Copyright Notice

The content of this User Manual is copyrighted by Kingdee Apusic Cloud Computing Co., Ltd. Reproduction, excerpting, or use of the text or viewpoints in any other manner should be credited with the source as "Kingdee Apusic Cloud Computing Co., Ltd.".

Introduction

Apusic In-Memory Data Cache（AMDC）is a comprehensive, high-throughput, and data-secure distributed cache software designed to provide safe and reliable caching capabilities for large-scale, high-concurrency, and highly available critical applications. It is compatible with the Redis protocol and persistent data files, enabling straightforward and seamless replacement of Redis.

This document will provide users with a detailed introduction to using Kingdee Apusic Distributed Cache, divided into five major sections: Console, Cache Core, Shell Client, RDB Cluster Data Migration Tool, and Performance Testing Tool.

Deployment Assistance

For detailed information on installation and deployment, please refer to theInstallation Manual。

Console

The AMDC Console is a web-based management and monitoring tool supporting cache monitoring, auto-deployment, cluster/node management, scaling, ACL management, automatic alerts, real-time configuration, web shell, access control, and more.

Console Configuration File

General Configuration Items

The AMDC Console configuration file is stored at: /Console Installation Directory/amdc-console/config.yaml

After modifying the configuration file, you must restart the console service for the new settings to take effect. Commonly used configuration items are as follows:

First-Level Parameter	Second-Level Parameter	Default Value	Notes
system	addr	9001	Listening port for the console
sqlite	path	../console.db	Storage file for the console
zap	director	log	Log file for the console
tls	isEnable	false	Enable HTTPS access

Keycloak Single Sign-On Integration Configuration

The AMDC Console configuration file is stored at: /Console Installation Directory/amdc-console/config.yaml

First-Level Parameter	Second-Level Parameter	Default Value	Notes
keycloak	isEnable	false	Enable Keycloak single sign-on mode
	configURL	""	Authentication URL
	clientID	""	Client ID
	clientSecret	""	Client credential
	redirectURL	""	Redirect URL
	state	"somestate"	Custom state request parameter used for requesting authorization token (OAuth2 protocol standard)
	accestoken_publickey	""	Public key for parsing RS256 algorithm accesstoken

Quick Start

Role Explanation

The Console employs a three-role management approach:

The three roles are:

• System Administrator（Account：SystemAdministrator）
• Security Keeper（Account：KeysKeeper）
• Safety Auditor（Account：SafetyAuditor）

The initial password for all three roles is 【admin!123】，and it is recommended to change the password after logging in.

For actual users, there are two types of roles:

• Administrator：Responsible for implementing and maintaining AMDC services but not a direct user, and does not require allocation to a tenant.
• Tenant Account：A role that uses caching services, only responsible for using caching services, not involved in implementation and maintenance processes. Tenants are grouped to achieve data isolation. Multiple tenant accounts can exist under a single tenant.

Creating Accounts

For a new user, the three-role management can be a bit complex. Here's how to quickly create an account:

1. Log in as the System Administrator（SystemAdministrator）and create tenants and accounts (multiple accounts can be created, accounts do not necessarily belong to tenant accounts, they can also serve as administrator accounts);
2. Log in as the Security Keeper (KeysKeeper) and click on the "Unauthorized" group, select an account, and click the "Authorize" button.
3. If authorized as an Administrator, it defaults to the administrator group and cannot be changed;if authorized as a Tenant Account, then you need to choose which tenant to authorize (the relationship between tenant accounts and tenants is many-to-one);
4. Select the account and click to modify the password, adding a password for it (this can also be used to modify the password for old users);
5. Select the account and click to activate, enabling the account;
6. Log in with the new account.

Common Features

These are functions and pages accessible by all roles.

Login

After deployment, open a browser（recommended: Chrome、firefox） and enter the address to access the Console login page: http://serverIP:serverPort。

For Example：

Deploying the AMDC Console on the server 192.168.0.1, with the default port 9001, the Console login address would be: http://192.168.0.1:9001

Refer to Chapter 8, "Passwords and Security," for initial username and password.

