SlideShare une entreprise Scribd logo
1  sur  61
Télécharger pour lire hors ligne
What's new
#RedisTLV Jan 21st 2016
in v3.2
v3.2RC1 - TL;DR
~/src/redis$ git rev-list 3.0..3.2 --count
1606
~/src/redis$ git diff 3.0..3.2 --shortstat
262 files changed, 46931 insertions(+),
28720 deletions(-)
Today's meetup: ~6.379 topics
1. Geospatial indices
2. Quadtree & redimension
3. Internals deep dive
4. Effects-based replication
5. Security & `protected-mode`
6. Redis Lua debugger
About our sponsor
- Home of open source Redis
- The commercial provider for managed and
downloadable Redis solutions
- HQ @ Mountain View, R&D @ Italy & Israel
- ~70 employees & growing fast - we're hiring :)
The newsletter: Redis Watch
http://bit.ly/RedisWatch
Geospatial
indices
OSS crossover:
Redis 2.x -> ardb ->
2D spatial index ->
Matt Stancliff @mattsta ->
Salvatore Sanfilippo @antirez ->
Redis 3.2
Geospatial indices redhistory
The Geohash
- A geocoding system, hierarchical spatial grid
- The hash value maps to a location (lon, lat)
- Is usually base-32 encoded (.e.g sv8y8v66bt0)
- By Gustavo Niemeyer, in the public's domain
since 28 Feb. 2009
PrimeMeridian=0°
lon -180° lon 180°
0 1
0 11
10
Equator = 0°
lat -85.05112878
lat -85.05112878
0 110
10
111
Geospatial Indices in Redis
- Redis' Geohashes are 52-bit integers (~0.6m)
- Redis' Sorted Sets' scores are IEEE 754
floats, i.e. 53-bit integer precision…
BINGO!
New! GEO API, part 1
Add a point:
GEOADD key longitude latitude member [...]
Get longitude & latitude / geohash:
GEOPOS|GEOHASH key member [...]
Get the distance between two points:
GEODIST key member1 member2 [unit]
GEO API - "Demo"
127.0.0.1:6379> GEOADD g 34.84076 32.10942 RL@TLV
(integer) 1
127.0.0.1:6379> GEOADD g -122.0678325 37.3775256 RL@MV
(integer) 1
127.0.0.1:6379> GEODIST g RL@TLV RL@MV km
"11928.692170353959"
127.0.0.1:6379> GEOADD g 34.8380433 32.1098095 Hudson
(integer) 1
Size (almost) doesn't matter
127.0.0.1:6379> GEOHASH g RL@TLV Hudson
1) "sv8y8v66bt0"
2) "sv8y8v2m1n0"
- Shorter hashes -> "same" location, bigger area
- Close spatial proximity usually means a
shared hash prefix
New! GEO API, part 2
Search for members in a radial area:
GEORADIUS key longitude latitude radius unit ...
GEORADIUSBYMEMBER key member radius unit ...
Overthrows ZREVRANGEBYSCORE!!! #RedisTrivia
Delete a point - no GEOREM for you:
ZREM key member [...]
GEO Use cases
Any ideas?
Quadtree &
redimension
Multi-dimensional queries
SELECT id
FROM users
WHERE age > 35 AND
salary BETWEEN 250 AND 350
http://stackoverflow.
com/questions/32911604/intersection-of-two-or-
more-sorted-sets
The Redis way - "ZQUERY"
ZUNIONSTORE t 1 age WEIGHTS 1
ZREMRANGEBYSCORE t -inf (25
ZREMRANGEBYSCORE t (35 +inf
ZINTERSTORE t 2 t salary WEIGHTS 0 1
ZRANGEBYSCORE t 250 350
DEL t
Works, but not too efficient.
Would indexing the data help?
rqtih.lua: Another Redis Way
https://gist.github.
com/itamarhaber/c1ffda42d86b314ea701
rqtih.lua is about 32.5 times faster than
ZQUERY on 100K users (age & salary)
rqtih.lua?!? A PoC for
R - Redis, duh
QT - Quadtree
IH - In Hash
.LUA - "object oriented", JiT reads, delayed
writes
Trillustration
a
b
d
e
f
g
h
i
a
d b f
h g c e i
c
{
/ : x, y, w, h, {a}
/00/ : x, y, w, h, {d}
/01/ : x, y, w, h, {b}
...}
* Node capacity = 1
A new Redis data structure?
- Discussions in proximity to Redis Developers
Day 2015 (London)
- k-d tree: similar principles for k dimensions,
but complex complexity
- Outcomes: topics/indexes & experimental API
that uses existing data types (Zorted & Hash)
Redimension: k-d query API
@antirez's idea: interleave the dimensions, store
"score"+data in a Zorted Set for lexicographical
ranges, maintain a Hash for lookups
redimension.rb - implementation by @antirez
redimension.lua - port by @itamarhaber
Redimension "Demo"
~/src/lua-redimension$ redis-cli SCRIPT LOAD
"$(cat redimension.lua)"
"4abdad23c459145cbd658c991c0c8ad93d984d91"
~/src/lua-redimension$ redis-cli EVALSHA
4abdad23c459145cbd658c991c0c8ad93d984d91 0
1) "KEYS[1] - index sorted set key"
2) "KEYS[2] - index hash key"
3) "ARGV[1] - command. Can be:"
Redimension "Demo", 2
4) " create - create an index with ARGV
[2] as dimension and ARGV[3] as precision"
5) " drop - drops an index"
6) " index - index an element ARGV[2]
with ARGV[3]..ARGV[3+dimension] values"
7) " unindex - unindex an element ARGV
[2] with ARGV[3]..ARGV[3+dimension] values"
Redimension "Demo", 3
9) " update - update an element ARGV[2]
with ARGV[3]..ARGV[3+dimension] values"
10) " query - query using ranges ARGV
[2], ARGV[3]..ARGV[2+dimension-1], ARGV
[2+dimension]"
11) " fuzzy_test - fuzzily tests the library
on ARGV[2] dimension with ARGV[3] items using
ARGV[4] queries"
redimension.next()
- Currently just an experiment
- Many improvements still needed
- Planned to become a part of the core project
- Need more feedback WRT functionality & API
- Any ideas?
Internals deep dive
Oran Agra @RedisLabs
changes that made it (or didn’t) to OSS redis
● merged into 3.0
○ Fix a race condition in processCommand() with freeMemoryIfNeeded()
○ diskless replication fixes
○ psync fixes
○ fixes in LRU eviction (dict random keys during rehasing)
● merged into 3.2
○ sds optimizations
○ jemalloc size class optimization
● changes not merged yet
○ diskless slave replication
○ dict.c improvements
● other changes i didn’t get to push yet
○
nothing is user facing.
only optimizations, and fixes 8-
(
diskless replication
● how normal replication works.
master->fork->rdb on disk->main process streams to slave
slave->save to disk while serving clients->flushdb->load rdb
● disadvantages of diskless replication
○ slaves must connect together
○ slave side flush before RDB was fully received
○ on slow network, longer fork duration
● a word about fork() and CoW?
diskless replication benchmark (replication time)
two instances of r3.2xl (60GB ram, with 160GB SSD),
