexes has very little to do with your performance problems.
The cardinality of that TINYINT is low in the quote table. So? If you look up all the rows with customer_type_id = 2 the Optimizer will probably ignore INDEX(customer_type_id). But, I don't think you have that. Let's see SHOW CREATE TABLE quotes.
The customer_type table is tiny. Its data and index(es) are so trivial that I don't even want to discuss them. And you are probably not adding much to it, ever.
During INSERT INTO customer_type ..., the FOREIGN KEY constraint needs to check that there is a matching row in customer_type; that takes a small amount of CPU, but probably zero I/O. There will be one read the first time, then that block will stay in cache (see innodb_buffer_pool_size) until shutdown.
Lately INSERTs to the database started being slow and one possible reason could be having too many indices.
I see 2 indexes in quotes.
The PRIMARY KEY is clustered with the data, and is necessary. And, since it is AUTO_INCREMENT, inserts will go at the "end" of the table -- quite efficient.
KEY(customer_type_id) -- I've already explained that it is probably useless. But it is not a big deal. That is, it would not explain your perceived slowdown.
There are other columns and indices in the quote table
Let's see them!
Normally 10 indexes is no big deal. But there could be a UUID or GUID or something else that raises a red flag. Also, if quotes is a billion rows long, other issues raise their ugly head. Or big text/blob columns.
(Of course, if you have a billion rows, then you are threatening to overflow AUTO_INCREMENT. That is messy to repair.
Show us some of the slow queries; I predict that you could use some "composite" indexes.
Batching inserts can greatly speed them up. Describe the inserts (randomly coming from multiple clients / 1000 at a time / whatever).
This is my table structure:
CREATE TABLE `channel_play_times_bar_chart` (
`playing_date` datetime NOT NULL,
`channel_report_tag` varchar(50) NOT NULL,
`country_code` varchar(50) NOT NULL,
`device_report_tag` int(11) NOT NULL,
`greater_than_30_minutes` decimal(10,0) NOT NULL,
`15_to_30_minutes` decimal(10,0) NOT NULL,
`0-15_minutes` decimal(10,0) NOT NULL,
PRIMARY KEY (`country_code`,`device_report_tag`,`channel_report_tag`,`playing_date`)
) ENGINE=InnoDB DEFAULT CHARSET=latin1;
When I run the following query:
EXPLAIN EXTENDED
SELECT
channel_report_tag,
SUM(`greater_than_30_minutes`) AS '>30 minutes',
SUM(`15_to_30_minutes`) AS '15-30 Minutes',
SUM(`0-15_minutes`) AS '0-15 Minutes'
FROM
channel_play_times_bar_chart USE INDEX (ABCDE)
WHERE country_code = 'US'
AND device_report_tag = 14
AND channel_report_tag = 'DUNYA NEWS'
AND playing_date BETWEEN '2016-09-01'
AND '2016-09-13'
GROUP BY channel_report_tag
ORDER BY SUM(`greater_than_30_minutes`) DESC
LIMIT 10
This is the output I get ( open it in another tab):
The index was defined as :
CREATE INDEX ABCDE
ON channel_play_times_bar_chart (
`country_code`,
`device_report_tag`,
`channel_report_tag`,
`playing_date`,
`greater_than_30_minutes`
)
I am a bit confused here ; The key column shows ABCDE being used the as the index , yet ref column shows NULL. What does this mean ? Is the index actually being used ? If not what did I do wrong ?
It is using the key you are showing in the create index, that is, ABCDE.
It would be nice if you did a
show create table channel_play_times_bar_chart
and just showed it all at once. That key might not be of much use to you as it replicates most of what your rather wide Primary Key already gives you.
Once the query uses the key up thru the 3rd segment of the composite key, it resumes with a WHERE range on playing_date in that composite and finds 8 rows.
Note EXPLAIN is an estimate.
Further, I would reconsider the strategy for your PRIMARY KEY (PK) ideas especially considering that you decided to dupe it up more or less with the creation of ABCDE. That means you are maintaining two indexes with little if anything gained on the second one (you added one column to the secondary index ABCDE).
