Shulou(Shulou.com)06/02 Report--

In order to solve the problem of how to achieve high availability distribution in Redis, this article introduces the corresponding analysis and solution in detail, hoping to help more partners who want to solve this problem to find a more simple and feasible method.

Slow query

Action

Log the statements whose query time exceeds a certain limit, and find the command of the bottleneck in the system.

Lifecycle of client request

There are two points:

1. Slow query occurs in the third stage, and the judgment is based on the length of execution.

two。 Client timeout is not necessarily a slow query (all four steps are possible), but slow query is a possible factor in client timeout

Two configurations

Config get slowlog-max-len=128 FIFO queues are usually set to 1000, fixed length and saved in memory

Config get slowlog-log-slower-than=10000 slow query threshold (in microseconds), i.e. 10ms, which is usually set to 1ms

= 0, that is, record all commands (view the time taken by the command), 50s

Use pipeline,1w hset-> 0.7s

Suggestions for use:

Pay attention to the amount of data carried by pipeline each time

Pipeline can only act on one redis node at a time

The difference between M operation and pipeline

4.3 publish and subscribe

Role: publisher subscriber channel

Subscribers can subscribe to multiple channels, but cannot receive pre-subscription messages

Command

Subscribe channel [channel...] Subscribe to information on one or more channels

Publish channel message information is sent to the specified channel

Unsubscribe [channel [channel...]] Means to unsubscribe from a given channel

Redis can be used not only as a cache server, but also as a message queue. Its list type is inherently supported as a message queue.

The message queue needs to be robbed, such as red packets.

Publish and subscribe, such as announcements.

4.4 Bitmap bitmap

Function: a solution to reduce memory

Command

Setbit key offset value sets the specified position offset to 0can1.

Getbit key offset gets the value of the location

Bitcount key [start end] gets the number of specified range values of 1. If not specified, it is all.

Bitop op destkey key [key...] Do multiple Bitmap and (intersection), or (union), not (non), xor (XOR), and save the results in destkey

Within the calculation range of bitpos key targetBit [start] [end], the offset is equal to targetBit, and the first position, if not specified, is all.

Actual combat: independent user statistics

100 million users, 5500 independent, using the difference between set and Bitmap

Only 100000 unique users

Use experience

Type=string, maximum 512MB

Pay attention to the offset in setbit, which may be time consuming

Bitmaps are not absolutely good.

4.5 HyperLogLog

Action

Minimal space completes independent data statistics, which is used to do cardinality statistics (non-repetitive element statistics). The advantage of HyperLogLog is that when the number or volume of input elements is very large, the space needed to calculate the cardinality is always fixed and small, and it is essentially a string.

Command

Pfadd key element [element...] Add the specified element to the HyperLogLog

Pfcount key [key...] Returns the cardinality estimate of a given HyperLogLog

Pfmerge destkey sourcekey [sourcekey...] Merge multiple HyperLogLog into a single HyperLogLog

Use experience

Whether errors can be tolerated or not, the cardinality estimation is to quickly calculate the cardinality within the acceptable range of errors.

Do you need one-on-one data

4.6 GEO Geographic Information

Action

Used to store longitude and latitude, calculate the distance between two places, range calculation, etc., in fact, the zset type implementation

Command

Geoadd key Longitude longitude Latitude latitude City member add Geographic location Information

Geopos key member acquires geolocation information

Geodist key member1 member2 [unit] acquires the distance between the two places in m, km, mi (miles), ft (ruler)

This is the answer to the question about how to achieve high availability distribution in Redis. I hope the above content can be of some help to you. If you still have a lot of doubts to be solved, you can follow the industry information channel for more related knowledge.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.