4,000,000 string keys of 1k random data.
(consuming 52GB of RAM), 19GB RDB file.
fully disk based: 513 seconds
only master diskless: 365 seconds
fully diskless: 231 seconds
only salve is diskless: 360 seconds
● we all know what fragmentation is
● history: on the search for the ultimate allocator
● how an allocator works (bins) to overcome that
● a word about virtual address space vs OS pages
○ RSS = VM pages mapped to physical RAM
● what’s internal fragmentation / used_memory
(maxmemory) includes internal frag
● RSS = used_memory (+external frag)
○ external frag are unused bins, and pages
●
unused
unused
fragmentation
allocators
16 byte bins pool
32 byte bins pool
internal fragmentation
22
byte
18
bytes
17
bytes
30
bytes
28
bytes
adding bin 24 bytes pull to jemalloc
used by: dictEntry, listNode, etc
redis-cli debug populate 10000000
original code's used_memory: 1,254,709,872
with patch used_memory: 1,094,714,048
memory optimization: 14%
size classes:
8
16
24
32
40
48
56
64
80
96
…
...
p1
FS
cache
kernel p1
FS
cache p2
p1+
p2
p2 p2 p2
FS
cache
FS
cache
FS
cache
physical ram (4k pages)
process 1(virtual address space) process 2(virtual address space)
unmap
ped
unmap
ped
unmap
ped
unmap
ped
unmap
ped
4k page
4k page
4k page
4k page
can be returned to os
(won’t be rss anymore)
4k page
ABCDEFn
char*
used unusedfree
4 bytes 4 bytes
old sds header
● grows in place(sometimes no need for realloc)
○ although realloc may nop instead of give new pointer and do
memcpy
● no need for strlen (search for null terminator)
● can be used in normal string functions like printf
struct sdshdr {
unsigned int len;
unsigned int free;
char buf[];
};
new sds header
ABCDEFn
char*
used unusedfree
4 bytes 4 bytes
old sds header
ABCDEFnused unusable
5 bits 3 bits
type5bit
ABCDEFnused unusedallocated
1 byte 1 byte 1 byte
type8bit
ABCDEFnused unusedallocated
2 bytes 2 bytes 1 byte
type16bit
ABCDEFnused unusedallocated
4 bytes 4 bytes 1 byte
type32bit
ABCDEFnused unusedallocated
8 bytes 8 bytes 1 byte
type64bit
struct __attribute__ ((__packed__)) sdshdr5 {
unsigned char flags; /* 3 lsb of type, and 5 msb of string length */
char buf[];
};
struct __attribute__ ((__packed__)) sdshdr8 {
uint8_t len; /* used */
uint8_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__ ((__packed__)) sdshdr16 {
uint16_t len; /* used */
uint16_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__ ((__packed__)) sdshdr32 {
uint32_t len; /* used */
uint32_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
struct __attribute__ ((__packed__)) sdshdr64 {
uint64_t len; /* used */
uint64_t alloc; /* excluding the header and null terminator */
unsigned char flags; /* 3 lsb of type, 5 unused bits */
char buf[];
};
sds size classes
debug populate 10000000
used_memory of original code: 1,254,709,872
used_memory with new code: 1,078,723,024
memory optimization: 16%
Intermission
600 seconds
Effects
replication
Script replication before v3.2
Lua scripts are pushed down to the slaves for
local execution. This reduces wire traffic in
cases such as:
for i = 1, 1000000 do
redis.call('LPUSH', KEYS[1], i)
end
Script replication caveats
Compute-intensive scripts (e.g. ZQUERY) waste
CPU time because they are run:
- 1+number of slaves times: wasteful
- When recovering from AOF: really bad
And then there's also...
Free will
vs.
> EVAL "redis.call('SET', KEYS[1], redis.call
('TIME')[1])" 1 foo
(error) ... Write commands
not allowed after non
deterministic commands
Script replication in v3.2
- Same defaults
- NEW! redis.replicate_commands()
causes the script's effects to be replicated
- NEW! redis.set_repl(...)
redis.REPL_[ALL|NONE|AOF|SLAVE]
Effect-based replication uses
Any ideas?
Security
"A few things about Redis security"
"The Redis security model is: it’s
totally insecure to let untrusted
clients access the system, please
protect it from the outside world yourself...
Let’s crack Redis for fun and no profit…"
HOWTO: http://antirez.com/news/96
The totally unexpected result
Script kiddies, cybercriminals and white hackers
3 critical points about security
Honesty is always the best option. That said:
1. Never leave an unprotected server open to
the outside world
2. If your server has been compromised, burn it
3. Always read the documentation
NEW! protected-mode directive
By default is enabled -> a breaking upgrade!
When (protected-mode && !requirepass && !bind):
- Allow only 127.0.0.1, ::1 or socket connections
- DENY (with the longest message ever!) others
Protection in Action - "Demo"
-DENIED Redis is running in protected mode because protected mode is enabled, no
bind address was specified, no authentication password is requested to clients. In
this mode connections are only accepted from the loopback interface. If you want to
connect from external computers to Redis you may adopt one of the following
solutions: 1) Just disable protected mode sending the command 'CONFIG SET protected-
mode no' from the loopback interface by connecting to Redis from the same host the
server is running, however MAKE SURE Redis is not publicly accessible from internet
if you do so. Use CONFIG REWRITE to make this change permanent. 2) Alternatively you
can just disable the protected mode by editing the Redis configuration file, and
setting the protected mode option to 'no', and then restarting the server. 3) If you
started the server manually just for testing, restart it with the '--protected-mode
no' option. 4) Setup a bind address or an authentication password. NOTE: You only
need to do one of the above things in order for the server to start accepting
connections from the outside.
Redis Lua
debugger
NEW! Integrated Lua debugger
- Step-by-step journey through history
- LDB: SCRIPT DEBUG yes/sync/no
- Demo: redis-cli, ZeroBrane Studio IDE plugin
http://redis.io/topics/ldb
https://redislabs.com/blog/zerobrane-studio-
plugin-for-redis-lua-scripts