The PK It is rather WIDE (118 bytes I believe). It dictates the physical ordering. And that idea could easily be a bad one if used throughout the way you architect things. Changes made to data via UPDATE that impact the columns in the PK force a reshuffle of physical ordering of the table. That fact would be a good indication why id INT AUTO_INCREMENT PRIMARY KEY is often used as a best practice use case as it never endures a reshuffle and is THIN (4 bytes).
The width of keys and their strategy with referencing (other) tables (in Foreign Key Constraints) impact key sizes and performance for lookups. Wide keys can measurably slow down that process.
This is not to suggest that you shouldn't have a key on those columns like your secondary index ABCDE. But in general that is not a good idea for the PK.
Note that it could be argued that ABCDE never gives you any benefit over your PK due to range queries ceasing the use of it near the end that just WHERE out with ranges once it hits the date. Just a thought.
A nice read and rather brief is the article Using EXPLAIN to Write Better MySQL Queries.
Your query does use the ABCDE index. As MySQL documentation on EXPLAIN Output Format explains :) (bolding is mine):
key (JSON name: key)
The key column indicates the key (index) that MySQL actually decided
to use. If MySQL decides to use one of the possible_keys indexes to
look up rows, that index is listed as the key value.
The ref field of the explain output is primarily used when joins are present to show the fields / constants / expressions the index was compared with.
I am building a website (LAMP stack) with an Amazon RDS MySQL instance as the back end (type db.m3.medium).
I am happy with database integrity, and it works perfectly with regards to SELECT/JOIN/ETC queries (everything is normalized, indexed, and foreign keyed, all tables have id primary keys and relevant secondary keys / unique keys).
I have a table 'df_products' with approx half a million products in it. The products need to be updated nightly. The process involves a PHP script reading over a large products data-file and inserting data into several tables (products table, product_colours table, brands table, etc), calling either INSERT or UPDATE depending on whether or not a row already exists. This is done as one giant transaction.
What I am seeing is the UPDATE commands are sufficiently fast (50/sec, not exactly lightning but it should do), however the INSERT commands are super slow (1/sec) and appear to be consuming 100% of the CPU. On a dual core instance we see 50% CPU use (i.e. one full core).
I assume that this is because indexes (1x PRIMARY + 5x INDEX + 1x UNIQUE + 1x FULLTEXT) are being rebuilt after every INSERT. However I though that putting the entire process into one transaction should stop indexes being rebuilt until the transaction is committed.
I have tried setting the following params via PHP but there is negligible performance improvement:
$this->db->query('SET unique_checks=0');
$this->db->query('SET foreign_key_checks=0;');
The process will take weeks to complete at this rate so we must improve performance. Google appears to suggest using LOAD DATA. However:
I would have to generate five files in order to populate five tables
The process would have to use UPDATE commands as opposed to INSERT since the tables already exist
I would still need to loop over the products and scan the database for what values already do and don't exist
The database is entirely InnoDB and I don't plan to move to MyISAM (I want transactions, foreign keys, etc). This means that I cannot disable indexes. Even if I did it would probably be a big performance drain as we need to check if a row already exists before we insert it, and without an index this will be super slow.
I have provided the products table defition below for information. Can you please provide advice to what process we should be using to achieve faster INSERT/UPDATE on multiple large related tables? Or what optimisations we can make to our existing process?
