Visit Amazon memorydb and build your own redis memory database

Speaking of cloud vendors , Your first reaction may be Alibaba cloud 、 Hua Wei Yun 、 Tencent, cloud, etc. , Alibaba cloud has the largest market share in China , But the global market share of Cloud Computing , Amazon is the big brother ：

At present, Amazon cloud technology provides 100 Of more than products   Free package  . among , Computing resources Amazon EC2 First year 12 It's free for three months ,750 Hours / month ; Storage resources Amazon S3 First year 12 It's free for three months ,5GB Standard storage capacity ; Database resources Amazon RDS First year 12 It's free for three months ,750 Hours ;Amazon Dynamo DB 25GB Storage capacity is permanently free, etc .

It has   Three different types of free offers   To choose from ：

• Free trial
• [ The short-term free trial offer is calculated from the date you activate a specific service ](aws.amazon.com/free/?all-f… Tier Types=tier%23trial&awsf.Free Tier Categories=*all#Free_Tier_details)
• 12 It's free for three months
• [ Since initial registration AWS From 12 Use these products for free within months ](aws.amazon.com/free/?all-f… Tier Types=tier%2312monthsfree&awsf.Free Tier Categories=*all#Free_Tier_details)
• Free forever
• [ These free package products will not expire , And applicable to all AWS Customer ](aws.amazon.com/free/?all-f… Tier Types=tier%23always-free&awsf.Free Tier Categories=*all&all-free-tier.sort-order=asc&awsf.Free Tier Types=*all&awsf.Free Tier Categories=*all#Free_Tier_details)

There's no formula , Pure is white whoring , AWS Free package , White whoring entrance

This article will be  Amazon MemoryDB  For example , Describe how to start from scratch and build your own Redis In-memory database .

MemoryDB advantage

Build a traditional self built Redis, Performance needs to be considered 、 Data persistence 、 Extensibility 、 A series of building environments such as firewalls , The steps are cumbersome , laborious . Build through cloud database Redis, from   efficiency 、 performance 、 Data security 、 Extensibility   And so on , It solves the pain points of traditional methods , Save a lot of time and operation and maintenance costs . Such as ：

adopt Redis Fast build

By being Stack Overflow named “ The most popular ” The database of Redis To quickly build applications . Use flexible Redis Data structure and API（ for example , Streaming 、 Lists and collections ）.

Achieve ultra fast performance

Access data with microsecond read latency and single digit millisecond write latency and high throughput .MemoryDB It can handle more than... Every day 13 Trillions of requests , And supports more than per second 1.6 The peak of billion requests .

Store data with persistence and high availability

MemoryDB Store data in memory , And use the transaction log of multiple availability zones to realize rapid database recovery and restart , Without losing data . because MemoryDB Data can be stored persistently , So you can use it as the primary database .

\

Easily expand

MemoryDB The storage space of each cluster can be changed from several GB Seamless expansion to 100 TB above , To meet your application needs .

Registered account

No, AWS Account partners need to register their accounts first ：  Registration and trial address （ On the bill registration page, you can use   Domestic credit cards  ）

Log in to the console

Console login portal

Enter your email address and password

When you come in Service There's so much content , For convenience , We can search here like the following figure MemoryDB, Select the first option for search results

You can see the overall interface , Click... To start using ：

MemoryDB Hands on Tutorial

Amazon MemoryDB for Redis It's a   And Redis compatible  、 A very persistent in memory database service , Ultra fast performance . Provide submillisecond latency for modern applications with a microservice architecture 、 High throughput and multi availability persistence .

Official documents ：
docs.aws.amazon.com/zh_cn/memor…

Create clusters

Select cluster... In the left navigation pane , And then click   Create clusters

• Cluster information

• Enter the name of the cluster   name  、 Description information

• \

• Subnet group

• Create a new   subnet   Group , Or select an existing subnet group to apply to this cluster from the available list . A subnet is similar to a local area network

• \

• Cluster settings

• Use  t4g.small  example ,2 Free within months （ Provide free instances and... Every month 20GB data ）

• • Apply to  Redis Version compatibility   in , Accept the default value  6.2
  • Apply to   port  , Accept the default Redis port 6379
  • Apply to   Parameter group   in , Accept  default.memorydb-redis6  Parameter group

• \

• Security

• Security group - act as   A firewall   To control network access to the cluster ,  Very important ！！！

• Static encryption – Enable encryption for data stored on disk

• After setting , Finally, click create

• \

When your cluster status is  available  when , To whom EC2 Access right , Connect to the cluster and start using it .

Grant access to

• establish EC2 example
• This section assumes that you are already familiar with Amazon EC2 Instance startup and connection . For more information , see also Amazon EC2 Getting started .
• Key pair ( Sign in ) Information , Never choose 【 Continue without a key pair 】 The option to , Otherwise, it will lead to EC2 After the instance is created, it cannot be accessed directly
• Authorized access
• adopt   Security group   Configure authorized access .
• all MemoryDB The purpose of clustering is to Amazon EC2 Instance to access . The most common situation is from the same Amazon Virtual Private Cloud (Amazon VPC) Medium Amazon EC2 Instance access MemoryDB colony . You have to authorize EC2 Instance access cluster , Then you can start from EC2 The instance is connected to the cluster .
• Here's a demonstration , Configure full inbound access directly , 0.0.0.0/0 That is, all devices can access
• \

Connect clusters

From you to MemoryDB Access data in nodes , You can use the secure socket layer (SSL) The client of , It can also be in Amazon Linux 2 Use on with TLS/SSL Of redis-cli.

If you want to use redis-cli Connect to Amazon Linux 2 Upper MemoryDB colony , Steps are as follows ：

• Sign in EC2 Command line console

• choice EC2 Instance Connect Connection type

• \

• Download and compile redis-cli Utilities

• stay EC2 Instance at the command prompt , Type the following command

• #Amazon Linux 2 $sudo yum -y install openssl-devel gcc$ wget download.redis.io/redis-stabl…$tar xvzf redis-stable.tar.gz$ cd redis-stable $make distclean$ make redis-cli BUILD_TLS=yes $ sudo install -m 755 src/redis-cli /usr/local/bin/

• stay EC2 Instance at the command prompt , Type the following command , And replace the corresponding content shown in this example with the terminal node of your cluster and port

#  Example 
# src/redis-cli -c -h Cluster Endpoint --tls -p 6379
$ src/redis-cli -c -h clustercfg.redis-free.uyejvs.memorydb.ap-southeast-1.amazonaws.com --tls -p 6379

among  Cluster Endpoint  be located MemoryDB Cluster information   Cluster endpoint

The actual operation results are as follows ：

be limited to Redis command

To provide a hosted service experience ,MemoryDB Restricted access to certain commands that require advanced privileges . The following commands are not available ：

acl deluser
acl load
acl save
acl setuser
bgrewriteaof
bgsave
cluster addslot
cluster delslot
cluster setslot
config
debug
migrate
module
psync
replicaof
save
shutdown
slaveof
sync

MemoryDB And Redis compatible , Is a very popular open source data storage , Enables you to use the same flexible and friendly Redis data structure 、API And commands to quickly build applications . Use MemoryDB, All your data is stored in memory , This enables you to achieve microsecond read and write latency per millisecond and high throughput .MemoryDB It also uses multi zone transaction logs to span multiple zones (AZ) Persistent storage of data , For fast failover 、 Database recovery and node restart .

Memory DB Both in memory performance and multi availability persistence , High performance master database for microservice applications , This eliminates the need to manage cache and persistent databases separately .