Redis

REmote Dictionary Server

References:

• Why Your MySQL Needs Redis - Redis Labs

Table of contents

1. Overview
2. Redis Command

Overview

Challenge MySQL facing:

• minimize the cost of data management while scaling (more tables, more computing resources)
• need flexible data structure to catch up the market
• keep applications responsive (require sub-millisecond latency)
• disk based databased are inherently limited (how to overcome the physical limits?)

Redis:

• performance: The Most Powerful Database (highest throughput at lowest latency in high volume of writes scenario)
• simplicity: data structures
• extensibility: modules

Use cases:

• Caching
  • Look-aside (frequent reads, infrequent writes)
  • write-through
• Session: frequent reads and writes; data is isolated between sessions
• Metering: Rate-limiting
  • Built-in counter
  • TTL
  • single-threaded architecture assures serializability
• Fast Data Ingest
  • Pub/Sub; Lists; Sorted Sets

Redis Command

keys *       -- show all keys
exists [key]
del [key]
set [key] [value]
set [key] [value] nx  -- fail if the key already exists
set [key] [value] xx  -- fail if the key doesn't exist
mset [key1] [value1] [key2] [value2] ...
incr [key]      --- += 1
incr [key] [delta]
get [key]
mget [key1] [key2] ...
expires [key] [seconds]
ttl [key]
type [key]
object encoding [key]    -- show the data encoding format (opaque to user)

Redis doesn’t support namespace. In general, we use something like ns:table:id:attr as keys.