Contenu connexe

Tendances

An Introduction to REDIS NoSQL database
An Introduction to REDIS NoSQL databaseAn Introduction to REDIS NoSQL database
An Introduction to REDIS NoSQL databaseAli MasudianPour
 
Redis 101 Data Structure
Redis 101 Data StructureRedis 101 Data Structure
Redis 101 Data StructureIsmaeel Enjreny
 
Redis modules 101
Redis modules 101Redis modules 101
Redis modules 101Dvir Volk
 
Redis - for duplicate detection on real time stream
Redis - for duplicate detection on real time streamRedis - for duplicate detection on real time stream
Redis - for duplicate detection on real time streamCodemotion
 
Redis Use Patterns (DevconTLV June 2014)
Redis Use Patterns (DevconTLV June 2014)Redis Use Patterns (DevconTLV June 2014)
Redis Use Patterns (DevconTLV June 2014)Itamar Haber
 
Power to the People: Redis Lua Scripts
Power to the People: Redis Lua ScriptsPower to the People: Redis Lua Scripts
Power to the People: Redis Lua ScriptsItamar Haber
 
Introduction to redis - version 2
Introduction to redis - version 2Introduction to redis - version 2
Introduction to redis - version 2Dvir Volk
 
Redis Functions, Data Structures for Web Scale Apps
Redis Functions, Data Structures for Web Scale AppsRedis Functions, Data Structures for Web Scale Apps
Redis Functions, Data Structures for Web Scale AppsDave Nielsen
 
A simple introduction to redis
A simple introduction to redisA simple introduction to redis
A simple introduction to redisZhichao Liang
 
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLab
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLabIntroduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLab
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLabCloudxLab
 
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2Antonios Giannopoulos
 
Sasi, cassandra on the full text search ride At Voxxed Day Belgrade 2016
Sasi, cassandra on the full text search ride At  Voxxed Day Belgrade 2016Sasi, cassandra on the full text search ride At  Voxxed Day Belgrade 2016
Sasi, cassandra on the full text search ride At Voxxed Day Belgrade 2016Duyhai Doan
 
Debugging & Tuning in Spark
Debugging & Tuning in SparkDebugging & Tuning in Spark
Debugging & Tuning in SparkShiao-An Yuan
 
Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to RedisArnab Mitra
 
Ceph Day KL - Bluestore
Ceph Day KL - Bluestore Ceph Day KL - Bluestore
Ceph Day KL - Bluestore Ceph Community
 
Fast track to getting started with DSE Max @ ING
Fast track to getting started with DSE Max @ INGFast track to getting started with DSE Max @ ING
Fast track to getting started with DSE Max @ INGDuyhai Doan
 
Building Scalable, Distributed Job Queues with Redis and Redis::Client
Building Scalable, Distributed Job Queues with Redis and Redis::ClientBuilding Scalable, Distributed Job Queues with Redis and Redis::Client
Building Scalable, Distributed Job Queues with Redis and Redis::ClientMike Friedman
 
A Brief Introduction to Redis
A Brief Introduction to RedisA Brief Introduction to Redis
A Brief Introduction to RedisCharles Anderson
 
Hive data migration (export/import)
Hive data migration (export/import)Hive data migration (export/import)
Hive data migration (export/import)Bopyo Hong
 

Tendances (20)

An Introduction to REDIS NoSQL database
An Introduction to REDIS NoSQL databaseAn Introduction to REDIS NoSQL database
An Introduction to REDIS NoSQL database
 
Redis 101 Data Structure
Redis 101 Data StructureRedis 101 Data Structure
Redis 101 Data Structure
 
Redis modules 101
Redis modules 101Redis modules 101
Redis modules 101
 
Redis and it's data types
Redis and it's data typesRedis and it's data types
Redis and it's data types
 
Redis - for duplicate detection on real time stream
Redis - for duplicate detection on real time streamRedis - for duplicate detection on real time stream
Redis - for duplicate detection on real time stream
 
Redis Use Patterns (DevconTLV June 2014)
Redis Use Patterns (DevconTLV June 2014)Redis Use Patterns (DevconTLV June 2014)
Redis Use Patterns (DevconTLV June 2014)
 
Power to the People: Redis Lua Scripts
Power to the People: Redis Lua ScriptsPower to the People: Redis Lua Scripts
Power to the People: Redis Lua Scripts
 
Introduction to redis - version 2
Introduction to redis - version 2Introduction to redis - version 2
Introduction to redis - version 2
 
