MySQL InnoDB Cluster and NDB Cluster

Copyright © 2016, Oracle and/or its affiliates. All rights reserved. |
MySQL High Availability
InnoDB Cluster and NDB Cluster
Mario Beck
mario.beck@oracle.com
Copyright © 2016, Oracle and/or its affiliates. All rights reserved.

Safe Harbor Statement
The following is intended to outline our general product direction. It is intended for
information purposes only, and may not be incorporated into any contract. It is not a
commitment to deliver any material, code, or functionality, and should not be relied upon
in making purchasing decisions. The development, release, and timing of any features or
functionality described for Oracle’s products remains at the sole discretion of Oracle.
2

Almost all organizations require their most
critical systems to be highly available
3
100%

InnoDB Cluster
4

InnoDB Cluster
5
App Servers with
MySQL Router
MySQL Group Replication
MySQL Shell
Setup, Manage,
Orchestrate
“High Availability becomes a core
first class feature of MySQL!”

Copyright © 2016, Oracle and/or its affiliates. All rights reserved. | 6
MySQL InnoDB Cluster: Vision
“A single product — MySQL — with high availability and scaling
features baked in; providing an integrated end-to-end solution
that is easy to use.”

• One Product: MySQL
– All components created together
– Tested together
– Packaged together
• Easy to Use
– One client: MySQL Shell
– Easy packaging
– Integrated orchestration
– Homogenous servers
• Flexible and Modern
– SQL and NoSQL together
– Protocol Buffers
– Developer friendly
• Support Read/Write Scale Out
– Sharded clusters
– Federated system of N replica sets
– Each replica set manages a shard
7
MySQL InnoDB Cluster: Goals

MySQL Connector
Application
MySQL Router
MySQL Connector
Application
MySQL Router
MySQL Shell
MySQL Connector
Application
MySQL Router
MySQL Connector
Application
MySQL Router
MySQL InnoDB Cluster: High Level Architecture
MySQL
InnoDB
cluster
MySQL Enterprise Monitor
…

MySQL
Group Replication
Natively distributed and highly available replica sets

MySQL Group Replication: What Is It?
• Group Replication library
– Implementation of Replicated Database State Machine
theory
• MySQL GCS is based on Paxos (variant of Mencius)
– Provides virtually synchronous replication for MySQL 5.7+
– Supported on all MySQL platforms
• Linux, Windows, Solaris, OSX, FreeBSD
“Multi-master update anywhere replication plugin for MySQL with
built-in conflict detection and resolution, automatic distributed
recovery, and group membership.”
10
App Servers with
MySQL Router
MySQL Shell
Setup, Manage,
Orchestrate

MySQL Group Replication: What Does It Provide?
• A highly available distributed MySQL database service
– Removes the need for manually handling server fail-over
– Provides distributed fault tolerance
– Enables Active/Active update anywhere setups
– Automates reconfiguration (adding/removing nodes, crashes, failures)
– Automatically detects and handles conflicts
11

MySQL Group Replication: Architecture
Node Types
R: Traffic routers/proxies: mysqlrouter, haproxy, sqlproxy, ...
M: mysqld nodes participating in Group Replication
12

MySQL Router
Transparent application connection routing

MySQL Router: 2.1
• Native support for InnoDB clusters
– Understands Group Replication topology
– Utilizes metadata schema stored on each member
• Bootstraps itself and sets up client routing for the InnoDB cluster
• Allows for intelligent client routing into the InnoDB cluster
• Supports multi-master and single primary modes
• Core improvements
– Logging
– Monitoring
– Performance
– Security
14
App Servers with
MySQL Router
MySQL Shell
Setup, Manage,
Orchestrate
labs.mysql.com

MySQL Shell
Single tool for development, setup,
management, orchestration, and monitoring

MySQL Shell
16
A single unified client for all administrative and operations tasks
App Servers with
MySQL Router
MySQL Shell
Setup, Manage,
Orchestrate
• Multi-Language: JavaScript, Python, and SQL
– Naturally scriptable
• Supports both Document and Relational models
• Exposes full Development and Admin API
”MySQL Shell provides the developer and DBA with a single intuitive,
flexible, and powerfull interface for all MySQL related tasks!”