Thank you,
CREATE TABLE `df_products` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`id_brand` int(11) NOT NULL,
`title` varchar(255) NOT NULL,
`id_gender` int(11) NOT NULL,
`id_colourSet` int(11) DEFAULT NULL,
`id_category` int(11) DEFAULT NULL,
`desc` varchar(500) DEFAULT NULL,
`seoAlias` varchar(255) CHARACTER SET ascii NOT NULL,
`runTimestamp` timestamp NOT NULL,
PRIMARY KEY (`id`),
UNIQUE KEY `seoAlias_UNIQUE` (`seoAlias`),
KEY `idx_brand` (`id_brand`),
KEY `idx_category` (`id_category`),
KEY `idx_seoAlias` (`seoAlias`),
KEY `idx_colourSetId` (`id_colourSet`),
KEY `idx_timestamp` (`runTimestamp`),
KEY `idx_gender` (`id_gender`),
FULLTEXT KEY `fulltext_title` (`title`),
CONSTRAINT `fk_id_colourSet` FOREIGN KEY (`id_colourSet`) REFERENCES `df_productcolours` (`id_colourSet`) ON DELETE NO ACTION ON UPDATE NO ACTION,
CONSTRAINT `fk_id_gender` FOREIGN KEY (`id_gender`) REFERENCES `df_lu_genders` (`id`) ON DELETE NO ACTION ON UPDATE NO ACTION
) ENGINE=InnoDB AUTO_INCREMENT=285743 DEFAULT CHARSET=utf8
How many "genders" are there? If the usual 2, don't normalize it, don't index it, don't us a 4-byte INT to store it, use a CHAR(1) CHARACTER SET ascii (only 1 byte) or an ENUM (1 byte).
Each unnecessary index is a performance drain on the load, regardless of how it is done.
For INSERT vs UPDATE, look into using INSERT ... ON DUPLICATE KEY UPDATE.
Load the nightly data into a separate table (this could be MyISAM with no indexes). Then run one query to update existing rows and one to insert new rows. (Each needs a JOIN.) See http://mysql.rjweb.org/doc.php/staging_table, especially the 2 SQLs used for "normalizing". They can be adapted to your situation.
Any kind of multi-row query runs noticeably faster than 1-row at a time. (A 100-row INSERT runs 10 times as fast as 100 1-row inserts.)
innodb_flush_log_at_trx_commit = 2 will let the individual write statements run much faster. (Batching them as I suggest won't speed up much.)
I have a client who has asked me to tune his MySQL database in order to implement some new features and to improve the performance of an already existing web app.
The biggest table (~90 GB) has over 200M rows, and is growing at periodic intervals (one per visit to any of the websites he owns). Having continuous INSERTs, each SELECT query performed from the backend page takes a while to complete, as indexes are regenerated each time.
I've done a simulation on my own server switching from BTREE indexes to HASH indexes. Both SELECTs and INSERTs are not running any faster. The table uses MyISAM as storage engine. There are only INSERTs and SELECTs, no UPDATEs or DELETEs.
I've came up with the idea of creating an auxiliary table updated together with each INSERT to speed up every SELECT query coming from the backend. I know this is bad practice, but, I'm sure the performance will improve for the statistics page.
I'm not a database performance expert, as you may have noticed... Is there a better approach for this?
By the way, from phpMyAdmin I've seen that most indexes on the table have a cardinality of 0. In my simulation, this didn't happen. I'm not sure why is this happening.
Thanks a lot.
1st update: I've just learned that hash index isn't available for MyISAM engine.
2nd update: OK. Here's the table schema.
CREATE TABLE `visits` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`datetime` int(8) NOT NULL,
`webmaster_id` char(18) NOT NULL,
`country` char(2) NOT NULL,
`connection` varchar(15) NOT NULL,
`device` varchar(15) NOT NULL,
`provider` varchar(100) NOT NULL,
`ip_address` varchar(15) NOT NULL,
`url` varchar(300) NOT NULL,
`user_agent` varchar(300) NOT NULL,
PRIMARY KEY (`id`),
KEY `datetime` (`datetime`),
KEY `webmaster_id` (`webmaster_id`),
KEY `country` (`country`),
KEY `connection` (`connection`),
KEY `device` (`device`),
KEY `provider` (`provider`)
) ENGINE=InnoDB;
So, instead of performing queries like select count(*) from visits where datetime=20140715 and device="ios", won't it be best to fetch this from select count from visits_stats where datetime=20140715 and device="ios"?
INSERTs are, as said, much more frequent than SELECTs, but my client wants to improve the performance of the backend used to retrieve aggregated data. Using my approach, each visit would imply one INSERT and one INSERT/UPDATE (or REPLACE) which would increment one or more counters (I haven't decided the schema for the visits_stats table yet, the above query was just an example).
Apart from this, I've decided to replace some of the fields by their appropriate IDs from a foreign table. So far, data is stored in strings like connection=cable, device=android, and so on. I'm not sure how would this affect performance.