Redis Functions, Data Structures for Web Scale Apps
Redis Functions, Data Structures for Web Scale AppsRedis Functions, Data Structures for Web Scale Apps
Redis Functions, Data Structures for Web Scale Apps
 
A simple introduction to redis
A simple introduction to redisA simple introduction to redis
A simple introduction to redis
 
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLab
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLabIntroduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLab
Introduction to Sqoop | Big Data Hadoop Spark Tutorial | CloudxLab
 
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2
New Indexing and Aggregation Pipeline Capabilities in MongoDB 4.2
 
Sasi, cassandra on the full text search ride At Voxxed Day Belgrade 2016
Sasi, cassandra on the full text search ride At  Voxxed Day Belgrade 2016Sasi, cassandra on the full text search ride At  Voxxed Day Belgrade 2016
Sasi, cassandra on the full text search ride At Voxxed Day Belgrade 2016
 
Debugging & Tuning in Spark
Debugging & Tuning in SparkDebugging & Tuning in Spark
Debugging & Tuning in Spark
 
Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to Redis
 
Ceph Day KL - Bluestore
Ceph Day KL - Bluestore Ceph Day KL - Bluestore
Ceph Day KL - Bluestore
 
Fast track to getting started with DSE Max @ ING
Fast track to getting started with DSE Max @ INGFast track to getting started with DSE Max @ ING
Fast track to getting started with DSE Max @ ING
 
Building Scalable, Distributed Job Queues with Redis and Redis::Client
Building Scalable, Distributed Job Queues with Redis and Redis::ClientBuilding Scalable, Distributed Job Queues with Redis and Redis::Client
Building Scalable, Distributed Job Queues with Redis and Redis::Client
 
A Brief Introduction to Redis
A Brief Introduction to RedisA Brief Introduction to Redis
A Brief Introduction to Redis
 
Hive data migration (export/import)
Hive data migration (export/import)Hive data migration (export/import)
Hive data migration (export/import)
 

En vedette

Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to RedisDvir Volk
 
A Weight Off Your Shoulders: MongoDB Atlas
A Weight Off Your Shoulders: MongoDB AtlasA Weight Off Your Shoulders: MongoDB Atlas
A Weight Off Your Shoulders: MongoDB AtlasMongoDB
 
Lua: the world's most infuriating language
Lua: the world's most infuriating languageLua: the world's most infuriating language
Lua: the world's most infuriating languagejgrahamc
 
Redis data design by usecase
Redis data design by usecaseRedis data design by usecase
Redis data design by usecaseKris Jeong
 
NoSQL CGN: Redis (05/2012)
NoSQL CGN: Redis (05/2012)NoSQL CGN: Redis (05/2012)
NoSQL CGN: Redis (05/2012)Sebastian Cohnen
 
Cloud computing
Cloud computingCloud computing
Cloud computingvcoulombe
 

En vedette (7)

Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to Redis
 
A Weight Off Your Shoulders: MongoDB Atlas
A Weight Off Your Shoulders: MongoDB AtlasA Weight Off Your Shoulders: MongoDB Atlas
A Weight Off Your Shoulders: MongoDB Atlas
 
Lua: the world's most infuriating language
Lua: the world's most infuriating languageLua: the world's most infuriating language
Lua: the world's most infuriating language
 
Redis introduction
Redis introductionRedis introduction
Redis introduction
 
Redis data design by usecase
Redis data design by usecaseRedis data design by usecase
Redis data design by usecase
 
NoSQL CGN: Redis (05/2012)
NoSQL CGN: Redis (05/2012)NoSQL CGN: Redis (05/2012)
NoSQL CGN: Redis (05/2012)
 
Cloud computing
Cloud computingCloud computing
Cloud computing
 

Similaire à What's new in Redis v3.2

MongoDB for Time Series Data Part 3: Sharding
MongoDB for Time Series Data Part 3: ShardingMongoDB for Time Series Data Part 3: Sharding
MongoDB for Time Series Data Part 3: ShardingMongoDB
 
SequoiaDB Distributed Relational Database
SequoiaDB Distributed Relational DatabaseSequoiaDB Distributed Relational Database
SequoiaDB Distributed Relational Databasewangzhonnew
 
Tuning and Debugging in Apache Spark
Tuning and Debugging in Apache SparkTuning and Debugging in Apache Spark
Tuning and Debugging in Apache SparkDatabricks
 
Nodejs性能分析优化和分布式设计探讨
Nodejs性能分析优化和分布式设计探讨Nodejs性能分析优化和分布式设计探讨
Nodejs性能分析优化和分布式设计探讨flyinweb
 
Web-Scale Graph Analytics with Apache Spark with Tim Hunter
Web-Scale Graph Analytics with Apache Spark with Tim HunterWeb-Scale Graph Analytics with Apache Spark with Tim Hunter
Web-Scale Graph Analytics with Apache Spark with Tim HunterDatabricks
 
Managing your black friday logs - Code Europe
Managing your black friday logs - Code EuropeManaging your black friday logs - Code Europe
Managing your black friday logs - Code EuropeDavid Pilato
 
MongoDB for Time Series Data: Sharding
MongoDB for Time Series Data: ShardingMongoDB for Time Series Data: Sharding
MongoDB for Time Series Data: ShardingMongoDB
 
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介クラウドDWHとしても進化を続けるPivotal Greenplumご紹介
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介Masayuki Matsushita
 
Challenging Web-Scale Graph Analytics with Apache Spark
Challenging Web-Scale Graph Analytics with Apache SparkChallenging Web-Scale Graph Analytics with Apache Spark
Challenging Web-Scale Graph Analytics with Apache SparkDatabricks
 
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui Meng
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui MengChallenging Web-Scale Graph Analytics with Apache Spark with Xiangrui Meng
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui MengDatabricks
 