Personal Information

Click on the avatar in the top right corner of the home page to enter the personal information page, where you can view user information or modify the current user's password, email, and phone number. The avatar is temporarily not modifiable.

Service Monitoring

Tenant Path: 【Service Management】>【Service List】

Administrator Path:【Tenant Management】>【Tenant Details】>【Monitoring】

Click the【Monitoring】button under the cluster section to enter the cluster monitoring page, where detailed information about the current cluster and key metric items will be displayed.

Analysis

Click the【Analysis】button above【Monitoring】 to enter the【Analysis】 page. Here, you can conduct data analysis and queries for the cache service's bigkeys, hotkeys, memkeys, monitor, and slowlog.

Administrator Features

Quick Start

• Import Instance： Tenant List->Tenant Details->Import

• Automatic Deployment： Upload License->Upload Installation Package->Add Machine (Server)->Edit Configuration->Automatic Deployment->Deployment Task

Tenant List

Page Path: Tenant Management - Tenant List

Displays information about all tenants.

Tenant Details

Page Path: Tenant Management - Tenant List - Tenant Details

Displays the list of caching services owned by this tenant. Provides the capability to automatically deploy services for the tenant, import existing clusters, enable/disable, configure, monitor, and delete caching services.

Import into an AMDC cluster or standalone machine

For AMDC caching services not managed from the console, you can import them into the console to enable monitoring, modification, command execution, and other operations. From the cluster home page, click the [+] Import button and follow the on-screen prompts to add information for the cluster or standalone instance to be monitored.

Parameter Name	Meaning
Cluster Name	Custom name for the cluster
AMDC Password	Password for the AMDC caching service
AMDC Node	IP:Port of the AMDC caching service node

Machine Management

Access【Deployment Management】>【Machine Management】, where the machine management page offers entry points for adding and deleting machines. The machine information required for automatic deployment will be created here.

Add Machine

Click home page【Deployment Management】>【Machine Management】o enter the machine management page.You need to have a machine group before you can add a machine; only machines with the same architecture chip can be added to the same machine group. Click the [+] button next to the machine group on the left to add a new machine group; Select the machine group, then click the [Add] button at the upper left, and input the corresponding information in the input fields to add a machine. Note: Machines can only be added if the connection is normal, so please verify the connection status of the target machine.

Parameter Name名	Meaning
Username	Username for logging into the remote machine, e.g., root
Password	Password for logging into the remote machine
SSH Port	Port for logging into the remote machine, e.g., 22
Host	Address of the remote machine, e.g., 192.168.0.213
Remarks	Can be left empty

Remove Machine

Click【Deployment - Machine Management】 to enter the machine management page. Select the machine(s) you wish to delete, then click the【Delete】button to the right of the machine list to remove the selected machine(s). Batch deletion is supported.

Edit Machine

Click【Deployment Management】>【Machine Management】 to enter the machine management page. Select the machine you wish to edit, then click the 【Edit】button to the right of the machine list to edit the information for the selected machine.

Configuration Templates

Click【Deployment Management】>【Configuration Templates】 to enter the configuration templates page. On this page, you can define configuration templates for the cache core and sentinels, which can be used during【Automatic Deployment】

Note: Non-universal parameters such as IP addresses and ports will not take effect in the configuration templates to ensure the proper operation of the automatic deployment process.

Files

File management is used to store AMDC installation packages and License files. Uploaded files will undergo preliminary validation by the control panel to determine their usability and can be selected for use in automatic deployment.

Package Management

Package management is used to upload installation packages for automatic deployment.

Uploading Installation Packages

Click the【Upload】button to open the installation package upload dialog.Click【Select File】to choose the installation package you wish to upload, then confirm. Multiple files can be uploaded simultaneously.