Thanks again.
Edit: I said before not to use partitions. But Bill is right that the way he described would work. Your only concern would be if you tried to select across the 101 partitions, then the whole thing would come to a standstill. If you don't intend to do this then partitioning would solve the problem. Fix your indexes first though.
Your primary problem is that MyISAM is not the best engine, neither is InnoDB. TokuDB would be your best bet, but you'd have to install that on the server.
Now, you need to prune your indexes. This is the major reason for the slowness. Remove an index on everything that isn't part of common SELECT statements. Add an multi-column index on exactly what is requested in the WHERE of your SELECT statements.
So (in addition to your primary key) you want an index on datetime, device only as a multi-column index, according to your posted SELECT statement.
If you change to TokuDB the inserts will be much faster, if you stick with MyISAM then you could speed the whole thing up by using INSERT DELAYED instead of INSERT. The only issue with this is that the inserts will not be live, but will be added whenever MySQL decides there is not too much load.
Alternatively, if the above still does not help, your final option would be to use two tables. One table that you SELECT from, and another that you INSERT to. Once an day or so you would then copy the insert table to the select table. Though this means the data in your select table could be up to 24 hours old.
Other than that you would have to completely change the table structure, for which I can't tell you how to do because it depends on what you are using it for exactly, or use something other than MySQL for this. However, my above optimizations should work.
I would suggest looking into partitioning. You have to add datetime to the primary key to make that work, because of a limitation of MySQL. The primary or unique keys must include the column by which you partition the table.
Also make the index on datetime into a compound index on (datetime, device). This will be a covering index for the query you showed, so the query can get its answer from the index alone, without having to touch table rows.
CREATE TABLE `visits` (
`id` int(11) NOT NULL AUTO_INCREMENT,
`datetime` int(8) NOT NULL,
`webmaster_id` char(18) NOT NULL,
`country` char(2) NOT NULL,
`connection` varchar(15) NOT NULL,
`device` varchar(15) NOT NULL,
`provider` varchar(100) NOT NULL,
`ip_address` varchar(15) NOT NULL,
`url` varchar(300) NOT NULL,
`user_agent` varchar(300) NOT NULL,
PRIMARY KEY (`id`, `datetime`), -- compound primary key is necessary in this case
KEY `datetime` (`datetime`,`device`), -- compound index for the SELECT
KEY `webmaster_id` (`webmaster_id`),
KEY `country` (`country`),
KEY `connection` (`connection`),
KEY `device` (`device`),
KEY `provider` (`provider`)
) ENGINE=InnoDB
PARTITION BY HASH(datetime) PARTITIONS 101;
So when you query for select count(*) from visits where datetime=20140715 and device='ios', your query is only scanning one partition, with about 1% of the rows in the table. Then within that partition, it narrows down even further using the index.
Inserts should also improve, because they are updating much smaller indexes.
I use a prime number when doing hash partitioning, to help the partitions remain more evenly filled in case the dates inserted follow a regular pattern.
Converting a 90GB table to partitioning is going to take a long time. You can use pt-online-schema-change to avoid blocking your application.
You can even make more partitions if you want, in theory up to 1024 in MySQL 5.5 and 8192 in MySQL 5.6. Although with thousands of partitions, you may run into different bottlenecks, like the number of open files.
P.S.: HASH indexes are not support by either MyISAM or InnoDB. HASH indexes are only supported by MEMORY and NDB storage engines.
You are in the problem which is called Big Data Querying / Big Data handling now a days. For handling big data there are many solutions available unfortunately none of them are easy enough to be implemented. You always need a team to structure Big Data to fulfill your need. Some of The solution I may define here are as Under.
1. Big Table
Google uses this technique to create a whole lot big table with thousands of column.(To minimize records vertically). For which you will have to analyze your data and then partition on the basis of similarity and then tag those similarity with appropriate name. Now you must have to write Query that will be first analyzed by some algorithm to check what column space have to be queried. Not Simple enough
2. Distribute Database Across multiple Machine
Hadoop file system is an open source Apache project which is totally created for solving the problem of storing and querying big data. In early days Space was issue and system were capable enough to process small data but now space is not an issue.Even Small organization have tera bytes of data stored locally. But this terabytes of data can not be be processed in one go at one machine. Even a giant machine can take days to process aggregate operation. That is why hadoop is there.