Managing your Black Friday Logs NDC Oslo
Managing your  Black Friday Logs NDC OsloManaging your  Black Friday Logs NDC Oslo
Managing your Black Friday Logs NDC OsloDavid Pilato
 
Индексируем базу: как делать хорошо и не делать плохо Winter saint p 2021 m...
Индексируем базу: как делать хорошо и не делать плохо   Winter saint p 2021 m...Индексируем базу: как делать хорошо и не делать плохо   Winter saint p 2021 m...
Индексируем базу: как делать хорошо и не делать плохо Winter saint p 2021 m...Андрей Новиков
 
RedisConf18 - Redis Memory Optimization
RedisConf18 - Redis Memory OptimizationRedisConf18 - Redis Memory Optimization
RedisConf18 - Redis Memory OptimizationRedis Labs
 
r2con 2017 r2cLEMENCy
r2con 2017 r2cLEMENCyr2con 2017 r2cLEMENCy
r2con 2017 r2cLEMENCyRay Song
 
LSFMM 2019 BPF Observability
LSFMM 2019 BPF ObservabilityLSFMM 2019 BPF Observability
LSFMM 2019 BPF ObservabilityBrendan Gregg
 
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...Rob Skillington
 
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...Databricks
 
TiDB vs Aurora.pdf
TiDB vs Aurora.pdfTiDB vs Aurora.pdf
TiDB vs Aurora.pdfssuser3fb50b
 

Similaire à What's new in Redis v3.2 (20)

MongoDB for Time Series Data Part 3: Sharding
MongoDB for Time Series Data Part 3: ShardingMongoDB for Time Series Data Part 3: Sharding
MongoDB for Time Series Data Part 3: Sharding
 
SequoiaDB Distributed Relational Database
SequoiaDB Distributed Relational DatabaseSequoiaDB Distributed Relational Database
SequoiaDB Distributed Relational Database
 
Tuning and Debugging in Apache Spark
Tuning and Debugging in Apache SparkTuning and Debugging in Apache Spark
Tuning and Debugging in Apache Spark
 
Nodejs性能分析优化和分布式设计探讨
Nodejs性能分析优化和分布式设计探讨Nodejs性能分析优化和分布式设计探讨
Nodejs性能分析优化和分布式设计探讨
 
Web-Scale Graph Analytics with Apache Spark with Tim Hunter
Web-Scale Graph Analytics with Apache Spark with Tim HunterWeb-Scale Graph Analytics with Apache Spark with Tim Hunter
Web-Scale Graph Analytics with Apache Spark with Tim Hunter
 
Programar para GPUs
Programar para GPUsProgramar para GPUs
Programar para GPUs
 
Managing your black friday logs - Code Europe
Managing your black friday logs - Code EuropeManaging your black friday logs - Code Europe
Managing your black friday logs - Code Europe
 
MongoDB for Time Series Data: Sharding
MongoDB for Time Series Data: ShardingMongoDB for Time Series Data: Sharding
MongoDB for Time Series Data: Sharding
 
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介クラウドDWHとしても進化を続けるPivotal Greenplumご紹介
クラウドDWHとしても進化を続けるPivotal Greenplumご紹介
 
Challenging Web-Scale Graph Analytics with Apache Spark
Challenging Web-Scale Graph Analytics with Apache SparkChallenging Web-Scale Graph Analytics with Apache Spark
Challenging Web-Scale Graph Analytics with Apache Spark
 
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui Meng
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui MengChallenging Web-Scale Graph Analytics with Apache Spark with Xiangrui Meng
Challenging Web-Scale Graph Analytics with Apache Spark with Xiangrui Meng
 
Managing your Black Friday Logs NDC Oslo
Managing your  Black Friday Logs NDC OsloManaging your  Black Friday Logs NDC Oslo
Managing your Black Friday Logs NDC Oslo
 
Индексируем базу: как делать хорошо и не делать плохо Winter saint p 2021 m...
Индексируем базу: как делать хорошо и не делать плохо   Winter saint p 2021 m...Индексируем базу: как делать хорошо и не делать плохо   Winter saint p 2021 m...
Индексируем базу: как делать хорошо и не делать плохо Winter saint p 2021 m...
 
RedisConf18 - Redis Memory Optimization
RedisConf18 - Redis Memory OptimizationRedisConf18 - Redis Memory Optimization
RedisConf18 - Redis Memory Optimization
 
r2con 2017 r2cLEMENCy
r2con 2017 r2cLEMENCyr2con 2017 r2cLEMENCy
r2con 2017 r2cLEMENCy
 
Quick Wins
Quick WinsQuick Wins
Quick Wins
 
LSFMM 2019 BPF Observability
LSFMM 2019 BPF ObservabilityLSFMM 2019 BPF Observability
LSFMM 2019 BPF Observability
 
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...
FOSDEM 2019: M3, Prometheus and Graphite with metrics and monitoring in an in...
 
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...
Spark SQL Catalyst Code Optimization using Function Outlining with Kavana Bha...
 
TiDB vs Aurora.pdf
TiDB vs Aurora.pdfTiDB vs Aurora.pdf
TiDB vs Aurora.pdf
 

Plus de Itamar Haber

Redis v5 & Streams
Redis v5 & StreamsRedis v5 & Streams
Redis v5 & StreamsItamar Haber
 
Redis Modules API - an introduction
Redis Modules API - an introductionRedis Modules API - an introduction
Redis Modules API - an introductionItamar Haber
 
Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to RedisItamar Haber
 
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...Itamar Haber
 
Redis Streams - Fiverr Tech5 meetup
Redis Streams - Fiverr Tech5 meetupRedis Streams - Fiverr Tech5 meetup
Redis Streams - Fiverr Tech5 meetupItamar Haber
 
Why Your MongoDB Needs Redis
Why Your MongoDB Needs RedisWhy Your MongoDB Needs Redis
Why Your MongoDB Needs RedisItamar Haber
 