Deleting Installation Packages

Select the installation package(s) you wish to delete, then click the【Delete】button at the top left corner. Supports deleting multiple files at once.

License Management

Managing Licence for uploading for use in automated deployments. After uploading, Licenses will be parsed, and invalid Licenses will not be saved in the console.

Uploading Licenses

Click the 【Upload】button to open the License upload dialog. Click【Select File】 to choose the License file you wish to upload, then confirm. Multiple files can be uploaded simultaneously.

Removing Licenses

Select the license(s) you wish to delete, then click the 【Delete】button at the top left corner. Multiple licenses can be deleted simultaneously.

Automatic Deployment

Navigate to【Tenant List】>【Tenant Management】>【Tenant Details】, then click【Automatic Deployment】to enter the automatic deployment form page.

Fill in the relevant deployment information completely, then click 【Save Task】.This will save the deployment information to【Task Center】>【Deployment Tasks】; at this point, the deployment task has not started. Click【Execute Task】 to both save the deployment information to【Task Center】>【Deployment Tasks】 and begin the deployment process immediately. The progress and status of the deployment can be viewed in【Deployment Tasks】。

Prerequisites: Completion of the three actions: adding new machines, uploading licenses, and uploading installation packages.

Page Parameters:

Parameter Name	Meaning
Cluster Name	Custom name for the cluster
AMDC Password	Password for the AMDC caching service client connection, optional
Mode	Standalone mode, master-slave mode; when deploying in cluster mode, select standalone
Environment	Default Machine
Installation Package	Select the already added installation package (see Package Management for details)
License	Select the already added License (see License Management for details)
Machine List	Select the already added machines (see Machine Management for details)
Start Port	The port on which the AMDC cache service starts, multiple instances will be laid out on the current port + 1
Installation Directory	Installation path for the AMDC caching service; when installing with sudo, it must be installed under the home/ directory
sudo	Whether to install using sudo
Auto Topology	Auto-topology represents the automatic planning of the distribution of masters and slaves on the machine, and conversely requires manual specification will be arranged according to the current port + 1

Note: Using the automatic deployment function requires that the target server has tar and ss commands available.

Automatic Single-Machine Deployment via the Console

Before automatic deployment via the console, ensure that the link to the target server is normal, the target server has been added to the machine list (refer to Machine Management), and the target server has the tar and ss commands available.

Procedure:

1. Select the mode as【Single-Machine】.In single-machine mode, the primary node defaults to 1 and the replica node defaults to 0. Enter the corresponding information as prompted on the automatic deployment interface, then click the【Install】 button to perform a one-click deployment.

Verify the results of automatic deployment: Click on the【Service List】menu to check whether the cluster has been generated and its status is healthy. You can use the【Command Line】for further verification.

Automatic Master-Slave Deployment via the Console

Before automatic master-slave deployment via the console, ensure that the link to the target server is normal and that the target server has been added to the machine list (refer to Machine Management).

Procedure:

1. Select the AMDC core installation package and license according to the page content, choose the machine based on the target server,and in the【Mode】select Master-Slave.The primary node defaults to 1, and installation can be initiated to automatically achieve master-slave deployment.

Automatic Sentinel Deployment via the Console

Before automatic sentinel deployment via the console, ensure that the link to the target server is normal and that the target server has been added to the machine list (refer to Machine Management).

Procedure:

1. Select the AMDC core installation package and license according to the page content, choose the machine based on the target server, and in the【Mode】select Master-Slave. The primary node defaults to 1, and installation can be initiated to automatically achieve master-slave deployment.

Automatic Cluster Deployment via the Console

Before automatic cluster deployment via the console, ensure that the link to the target server is normal and that the target server has been added to the machine list (refer to Machine Management).

Procedure:

1. Select the AMDC core installation package and license according to the page content, choose the machine based on the target server, and in the【Mode】 select Cluster.Installation can be initiated to automatically achieve cluster deployment.