MySQL Shell: Admin API
• mysql-js> dba.help()
• The global variable 'dba' is used to access the
MySQL AdminAPI
• Perform DBA operations
– Manage MySQL InnoDB clusters
• Create clusters
• Deploy MySQL instances
• Get cluster info
• Start/Stop MySQL Instances
• Validate MySQL instances …
Database Administration Interface
labs.mysql.com
App Servers with
MySQL Router
MySQL Shell
Setup, Manage,
Orchestrate

MySQL Shell – Deploy MySQL Instances
shell> mysqlsh
mysql-js> dba.deployLocalInstance(3306)
mysql-js> dba.deployInstance(‘hanode2:3306’)
mysql-js> dba.deployInstance(‘hanode3:3306’)
18

MySQL Shell – Create an InnoDB Cluster
mysql-js> connect root@hanode1:3306
mysql-js> cluster = dba.createCluster(‘NewAppCluster')
mysql-js> cluster.addInstance('root@hanode2:3306')
mysql-js> cluster.addInstance('root@hanode3:3306')
19

MySQL Shell – Add a MySQL Router
shell> mysqlrouter --bootstrap hanode1:3306
shell> mysqlrouter &
shell> mysqlsh --uri root@localhost:6446
20

MySQL Document Store
Relational & Document Models
MySQL HA
Out-Of-Box HA
Read Scale-Out
Async Replication + Auto Failover
Write Scale-Out
Sharding
S1
S2
S3
S4
* Download InnoDB Cluster Preview Release from labs.mysql.com
What’s Next

MySQL Enterprise Monitor
• Native holistic support for InnoDB clusters
– Topology views
– Detailed metrics and graphs
– Best Practice advice
• Monitoring of MySQL Routers
• Monitoring of Group Replication
22

NDB Cluster
23

Driving Database Requirements
9th April 2015 Copyright 2015, oracle and/or its affiliates. All rights reserved 24
Rock Solid
Availability
In-Memory
Real-Time
Performance
Extreme
Read & Write
Scalability
Elasticity

HLR / HSS
Billing, AuC, VLR
AuC, Call
Routing, Billing
Location
Updates
Pre & Post Paid
• Massive volumes of write traffic
• < 3ms database response
• Downtime & lost transactions = lost $
• Extremly cost sensitive market
MySQL Cluster in Action: http://bit.ly/oRI5tF
No Trade-Offs: Cellular Network

User Session
Session
management
Session
management
Location Updates,
Character movements, ...
• Massive volumes of write traffic
• < 3ms database response
• Downtime & lost transactions = lost $
• Extremly sensitive crowd
No Trade-Offs: Massive Parallel Online Games

Copyright © 2016, Oracle and/or its affiliates. All rights reserved. | Confidential – Oracle Internal/Restricted/Highly Restricted 27
No Trade-offs: Hadoop
NameNodes
Leader
HDFS Client
DataNodes
Hops.io
MySQL NDB Cluster Data Nodes
ClusterJ
NDB Cluster powering the
world‘s fastest and most
scalable Hadoop
filesystem.
Scaling Hadoop 20x!

MySQL Cluster Overview
• Memory optimized tables with durability
• Predictable Low-Latency, Bounded Access Time
REAL-TIME
• Auto-Sharding, Active-Active
• ACID Compliant, OLTP + Real-Time Analytics
HIGH SCALE, READS + WRITES
FULLY ELASTIC
• Active-Active, Shared nothing, no Single Point of Failure
• Self Healing + On-Line Operations
99.999% AVAILABILITY
• Key/Value + Complex, Relational Queries
• SQL + Memcached + JavaScript + Java + HTTP/REST & C++
SQL + NoSQL
• Open Source + Commercial Editions, Easy to use and deploy
• Commodity hardware + Management, Monitoring Tools
LOW TCO
• Add and remove storage and performance capacity in seconds
• Fully cloud enabled

NoSQL Access to MySQL Cluster data
Apps Apps Apps Apps Apps Apps Apps Apps Apps Apps Apps Apps
JPA
Cluster JPA
PHP Perl Python Ruby JDBC Cluster J JS Apache Memcached
MySQL JNI Node.JS mod_ndb ndb_eng
NDB API (C++)
MySQL Cluster Data Nodes

MySQL Cluster Architecture
Clients
Application Layer
Data Layer

Data Partitioning / Auto-sharding
25/11/2016 Copyright 2015, oracle and/or its affiliates. All rights reserved 31
User-id (PK) Service (PK) Data
1773467253 chat xxx
6257346892 chat xxx
1773467253 photos xxx
8235602099 reminders xxx
8437829249 location xxx

1773467253 chat xxx
6257346892 chat xxx
Shard Key

Node Group
Data Node 1
• Data transparently sharded between Node Groups
• Each fragment active in one Data Node with synchronous replication to
2nd Data Node in same Node Group
Data Node 2
Node Group
Data Node 3 Data Node 4