Redis & MongoDB: Stop Big Data Indigestion Before It Starts
Redis & MongoDB: Stop Big Data Indigestion Before It StartsRedis & MongoDB: Stop Big Data Indigestion Before It Starts
Redis & MongoDB: Stop Big Data Indigestion Before It StartsItamar Haber
 
Benchmarking Redis by itself and versus other NoSQL databases
Benchmarking Redis by itself and versus other NoSQL databasesBenchmarking Redis by itself and versus other NoSQL databases
Benchmarking Redis by itself and versus other NoSQL databasesItamar Haber
 

Plus de Itamar Haber (9)

Redis v5 & Streams
Redis v5 & StreamsRedis v5 & Streams
Redis v5 & Streams
 
Redis Modules API - an introduction
Redis Modules API - an introductionRedis Modules API - an introduction
Redis Modules API - an introduction
 
Redis Lua Scripts
Redis Lua ScriptsRedis Lua Scripts
Redis Lua Scripts
 
Introduction to Redis
Introduction to RedisIntroduction to Redis
Introduction to Redis
 
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...
How I Implemented the #1 Requested Feature In Redis In Less than 1 Hour with ...
 
Redis Streams - Fiverr Tech5 meetup
Redis Streams - Fiverr Tech5 meetupRedis Streams - Fiverr Tech5 meetup
Redis Streams - Fiverr Tech5 meetup
 
Why Your MongoDB Needs Redis
Why Your MongoDB Needs RedisWhy Your MongoDB Needs Redis
Why Your MongoDB Needs Redis
 
Redis & MongoDB: Stop Big Data Indigestion Before It Starts
Redis & MongoDB: Stop Big Data Indigestion Before It StartsRedis & MongoDB: Stop Big Data Indigestion Before It Starts
Redis & MongoDB: Stop Big Data Indigestion Before It Starts
 
Benchmarking Redis by itself and versus other NoSQL databases
Benchmarking Redis by itself and versus other NoSQL databasesBenchmarking Redis by itself and versus other NoSQL databases
Benchmarking Redis by itself and versus other NoSQL databases
 

Dernier

Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesSinan KOZAK
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024The Digital Insurer
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Scriptwesley chun
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024Results
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...Neo4j
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking MenDelhi Call girls
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024Rafal Los
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonetsnaman860154
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Servicegiselly40
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slidevu2urc
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking MenDelhi Call girls
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...apidays
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfEnterprise Knowledge
 
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsTop 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsRoshan Dwivedi
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 

Dernier (20)

Unblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen FramesUnblocking The Main Thread Solving ANRs and Frozen Frames
Unblocking The Main Thread Solving ANRs and Frozen Frames
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024Partners Life - Insurer Innovation Award 2024
Partners Life - Insurer Innovation Award 2024
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024A Call to Action for Generative AI in 2024
A Call to Action for Generative AI in 2024
 
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...Workshop - Best of Both Worlds_ Combine  KG and Vector search for  enhanced R...
Workshop - Best of Both Worlds_ Combine KG and Vector search for enhanced R...
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men08448380779 Call Girls In Greater Kailash - I Women Seeking Men
08448380779 Call Girls In Greater Kailash - I Women Seeking Men
 
The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024The 7 Things I Know About Cyber Security After 25 Years | April 2024
The 7 Things I Know About Cyber Security After 25 Years | April 2024
 
How to convert PDF to text with Nanonets
How to convert PDF to text with NanonetsHow to convert PDF to text with Nanonets
How to convert PDF to text with Nanonets
 
CNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of ServiceCNv6 Instructor Chapter 6 Quality of Service
CNv6 Instructor Chapter 6 Quality of Service
 
Histor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slideHistor y of HAM Radio presentation slide
Histor y of HAM Radio presentation slide
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men08448380779 Call Girls In Friends Colony Women Seeking Men
08448380779 Call Girls In Friends Colony Women Seeking Men
 
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
Apidays Singapore 2024 - Building Digital Trust in a Digital Economy by Veron...
 
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdfThe Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
The Role of Taxonomy and Ontology in Semantic Layers - Heather Hedden.pdf
 
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live StreamsTop 5 Benefits OF Using Muvi Live Paywall For Live Streams
Top 5 Benefits OF Using Muvi Live Paywall For Live Streams
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 

