Several implementations of PHP to solve concurrency problems

For concurrent scenarios such as rush purchase of goods , There may be oversold , At this time, we need to solve these problems caused by concurrency

stay PHP There is no native solution to concurrency in the language , Therefore, we need to use other methods to achieve concurrency control .

Scheme 1 ： Use file lock exclusive lock

flock The function is used to obtain the lock of the file , This lock can only be obtained by one thread at the same time , Other threads that do not acquire locks are either blocked , Or the acquisition fails

When you get the lock , Query inventory first , If the inventory is greater than 0, Then place the order , Reduce inventory , Then release the lock

Option two ： Use Mysql Pessimistic lock provided by database

Innodb The storage engine supports row level locking , When a row of data is locked , Other processes cannot operate on this row of data

First query and lock the row ：select stock_num from table where id=1 for update

if(stock_num > 0){

// Place the order

update table set stock_num=stock-1 where id=1

}

Option three ： Using the queue

Put the user's order request into a queue in turn , In the background, a separate process is used to process the order placing request in the queue

Option four ： Use Redis

redis All operations are atomic , The inventory of goods can be stored in redis in , Check the inventory before placing an order decr operation , If the returned value is greater than or equal to 0 You can place an order later , Otherwise, you cannot place an order , This way is more efficient

if(redis->get('stock_num') > 0){

 stock_num = redis->decr('stock_num')

 if(stock_num >= 0){

 // Place the order

 }else{

 // Insufficient inventory

 }

}else{

// Insufficient inventory

}

Other concurrent problems ：

In practical applications , In many cases, data will be stored in the cache , When the cache fails , Fetch data from the database and reset the cache , If the concurrency is large , There will be many processes going to the database to get data at the same time , Lead to many requests

Penetrating into the database , And make the database crash , File lock can be used here

$data = $cache->get('key');
if(!$data){
	$fp = fopen('lockfile');
	if(flock($fp, LOCK_EX)){
		$data = $cache->get('key');// After getting the lock, check the cache again , There may already be 
		if(!$data){
			$data = mysql->query();
			$cache->set('key', $data);
		}
		flock($fp, LOCK_UN);
	}
	fclose($fp);
}

To put it bluntly , To solve the concurrency problem, we must lock , The essence of various schemes is locking