Deployment Tasks

Go to 【Task Center】-【Deployment Tasks】，where all task lists are displayed. You can view the deployment status through the field information in the task list; if a task is in a not executedstate, you can modify the deployment information through【Edit】;if the deployment fails, you can click the【Tasks Details】button in the list operation to view the specific reasons.

Settings

Provides the AMDC service command line, configuration modifications, stop, start, and restart of cluster nodes, viewing of info and config information, association with machines, data backup, and data recovery operations.

Command Line

Enter the【Service List】,click the【Command Line】button within the service ,or click【Settings】to enter the service settings interface,then click the【Command Line】button.

Import

To add nodes to the cluster, you can import AMDC cache services created through other channels via the console.

Procedure: Click the【Import】button to enter the cluster node import page and import AMDC services that have not been expanded through the console.

Note: Only services that are inherently part of the current cluster's nodes can be imported.

Cluster Expansion

Click the【Expand】button to navigate to the cluster expansion page, allowing you to expand the slave nodes in master-slave mode or extend the nodes in cluster mode.

Update License

Click the【Update License】 button to open the license update pop-up window. Select the license that has already been uploaded and click【Confirm】 to proceed with the update.When the license is due to expire within 1 month, it will be reminded in the【Service List】.

Associate Machines

Click the【Settings】button in the cluster information bar on the home page to enter the cluster settings page. Select the corresponding node and click the【Associate Machines】button to associate the imported nodes with machine information.This provides comprehensive management functionality for the nodes.

Start, Stop, Restart, Delete Nodes

Select the corresponding node and click the【Stop】、【Restart】、【Start】、【Delete】button to perform the operations of stopping, starting, restarting, or deleting nodes, which can be done in batches.

Tenant Features

Service Management

The【Service List】provides entry points for operations such as cluster browsing, cluster monitoring, cluster alerts, cluster settings, cluster editing, and deleting clusters.

Command Line

Click on the 【Service List】，select the【More-Command Line】button for a cluster toenter the cluster command line interface, where you can simulate interaction as a cluster client.

ACL Management

Click on the【Service List】, select the 【ACL】button for a cluster to enter the ACL management interface, where you can manage access control lists for the cluster.

Corresponding to acl setuser username >password on ~* &* +@all（+command） to explain.

Parameter Name	Meaning
Username (username)	The name of the new user
Enabled (on/off)	Whether to enable or disable the user
Password (>password)	The password for the user
Accessible Data (~*)	A regular expression matching accessible data
Command Set (+@all)	The name of the command set in the acl cat list.
Accessible Channels (&*)）	A regular expression matching accessible publish/subscribe channels.
Commands (+command)	Specific commands from the acl cat list.

All parametersdo not require prefix symbols (such as: >, ~, &, +@, -, +, etc.) to be added, as shown in the following image.

Settings

Cluster settings provide the ability to modify configurations of the AMDC service, add, delete, and update nodes, clear memory, remove the cluster, stop, start, restart, and delete cluster nodes, view cluster info and config details, associate machines, backup data, expand the cluster, and import operations.。

Clear Cluster Memory

Select the corresponding node and click【Clear】to clear the distributed data on the current node.

Start, Stop, Restart Nodes

Select the corresponding node and click the【Stop】、【Restart】、【Start】 button to perform the respective operations on the node.

Command Line

Enter the【Service List】and click the【Command Line】button under the service or click【Settings】to enter the service settings interface, then click the【Command Line】button.

Data Backup

Click the 【Data Backup】 button; after clicking, it will prompt download information, backing up all data from all nodes in the service to your local machine.

Restore Data

Click the 【Restore Data】 button, upload the tar.gz/.rdb file, and restore the data into the service.

Configuration

Click the Configuration】 button to open a dropdown list with two options: 1.Dynamic Update — real-time effective configuration;2.Static Update — updates all configurations, changes take effect after node restart.

Dynamic Update:

In the pop-up window, select the configuration item to update, enter the new configuration parameters, and click 【Confirm】.