What's new in Redis v3.2

  • 1. What's new #RedisTLV Jan 21st 2016 in v3.2
  • 2. v3.2RC1 - TL;DR ~/src/redis$ git rev-list 3.0..3.2 --count 1606 ~/src/redis$ git diff 3.0..3.2 --shortstat 262 files changed, 46931 insertions(+), 28720 deletions(-)
  • 3. Today's meetup: ~6.379 topics 1. Geospatial indices 2. Quadtree & redimension 3. Internals deep dive 4. Effects-based replication 5. Security & `protected-mode` 6. Redis Lua debugger
  • 4. About our sponsor - Home of open source Redis - The commercial provider for managed and downloadable Redis solutions - HQ @ Mountain View, R&D @ Italy & Israel - ~70 employees & growing fast - we're hiring :)
  • 5. The newsletter: Redis Watch http://bit.ly/RedisWatch
  • 7. OSS crossover: Redis 2.x -> ardb -> 2D spatial index -> Matt Stancliff @mattsta -> Salvatore Sanfilippo @antirez -> Redis 3.2 Geospatial indices redhistory
  • 8. The Geohash - A geocoding system, hierarchical spatial grid - The hash value maps to a location (lon, lat) - Is usually base-32 encoded (.e.g sv8y8v66bt0) - By Gustavo Niemeyer, in the public's domain since 28 Feb. 2009
  • 9.
  • 11. 0 1
  • 12. 0 11 10 Equator = 0° lat -85.05112878 lat -85.05112878
  • 14. Geospatial Indices in Redis - Redis' Geohashes are 52-bit integers (~0.6m) - Redis' Sorted Sets' scores are IEEE 754 floats, i.e. 53-bit integer precision… BINGO!
  • 15. New! GEO API, part 1 Add a point: GEOADD key longitude latitude member [...] Get longitude & latitude / geohash: GEOPOS|GEOHASH key member [...] Get the distance between two points: GEODIST key member1 member2 [unit]
  • 16. GEO API - "Demo" 127.0.0.1:6379> GEOADD g 34.84076 32.10942 RL@TLV (integer) 1 127.0.0.1:6379> GEOADD g -122.0678325 37.3775256 RL@MV (integer) 1 127.0.0.1:6379> GEODIST g RL@TLV RL@MV km "11928.692170353959" 127.0.0.1:6379> GEOADD g 34.8380433 32.1098095 Hudson (integer) 1
  • 17. Size (almost) doesn't matter 127.0.0.1:6379> GEOHASH g RL@TLV Hudson 1) "sv8y8v66bt0" 2) "sv8y8v2m1n0" - Shorter hashes -> "same" location, bigger area - Close spatial proximity usually means a shared hash prefix
  • 18. New! GEO API, part 2 Search for members in a radial area: GEORADIUS key longitude latitude radius unit ... GEORADIUSBYMEMBER key member radius unit ... Overthrows ZREVRANGEBYSCORE!!! #RedisTrivia Delete a point - no GEOREM for you: ZREM key member [...]
  • 21. Multi-dimensional queries SELECT id FROM users WHERE age > 35 AND salary BETWEEN 250 AND 350 http://stackoverflow. com/questions/32911604/intersection-of-two-or- more-sorted-sets
  • 22. The Redis way - "ZQUERY" ZUNIONSTORE t 1 age WEIGHTS 1 ZREMRANGEBYSCORE t -inf (25 ZREMRANGEBYSCORE t (35 +inf ZINTERSTORE t 2 t salary WEIGHTS 0 1 ZRANGEBYSCORE t 250 350 DEL t Works, but not too efficient.
  • 23. Would indexing the data help? rqtih.lua: Another Redis Way https://gist.github. com/itamarhaber/c1ffda42d86b314ea701 rqtih.lua is about 32.5 times faster than ZQUERY on 100K users (age & salary)
  • 24. rqtih.lua?!? A PoC for R - Redis, duh QT - Quadtree IH - In Hash .LUA - "object oriented", JiT reads, delayed writes
  • 25. Trillustration a b d e f g h i a d b f h g c e i c { / : x, y, w, h, {a} /00/ : x, y, w, h, {d} /01/ : x, y, w, h, {b} ...} * Node capacity = 1
  • 26. A new Redis data structure? - Discussions in proximity to Redis Developers Day 2015 (London) - k-d tree: similar principles for k dimensions, but complex complexity - Outcomes: topics/indexes & experimental API that uses existing data types (Zorted & Hash)
  • 27. Redimension: k-d query API @antirez's idea: interleave the dimensions, store "score"+data in a Zorted Set for lexicographical ranges, maintain a Hash for lookups redimension.rb - implementation by @antirez redimension.lua - port by @itamarhaber
  • 28. Redimension "Demo" ~/src/lua-redimension$ redis-cli SCRIPT LOAD "$(cat redimension.lua)" "4abdad23c459145cbd658c991c0c8ad93d984d91" ~/src/lua-redimension$ redis-cli EVALSHA 4abdad23c459145cbd658c991c0c8ad93d984d91 0 1) "KEYS[1] - index sorted set key" 2) "KEYS[2] - index hash key" 3) "ARGV[1] - command. Can be:"
  • 29. Redimension "Demo", 2 4) " create - create an index with ARGV [2] as dimension and ARGV[3] as precision" 5) " drop - drops an index" 6) " index - index an element ARGV[2] with ARGV[3]..ARGV[3+dimension] values" 7) " unindex - unindex an element ARGV [2] with ARGV[3]..ARGV[3+dimension] values"
  • 30. Redimension "Demo", 3 9) " update - update an element ARGV[2] with ARGV[3]..ARGV[3+dimension] values" 10) " query - query using ranges ARGV [2], ARGV[3]..ARGV[2+dimension-1], ARGV [2+dimension]" 11) " fuzzy_test - fuzzily tests the library on ARGV[2] dimension with ARGV[3] items using ARGV[4] queries"
  • 31. redimension.next() - Currently just an experiment - Many improvements still needed - Planned to become a part of the core project - Need more feedback WRT functionality & API - Any ideas?
  • 32. Internals deep dive Oran Agra @RedisLabs
  • 33. changes that made it (or didn’t) to OSS redis ● merged into 3.0 ○ Fix a race condition in processCommand() with freeMemoryIfNeeded() ○ diskless replication fixes ○ psync fixes ○ fixes in LRU eviction (dict random keys during rehasing) ● merged into 3.2 ○ sds optimizations ○ jemalloc size class optimization ● changes not merged yet ○ diskless slave replication ○ dict.c improvements ● other changes i didn’t get to push yet ○
  • 34. nothing is user facing. only optimizations, and fixes 8- (
  • 35. diskless replication ● how normal replication works. master->fork->rdb on disk->main process streams to slave slave->save to disk while serving clients->flushdb->load rdb ● disadvantages of diskless replication ○ slaves must connect together ○ slave side flush before RDB was fully received ○ on slow network, longer fork duration ● a word about fork() and CoW?
  • 36. diskless replication benchmark (replication time) two instances of r3.2xl (60GB ram, with 160GB SSD), 4,000,000 string keys of 1k random data. (consuming 52GB of RAM), 19GB RDB file. fully disk based: 513 seconds only master diskless: 365 seconds fully diskless: 231 seconds only salve is diskless: 360 seconds
