Redis~geospatial (geospatial), hyperloglog (cardinality Statistics)

Get the coordinates of Xi'an ——GEOPOS

127.0.0.1:6379> GEOPOS CHINA xian

1. 1. “108.94683212041854858”

2. “34.29296115814533863”

In longitude 120 latitude 35 Position centered , Get the radius 1000 Cities within kilometers ——GEORADIUS

127.0.0.1:6379> GEORADIUS CHINA 120 35 1000 km

1. “beijing”

2. “shanghai”

Get the radius in Guangzhou 500 Cities within kilometers ——GEORADIUSBYMEMBER

127.0.0.1:6379> GEORADIUSBYMEMBER CHINA guangzhou 500 km

1. “shenzhen”

2. “guangzhou”

3. “hainan”

Convert the coordinates of Guangzhou and Shenzhen to 11 For the purpose of GEO Hash value ——GEOHASH

127.0.0.1:6379> GEOHASH CHINA guangzhou shenzhen

1. “ws0e9cb3yj0”

2. “ws10k0dcg10”

[](()Hyperloglog( Base Statistics )

-------------- 《 A big factory Java Analysis of interview questions + Back end development learning notes + The latest architecture explanation video + Practical project source code handout 》 Free open source Prestige search official account 【 Advanced programming 】 ------------------------------------------------------------------

• After we count the number of visits to the website 、 Daily activity hours , Because we count the number of users, not the number of visits , Therefore, even if a user visits multiple times, it will only be counted once , This non repeating data is often referred to as cardinality .

• In traditional practice , We usually use set To save the user's ID To count , Because it has the function of de duplication , But what we need is to count the users , If all users' ID Save the method to complete , When the number of users is large, there will be a huge pressure on the memory , And the efficiency is greatly reduced .

• To solve this problem ,Redis stay 2.8.9 Version added HyperLogLog structure .

[](() Realization principle

• Redis HyperLogLog It's an algorithm for cardinality statistics ,HyperLogLog The advantages of , When the number or volume of input elements is very, very large , The space needed to calculate the cardinality is always fixed 、 And it's very small .

• stay Redis Inside , Every HyperLogLog Keys only cost 12KB Memory , So we can calculate the proximity 2^64 Cardinality of different elements . This is the same as calculating the cardinality , The more elements consume memory, the more collections there are .

• HyperLogLog Using a probabilistic algorithm , By storing elements of hash The first one worth it 1 The location of , To calculate the number of elements , therefore HyperLogLog The element itself is not stored , There may also be some errors when the amount of data is large . But in terms of cardinality Statistics , Its effect is incomparable to other structures .

[](() Use

PFADD key value Add the specified value to Hyperloglog in

PFCONUT key Return to a given Hyperloglog The base estimate of

PFMERGE destkey sourcekey The target Hyperloglog Merge to source Hyperloglog in

127.0.0.1:6379> PFADD NUMS1 1 2 3 4 # towards NUMS1 Insert 1-4

(integer) 1

127.0.0.1:6379> PFADD NUMS1 1 # Data already exists , Don't insert

(integer) 0

127.0.0.1:6379> PFCOUNT NUMS1 # View the current base quantity

(integer) 4

127.0.0.1:6379> PFADD NUMS2 3 4 5 6 # towards NUMS2 Insert 3-6

(integer) 1

127.0.0.1:6379> PFMERGE NUMS1 NUMS2 # take NUMS2 Merge into NUMS1 in

OK

127.0.0.1:6379> PFCOUNT NUMS1 # here NUMS1 Recorded in 1-6, Six elements

(integer) 6

[](()Bitmap( Bitmap )

• Bitmap is actually a deformation of hash , He processes data through hash mapping , Bitmaps themselves do not store data , Instead, store tags . By a bit , namely 0/1 To mark two states of a data

• Bitmaps are usually used for large amounts of data , Mark two states of a certain data under the scenario that the data is not repeated . We can use bitmap to record the current user's Login status 、 Or punch in 、 Check in and other functions .

[](() Use

GETBIT key offset value(0/1) Set up Bitmap The middle offset is offset The value of the position of

SETBIT key offset value return Bitmap The middle offset is offset The value of the position of

BITCOUNT key Calculate how many are in the bitmap 1

127.0.0.1:6379> SETBIT TEST 1 1 # In the bitmap 1,3,5 Bit is set to 1

(integer) 0

127.0.0.1:6379> SETBIT TEST 3 1

(integer) 0

127.0.0.1:6379> SETBIT TEST 5 1

(integer) 0

127.0.0.1:6379> GETBIT TEST 1 # View the bitmap 1,2,3 The value of a

(integer) 1

127.0.0.1:6379> GETBIT TEST 2

(integer) 0

127.0.0.1:6379> GETBIT TEST 3