What should I do with the Redis cluster solution?

The high-performance computer cluster system is a small high-performance parallel computing system based on the network and oriented to scientific research, which can swim through a set of loosely integrated computer software and hardware to work closely together to complete the computing work. Connect a single computer node in the cluster system through the LAN to make it complete the same job at the same time to achieve high work efficiency, high computing speed and high reliable performance.

The system is designed in a hierarchical manner, deployed in accordance with Intel's high-performance computing ecosystem, top-down, and in accordance with the deployment of "HPC parallel applications, middleware cluster management and communication libraries, as well as various software optimization tools, operating systems, hardware platforms for computing nodes and master control nodes, and system environment", including standardized design for heat dissipation, power supply, and space layout.

Redis clusters have some functional limitations compared with single nodes, which should be avoided when using them. Note the following:

1) Limited support for key batch operations. For example, mset and mget can only have the same slot value.

key to perform batch operations. For the cherry key mapped to different slot values, it is possible to perform operations such as mget and mget.

can exist on multiple nodes and is therefore not supported.

2) Limited support for key transaction operations. In the same way, only multi-key transactions on the same node are supported.

Transactions cannot be used when multiple keys are distributed across different nodes.

3) key as the minimum granularity of data partitioning, so you can't put a large key-value object such as.

hash, list, etc. are mapped to different nodes.

4) Multi-database space is not supported. Redis on a single node can support 16 databases and cluster modules.

Only one database space, db0, can be used.

5) The replication structure only supports one layer, and the slave node can only copy the master node, and does not support nested tree complexes.

A minimum of 3 master nodes are required to build a Redis Cluster environment, and here you refer to the example on the official website to create 6 nodes, including 3 master nodes and 3 slave nodes, and the corresponding Redis node IP and port are as follows:

Here's a configuration file with the least options for clustering

Create a new directory and create six subdirectories named after the port number, and later we'll run a Redis instance in each directory: Here's the command:

In folders 7000 to 7005, create a file for each, and the contents of the file can be used in the example configuration file above, but remember to change the port number in the configuration to the same number as the same server name from 7000 to the same number as the folder name.

Start the cluster instance

The log printed by the instance shows that because the file does not exist, each node has a new ID assigned to itself:

Instances will always use the same ID, keeping a unique name in the cluster.

Go to the Redis directory and run the following command to create a cluster:

Just install Ruby.

Missing rubygems component, install with yum.

The prompt cannot be loaded because the Redis and Ruby interfaces are missing, and the GEM is used to install it.

Cluster commands are being executed.

At this point, the cluster mode is set up.

Use the redis-cli command to enter the cluster environment, and you need to bring it with you to enter the cluster mode-cIf port 7000 is not included, it indicates that the normal Redis port 7000 is entered.

Clusters

node

Slot

key