  • 37. ● we all know what fragmentation is ● history: on the search for the ultimate allocator ● how an allocator works (bins) to overcome that ● a word about virtual address space vs OS pages ○ RSS = VM pages mapped to physical RAM ● what’s internal fragmentation / used_memory (maxmemory) includes internal frag ● RSS = used_memory (+external frag) ○ external frag are unused bins, and pages ●
  • 39. allocators 16 byte bins pool 32 byte bins pool internal fragmentation 22 byte 18 bytes 17 bytes 30 bytes 28 bytes
  • 40. adding bin 24 bytes pull to jemalloc used by: dictEntry, listNode, etc redis-cli debug populate 10000000 original code's used_memory: 1,254,709,872 with patch used_memory: 1,094,714,048 memory optimization: 14% size classes: 8 16 24 32 40 48 56 64 80 96 … ...
  • 41. p1 FS cache kernel p1 FS cache p2 p1+ p2 p2 p2 p2 FS cache FS cache FS cache physical ram (4k pages) process 1(virtual address space) process 2(virtual address space) unmap ped unmap ped unmap ped unmap ped unmap ped
  • 42. 4k page 4k page 4k page 4k page can be returned to os (won’t be rss anymore) 4k page
  • 43. ABCDEFn char* used unusedfree 4 bytes 4 bytes old sds header ● grows in place(sometimes no need for realloc) ○ although realloc may nop instead of give new pointer and do memcpy ● no need for strlen (search for null terminator) ● can be used in normal string functions like printf struct sdshdr { unsigned int len; unsigned int free; char buf[]; };
  • 44. new sds header ABCDEFn char* used unusedfree 4 bytes 4 bytes old sds header ABCDEFnused unusable 5 bits 3 bits type5bit ABCDEFnused unusedallocated 1 byte 1 byte 1 byte type8bit ABCDEFnused unusedallocated 2 bytes 2 bytes 1 byte type16bit ABCDEFnused unusedallocated 4 bytes 4 bytes 1 byte type32bit ABCDEFnused unusedallocated 8 bytes 8 bytes 1 byte type64bit struct __attribute__ ((__packed__)) sdshdr5 { unsigned char flags; /* 3 lsb of type, and 5 msb of string length */ char buf[]; }; struct __attribute__ ((__packed__)) sdshdr8 { uint8_t len; /* used */ uint8_t alloc; /* excluding the header and null terminator */ unsigned char flags; /* 3 lsb of type, 5 unused bits */ char buf[]; }; struct __attribute__ ((__packed__)) sdshdr16 { uint16_t len; /* used */ uint16_t alloc; /* excluding the header and null terminator */ unsigned char flags; /* 3 lsb of type, 5 unused bits */ char buf[]; }; struct __attribute__ ((__packed__)) sdshdr32 { uint32_t len; /* used */ uint32_t alloc; /* excluding the header and null terminator */ unsigned char flags; /* 3 lsb of type, 5 unused bits */ char buf[]; }; struct __attribute__ ((__packed__)) sdshdr64 { uint64_t len; /* used */ uint64_t alloc; /* excluding the header and null terminator */ unsigned char flags; /* 3 lsb of type, 5 unused bits */ char buf[]; };
  • 45. sds size classes debug populate 10000000 used_memory of original code: 1,254,709,872 used_memory with new code: 1,078,723,024 memory optimization: 16%
  • 47.
  • 49. Script replication before v3.2 Lua scripts are pushed down to the slaves for local execution. This reduces wire traffic in cases such as: for i = 1, 1000000 do redis.call('LPUSH', KEYS[1], i) end
  • 50. Script replication caveats Compute-intensive scripts (e.g. ZQUERY) waste CPU time because they are run: - 1+number of slaves times: wasteful - When recovering from AOF: really bad And then there's also...
  • 51. Free will vs. > EVAL "redis.call('SET', KEYS[1], redis.call ('TIME')[1])" 1 foo (error) ... Write commands not allowed after non deterministic commands
  • 52. Script replication in v3.2 - Same defaults - NEW! redis.replicate_commands() causes the script's effects to be replicated - NEW! redis.set_repl(...) redis.REPL_[ALL|NONE|AOF|SLAVE]
  • 55. "A few things about Redis security" "The Redis security model is: it’s totally insecure to let untrusted clients access the system, please protect it from the outside world yourself... Let’s crack Redis for fun and no profit…" HOWTO: http://antirez.com/news/96
  • 56. The totally unexpected result Script kiddies, cybercriminals and white hackers
  • 57. 3 critical points about security Honesty is always the best option. That said: 1. Never leave an unprotected server open to the outside world 2. If your server has been compromised, burn it 3. Always read the documentation
  • 58. NEW! protected-mode directive By default is enabled -> a breaking upgrade! When (protected-mode && !requirepass && !bind): - Allow only 127.0.0.1, ::1 or socket connections - DENY (with the longest message ever!) others
  • 59. Protection in Action - "Demo" -DENIED Redis is running in protected mode because protected mode is enabled, no bind address was specified, no authentication password is requested to clients. In this mode connections are only accepted from the loopback interface. If you want to connect from external computers to Redis you may adopt one of the following solutions: 1) Just disable protected mode sending the command 'CONFIG SET protected- mode no' from the loopback interface by connecting to Redis from the same host the server is running, however MAKE SURE Redis is not publicly accessible from internet if you do so. Use CONFIG REWRITE to make this change permanent. 2) Alternatively you can just disable the protected mode by editing the Redis configuration file, and setting the protected mode option to 'no', and then restarting the server. 3) If you started the server manually just for testing, restart it with the '--protected-mode no' option. 4) Setup a bind address or an authentication password. NOTE: You only need to do one of the above things in order for the server to start accepting connections from the outside.
  • 61. NEW! Integrated Lua debugger - Step-by-step journey through history - LDB: SCRIPT DEBUG yes/sync/no - Demo: redis-cli, ZeroBrane Studio IDE plugin http://redis.io/topics/ldb https://redislabs.com/blog/zerobrane-studio- plugin-for-redis-lua-scripts