If you are individual then definitely you are in trouble you will need resource for doing this painful task for You. But you can use the essence of these techniques without employing these technologies.
You are free to give a try to these technique. Just study articles about handling big data. Relational database queries are not gonna work in your case
I have this simple query:
INSERT IGNORE INTO beststat (bestid,period,rawView) VALUES ( 4510724 , 201205 , 1 )
On the table:
CREATE TABLE `beststat` (
`bestid` int(11) unsigned NOT NULL,
`period` mediumint(8) unsigned NOT NULL,
`view` mediumint(8) unsigned NOT NULL DEFAULT '0',
`rawView` mediumint(8) unsigned NOT NULL DEFAULT '0',
PRIMARY KEY (`bestid`,`period`),
) ENGINE=InnoDB AUTO_INCREMENT=2020577 DEFAULT CHARSET=utf8
And it takes 1 sec to completes.
Side Note: actually it doesn't take always 1sec. Sometime it's done even in 0.05 sec. But often it takes 1 sec
This table (beststat) currently has ~500'000 records and its size is: 40MB. I have 4GB RAM and innodb buffer pool size = 104,857,600, with: Mysql: 5.1.49-3
This is the only InnoDB table in my database (others are MyISAM)
ANALYZE TABLE beststat shows: OK
Maybe there is something wrong with InnoDB settings?
I ran some simulations about 3 years ago as part of some evaluation project for a customer. They had a requirement to be able to search a table where data is constantly being added, and they wanted to be up to date up to a minute.
InnoDB has shown much better results in the beginning, but has quickly deteriorated (much before 1mil records), until I have removed all indexes (including primary). At that point InnoDB has become superior to MyISAM when executing inserts/updates. (I have much worse HW then you, executing tests only on my laptop.)
Conclusion: Insert will always suffer if you have indexes, and especially unique.
I would suggest following optimization:
Remove all indexes from your beststat table and use it as a simple dump.
If you really need these unique indexes, consider some programmable solution (like remembering the max bestid at all time, and insisting that the new record is above that number - and immediately increasing this number. (But do you really need so many unique fields - and they all sound to me just like indexes.)
Have a background thread move new records from InnoDB to another table (which can be MyISAM) where they would be indexed.
Consider dropping indexes temporarily and then after bulk update re-indexing the table, possibly switching two tables so that querying is never interrupted.
These are theoretical solutions, I admit, but is the best I can say given your question.
Oh, and if your table is planned to grow to many millions, consider a NoSQL solution.
So you have two unique indexes on the table. You primary key is a autonumber. Since this is not really part of the data as you add it to the data it is what you call a artificial primary key. Now you have a unique index on bestid and period. If bestid and period are supposed to be unique that would be a good candidate for the primary key.
Innodb stores the table either as a tree or a heap. If you don't define a primary key on a innodb table it is a heap if you define a primary key it is defined as a tree on disk. So in your case the tree is stored on disk based on the autonumber key. So when you create the second index it actually creates a second tree on disk with the bestid and period values in the index. The index does not contain the other columns in the table only bestid, period and you primary key value.
Ok so now you insert the data first thing myself does is to ensure the unique index is always unique. Thus it read the index to see if you are trying to insert a duplicate value. This is where the slow down comes into play. It first has to ensure uniqueness then if it passes the test write data. Then it also has to insert the bestid, period and primary key value into the unique index. So total operation would be 1 read index for value 1 insert row into table 1 insert bestid and period into index. A total of three operations. If you removed the autonumber and used only the unique index as the primary key it would read table if unique insert into table. In this case you would have the following number of operations 1 read table to check values 1 insert into tables. This is two operations vs three. So you do 33% less work by removing the redundant autonumber.
I hope this is clear as I am typing from my Android and autocorrect keeps on changing innodb to inborn. Wish I was at a computer.