Everybody Loves Redis And So Can You!

(This talk was presented at Steel City Ruby Conf 2012)

To call it a key-value store would be an understatement. To call it a database would be an overstatement. To call it NoSQL would be cause for defenestration. Redis, quite simply, stands on its own as a remarkable way to store data structures: strings, lists, sets, sorted sets, and hashes.

This talk will articulate the significance and elegance of its design and explore some conventional and rather unconventional ways that you can use Redis in your own applications. If you’re new to Redis, prepare to fall in love, and even if you’re already quite enamored, you’ll definitely enjoy the section about best practices and advanced usage.

Transcript

1. Everybody Loves Redis And So Can You! Mattt Thompson Steel City Ruby Conf 2012
2. @mattt
4. key-value store
5. > SET foo 42
6. > GET foo
7. •strings •hashes •lists •sets •sorted sets
9. http://try.redis-db.com
10. $ redis-cli
11. Strings
12. Strings
13. Strings > SET foo 42
14. Strings > SET foo 42 > GET foo
15. Strings
16. Strings > SET count 1
17. Strings > SET count 1 > INCR count
18. Strings > SET count 1 > INCR count > GET count
19. Hashes
20. {:key => “value”}
22. Hashes
23. Hashes > HMSET users:1 name Mattt
24. Hashes > HMSET users:1 name Mattt > HMSET users:1 age 25
25. Hashes > HMSET users:1 name Mattt > HMSET users:1 age 25 > HMSET users:1 company Heroku
26. Hashes > HMSET users:1 name Mattt > HMSET users:1 age 25 > HMSET users:1 company Heroku > HKEYS users:1
27. Hashes > HMSET users:1 name Mattt > HMSET users:1 age 25 > HMSET users:1 company Heroku > HKEYS users:1 > HGET users:1 name
28. Hashes user = {} user[:name] = “Mattt” user[:age] = 25 user[:company] = “Heroku” user.keys user[:name]
29. $ gem install redis
30. Hashes require ‘redis’ REDIS = Redis.new REDIS.hset “users:1”, “name”, “Mattt” REDIS.hset “users:1”, “age”, “25” REDIS.hset “users:1”, “company”, “Heroku”
31. Hashes require ‘redis’ REDIS = Redis.new user = { name: “Mattt”, age: 25, company: “Heroku” } REDIS.hmset “users:1”, *user.flatten
32. Key Namespacing
33. Key Namespacing users
34. Key Namespacing users:1
35. Key Namespacing users:1:friends
36. Lists
37. [0, 1, 2, 3]
38. Lists
39. Lists > LPUSH fruit Apple Banana Cherry
40. Lists > LPUSH fruit Apple Banana Cherry > LLEN fruit
41. Lists > LPUSH fruit Apple Banana Cherry > LLEN fruit > LINDEX fruit 1
42. Lists > LPUSH fruit Apple Banana Cherry > LLEN fruit > LINDEX fruit 1 > LPOP fruit
43. Lists > LPUSH fruit Apple Banana Cherry > LLEN fruit > LINDEX fruit 1 > LPOP fruit > RPOP fruit
44. Lists > LPUSHX tattoos Kanji
45. Command Naming Conventions
47. Command Naming Conventions
48. • _____ Command Naming Conventions
49. • _____ • H____ Command Naming Conventions
50. • _____ • H____ • L_____ Command Naming Conventions
51. • _____ • H____ • L_____ • R_____ Command Naming Conventions
52. • _____ • H____ • L_____ • R_____ • S_____ Command Naming Conventions
53. • _____ • H____ • L_____ • R_____ • S_____ • Z_____ Command Naming Conventions
54. • _____ • H____ • L_____ • R_____ • S_____ • Z_____ • *M___ Command Naming Conventions
55. • _____ • H____ • L_____ • R_____ • S_____ • Z_____ • *M___ • B____ Command Naming Conventions
56. • _____ • H____ • L_____ • R_____ • S_____ • Z_____ • *M___ • B____ • ____X Command Naming Conventions
57. Sets
58. (:a, :b, :c)
59. Sets > SADD users:1:friends 2 5 7 14 23 > SADD users:2:friends 1 5 6 12 14 22 > SCARD users:1:friends > SISMEMBER users:2:friends 3 > SINTER users:1:friends users:2:friends
60. Time Complexity
61. Big-O Notation
62. Time Complexity
63. • O(1) Time Complexity
64. • O(1) • O(log(n)) Time Complexity
65. • O(1) • O(log(n)) • O(n) Time Complexity
66. • O(1) • O(log(n)) • O(n) • O(n • log(n)) Time Complexity
67. • O(1) • O(log(n)) • O(n) • O(n • log(n)) • O(n2) Time Complexity
69. Sorted Sets
70. *NameError*
72. (foo, bar, baz) + {“foo” => 2}
73. Sorted Sets
74. Sorted Sets > ZADD numbers 1 "one" > ZADD numbers 2 "two" > ZADD numbers 3 "three"
75. Sorted Sets > ZADD numbers 1 "one" > ZADD numbers 2 "two" > ZADD numbers 3 "three" > ZSCORE numbers "one"
76. Sorted Sets > ZADD numbers 1 "one" > ZADD numbers 2 "two" > ZADD numbers 3 "three" > ZSCORE numbers "one" > ZRANGEBYSCORE numbers -inf +inf
77. Sorted Sets > ZADD numbers 1 "one" > ZADD numbers 2 "two" > ZADD numbers 3 "three" > ZSCORE numbers "one" > ZRANGEBYSCORE numbers -inf +inf > ZRANGEBYSCORE numbers 1 2
78. •strings •hashes •lists •sets •sorted sets
81. after hours
82. Json
83. json artist = { “name”: “Radiohead”, “songs”: [ “Kid A”, “Creep”, “Myxamatosis”, ... ] }
84. json
85. REDIS.set “artists:radiohead”, artist.to_json json
86. REDIS.set “artists:radiohead”, artist.to_json json = REDIS.get(“artists:1”) user = Artist.from_json(json) json
87. Concurrency
88. > MGET foo bar baz ... > SADD 1 2 3 5 8 13 21 34 ... > HMSET users:1 k1 v1 k2 v2 ...
89. Concurrency
90. REDIS.multi do Concurrency
91. REDIS.multi do redis.set "foo", "bar" Concurrency
92. REDIS.multi do redis.set "foo", "bar" redis.incr "baz" Concurrency
93. REDIS.multi do redis.set "foo", "bar" redis.incr "baz" end Concurrency
95. > MULTI
96. > MULTI > SET foo bar
97. > MULTI > SET foo bar > INCR baz
98. > MULTI > SET foo bar > INCR baz > EXEC
99. Timestamps
100. { timestamp: “2012-08-03 15:00”, ip_address: “127.0.0.1” }
101. REDIS.zadd “visitors”, timestamp.to_i ip_address
102. REDIS.zrangebyscore “visitors”, Time.now.to_i - 86400 Time.now.to_i
103. • Pub / Sub • Authentication • Lua Scripting • Transactions
104. Have Fun!
105. Resources • http://redis.io • Homebrew • Redis Gem • Little Redis Book - Karl Seguin
106. http://inductionapp.com
107. Thanks!