• DBA chooses which part of
Primary Key to use as shard
key
• Fragment for each row
decided by hashing the
sharding key
1773467253 chat xxx
6257346892 chat xxx
Shard Key
Node Group
Node Group

• Data automatically rearranged to use new capacity
• Designed to be a slow background process not impacting real-time
performance.
Node Group
Node Group
Node Group
On-line Scaling and Elasticity
Re-partioning

MySQL Cluster 7.2: 1.2 Billion UPDATEs per Minute
• NoSQL C++ API,
flexaSynch benchmark
• 30 x Intel E5-2600 Intel
Servers, 2 socket, 64GB
• ACID Transactions, with
Synchronous Replication
http://www.mysql.com/why-
mysql/white-papers/mysql-cluster-
benchmarks-1-billion-writes-per-minute/0
5
10
15
20
25
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
MillionsofUPDATEsperSecond

• Memory optimized tables
– Durable
– Mix with disk-based tables
• Massively concurrent OLTP
• Distributed Joins for analytics
• Parallel table scans for non-indexed
searches
• MySQL Cluster 7.4 FlexAsych
– 200M NoSQL Reads/Second
25/11/2016 38
MySQL Cluster 7.4 NoSQL Performance
200 Million NoSQL Reads/Second
Copyright 2015, oracle and/or its affiliates. All rights reserved
-
50,000,000
100,000,000
150,000,000
200,000,000
250,000,000
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32
Readspersecond
Data Nodes
FlexAsync Reads

• Memory optimized tables
– Durable
– Mix with disk-based tables
• Massively concurrent OLTP
• Distributed Joins for analytics
• Parallel table scans for non-indexed
searches
• MySQL Cluster 7.4 DBT2 BM
– 2.5M SQL Statements/Second
25/11/2016 39
MySQL Cluster 7.4 SQL Performance
2.5M SQL Statements/Second
Copyright 2015, oracle and/or its affiliates. All rights reserved
-
500,000
1,000,000
1,500,000
2,000,000
2,500,000
3,000,000
2 4 6 8 10 12 14 16
SQLStatements/sec
Data Nodes
DBT2 SQL Statements per Second

MySQL NDB Cluster 7.5
• Read Optimized
Tables
• Fully Replicated
Capacity and
Scale Out
• MySQL 5.7
• JSON Data Type
• Generated Columns
• Records-Per-Key
Optimization
Improved
SQL
• Improved Reporting
• Improved Logging
• Improved Visibility
• Improved Restore
Management
GA

• Reading from backup allows to read
from any copy
• Previously all reads were directed
towards the primary fragment only
Confidential – Oracle Internal/Restricted/Highly Restricted 41
MySQL NDB 7.5: Reading from backup
Node Group

MySQL NDB Cluster 7.5: Fully replicated
• Fully replicated allows a table to be read/written locally on any node
• Ideal for static data, faster join performance
Confidential – Oracle Internal/Restricted/Highly Restricted 42
Node Group
Node Group

Next Steps
Learn More
• www.mysql.com/cluster
• https://dev.mysql.com/doc/mysql-innodb-cluster/en/
• Authentic MySQL Curriculum: http://oracle.com/education/mysql
Try it Out
• dev.mysql.com/downloads/cluster/
7.4 GA and 7.5 DMR 1
• https://labs.mysql.com/
Let us know what you think
• http://mysqlhighavailability.com/
• forums.mysql.com/list.php?25

InnoDB Cluster
44
or
NDB Cluster

• MySQL InnoDB Cluster
– Easy HA built into MySQL 5.7+
– Write consistency
– Read Scalability
– App failover using MySQL Router
– Management via mysqlsh
– InnoDB storage engine
• MySQL NDB Cluster
– In memory database
– Automatic sharding
– Native access via several API
– Read/write consistency
– Read/write scalability
– NDB storage engine
How Do The Two Compare I

How Do The Two Compare II
MySQL InnoDB Cluster MySQL NDB Cluster
Storage Engine InnoDB NDBCLUSTER
Distributed Architecture Shared nothing Shared nothing
Consistency Model Weak Consistency Strong Consistency
Sharding No Yes
Arbitration No Yes
Load Balancing No Yes
NoSQL APIs MySQL Document Store Native NDB API
Operational Complexity Medium High
Administration Standard (MySQL) Custom (MySQL + NDB)

Thank you!

MySQL InnoDB Cluster and NDB Cluster

More Related Content

MySQL InnoDB Cluster and NDB Cluster