How to implement Redis distributed cache and flash sales

1. Problems with single-point Redis

1. Data loss problem

Redis data persistence.

2. Concurrency issue

Our master-slave cluster realizes the separation of reading and writing.

3. Failure recovery issues

Use Redis Sentinel to implement health detection and automatic recovery.

4. Storage capacity issue

Build a sharded cluster and use the slot mechanism to achieve dynamic expansion.

2. RDB

RDB's full name is Redis Database Backup file (Redis data backup file), also called Redis data snapshot. To put it simply, all the data in the memory is recorded to the disk. When the Redis instance fails and restarts, the snapshot file is read from the disk and the data is restored.
The snapshot file is called an RDB file and is saved in the current running directory by default.

There is a mechanism to trigger RDB inside Redis, which can be found in the redis.conf file. The format is as follows:

When the bgsave command is executed, fork will be passed. The system call creates a child process that shares memory data with the main process. After completing the fork, read the memory data and write it to the RDB file.

Fork uses copy-on-write technology:

• When the main process performs a read operation, it accesses the shared memory;

• When the main process performs a write operation, it will copy a copy of the data and perform the write operation;

The basic process of RDB mode bgsave?

1. Fork the main process to get a child process and share the memory space;

2. The child process reads the memory data and writes a new RDB file;

3. Replace the old RDB file with the new RDB file;

When will RDB be executed? What does save 60 1000 mean?

• The default is when the service is stopped;

• means that RDB will be triggered if at least 1000 modifications are performed within 60 seconds;

Disadvantages of RDB?

• The RDB execution interval is long, and there is a risk of data loss between two RDB writes;

• fork sub-process, compression, write It is time-consuming to export RDB files;

The frequency of AOF command recording can also be configured through the redis.conf file:

3. AOF

AOF stands for Append Only File. Every write command processed by Redis will be recorded in the AOF file, which can be regarded as a command log file.

AOF is turned off by default. You need to modify the redis.conf configuration file to enable AOF:

The frequency of AOF command recording can also be passed through redis.conf File to match:

Configuration items	Flush timing	Advantages	Disadvantages
Always	Synchronous disk flush	High reliability, almost no data loss	Great impact on performance
everysec	Flash disk per second	Moderate performance	Lost data for up to one minute
no	Operating system control	Best performance	Poor reliability, may lose a lot of data

Because it is a recording command, the AOF file will be much larger than the RDB file. Although AOF will record multiple write operations to the same key, only the last write operation among them is meaningful. You can use the bgrewriteaof command to complete the AOF file rewriting function with the minimum number of commands.

set id 1
set name nezha
set id 2

bgrewriteaof

mset name nezha id 2

Redis will also automatically rewrite the AOF file when the threshold is triggered. The threshold can also be configured in redis.conf:

# If the AOF file grows by more than the last file, the rewrite will be triggered. auto-aof-rewrite-percentage 100# What is the minimum size of the AOF file? Trigger rewrite auto-aof-rewrite-min-size 64mb

RDB and AOF each have their own advantages and disadvantages. If the data security requirements are high, the two are often combined in actual development to use.

##RDBAOFPersistence methodRegular snapshots of the entire memoryRecord every executed commandData integrityIncomplete, between two backups Will loseRelatively complete, depends on the brushing strategyFile sizeThere will be compression, the file size is smallrecords command, the file size is very largeDowntime recovery speedQuicklySlowData Recovery priorityLow, because data integrity is not lowHigh, because data integrity is higherSystem resource usageHigh, a lot of CPU and memory consumptionLow, mainly disk IO resources, but AOF rewriting will occupy a lot of CPU and memory resourcesUsage scenariosCan tolerate data loss for several minutes and pursue faster startup speedCommon with higher requirements for data security 4. Redis optimization flash sale process

1. Flash sale steps:

1. Query coupons;

2. Judge flash sale product inventory;

3. Query order

4. Verify one order per person;

5. Reduce inventory;

6. Create an order;

2. Redis optimization flash sale steps:

1. Add flash sale coupons and save the coupon information to Redis;

2. Based on Lua script, determine the flash sale product inventory, one person per order, determine whether the user's flash sale is successful;

3. If the flash sale is successful, encapsulate the coupon id, user id, and product id into the blocking queue;

4. Start the asynchronous task and continuously remove the items from the blocking queue Read information and implement asynchronous ordering function;

3. Lua script for flash sale

4. Call the flash kill lua script

public Result seckillVoucher(Long voucherId) {
     Long userId = UserHolder.getUser().getId();
     long orderId = redisIdWorker.nextId("order");
     // 1.执行lua脚本
     Long result = stringRedisTemplate.execute(
             SECKILL_SCRIPT,
             Collections.emptyList(),
             voucherId.toString(), userId.toString(), String.valueOf(orderId)
     );
     int r = result.intValue();
     // 2.判断结果是否为0
     if (r != 0) {
         // 2.1.不为0 ，代表没有购买资格
         return Result.fail(r == 1 ? "库存不足" : "不能重复下单");
     }
     // 3.返回订单id
     return Result.ok(orderId);
 }

5. Through the thread pool, operate the blocking queue

// 线程池
private static final ExecutorService SECKILL_ORDER_EXECUTOR = Executors.newSingleThreadExecutor();

/**
* 在类初始化完成后执行
*/
@PostConstruct
private void init() {
    SECKILL_ORDER_EXECUTOR.submit(new VoucherOrderHandler());
}

// 阻塞队列
private BlockingQueue<VoucherOrder> orderTasks = new ArrayBlockingQueue<>(1024 * 1024);
private class OrderHandler implements Runnable{

    @Override
    public void run() {
        while (true){
            try {
                doSomething();
            } catch (Exception e) {
                log.error("处理订单异常", e);
            }
        }
    }
}

5. Implement shared session login based on Redis

Login based on session

Implementing shared session login based on Redis

public class RefreshTokenInterceptor implements HandlerInterceptor {

    private StringRedisTemplate stringRedisTemplate;

    public RefreshTokenInterceptor(StringRedisTemplate stringRedisTemplate) {
        this.stringRedisTemplate = stringRedisTemplate;
    }

    @Override
    public boolean preHandle(HttpServletRequest request, HttpServletResponse response, Object handler) throws Exception {
        // 1、获取请求头中的token
        String token = request.getHeader("authorization");
        if (StrUtil.isBlank(token)) {
            return true;
        }
        // 2、基于TOKEN获取redis中的用户
        String key  = LOGIN_USER_KEY + token;
        Map<Object, Object> userMap = stringRedisTemplate.opsForHash().entries(key);
        // 3、判断用户是否存在
        if (userMap.isEmpty()) {
            return true;
        }
        // 5、将查询到的hash数据转为UserDTO
        UserDTO userDTO = BeanUtil.fillBeanWithMap(userMap, new UserDTO(), false);
        // 6、存在，保存用户信息到 ThreadLocal
        UserHolder.saveUser(userDTO);
        // 7、刷新token有效期
        stringRedisTemplate.expire(key, LOGIN_USER_TTL, TimeUnit.MINUTES);
        // 8、放行
        return true;
    }

    @Override
    public void afterCompletion(HttpServletRequest request, HttpServletResponse response, Object handler, Exception ex) throws Exception {
        // 移除用户
        UserHolder.removeUser();
    }
}

The above is the detailed content of How to implement Redis distributed cache and flash sales. For more information, please follow other related articles on the PHP Chinese website!