Static Update:

Navigate to the 【Configuration Template】Page, make modifications, and click 【Confirm】when done.

Alerts

Alerts are an additional feature of cache monitoring, automatically issuing alerts when cache triggers alert rules, notifying recipients.

Alert Channel

An alert channel refers to the notification method used to notify recipients. The AMDC console provides an Email alert channel.

Add Alert Channel

Click Alerts - 【Alert Channel】on the homepage to enter the alert channel page, click the 【Add】 button to create a new alert channel.

Parameter Name	Meaning
Channel Name	Customizable
Channel Type	Email
SMTP Address	Mail Transfer Protocol server
SMTP Port	Port used by the mail server
Sender Email	Username for sending emails
Password	Password for the sender's email
Recipients	Users under the current tenant, multiple selections allowed, notifying multiple people simultaneously
Remarks	Remark information

Edit Alert Channel

Click 【Alerts - Alert Channel】 on the homepage to enter the alert channel page, select the alert channel that needs editing and click the 【Edit】 button on the right side of the alert channel list to edit the alert channel.

Delete Alert Channel

Click 【Alerts - Alert Channel】 on the homepage to enter the alert channel page, check the alert channel(s) to be deleted, click the 【Delete】 button on the right side of the alert channel list, and confirm deletion in the confirmation box.

Alert Rules

Set alert rules in the console, which will trigger alerts within the set time frame once the set conditions are met! Alerts will be sent according to the frequency set!

Create Alert Rule

Alert rules are divided into global alert rules that are effective for all clusters and local alert rules that apply to individual clusters. Local cluster alert rules need to be added within the cluster (refer to ##### Add Cluster Alert Rules). When the global switch is turned on, it applies to all clusters. Click 【Alerts - Alert Rules】 on the homepage to enter the alert rules page, click the Create button at the top right corner of the page to create an alert rule.

• Node Status Alert: Node alert switch, turn on to activate node alerts
• Rule Name: Custom alert rule name
• Rule: Set the condition for triggering the alert
• Comparator: Select the comparison operator
• Comparison Value: Input the numerical value for the alert comparison. An alert will be triggered upon meeting this value.
• Interval: The period (in minutes) for checking the rule: 5, 10, 15, 20, 30, 45, 60, 120 (min)
• Active: Switch to activate the alert rule
• Global: Default is No. When Global is selected, this rule will apply to all clusters under the current tenant.

Edit Alert Rule

Click 【Alerts - Alert Rules】 on the homepage to enter the alert rules page, select the relevant alert rule, click the 【Edit】 button on the right side of the list to edit the alert rule.

Delete Alert Rule

Click 【Alerts - Alert Rules】 on the homepage to enter the alert rules page, check the alert rule(s) to be deleted, click the 【Delete】 button on the right side of the list to delete the alert rule. Multiple alert rules can also be selected and the 【Delete】 button at the top left corner clicked to delete them in batch.

Service Alerts

Click 【Services】 on the homepage to enter the cluster management page, click the 【Alerts】 button under the cluster section to enter the cluster alerts page. This page displays alert information for the current cluster, including the alert records of the current cluster, alert rules for the current cluster, and alert channels for the current cluster.

Add Service Alert Channels

The steps to add an alert channel for a cluster are as follows:

1. Add a new alert channel in 【Alerts > Alert Channels】 (refer to creating an 【Alert Channel】).

2. Go to 【Settings】 > 【Alerts】, switch to the Alerts tab, click the 【Add】 button, and select the relevant alert channel to add.

Add Service Alert Rules

Alert rules are divided into global alert rules and local alert rules. Global alert rules apply to all services without needing to be added; alert rules added for a single cluster can only be local alert rules. To add a global alert rule (refer to adding an alert rule), follow these steps to add cluster alert rules:

1. In 【Alerts】 > 【Alert Rules】, add a new alert rule and turn off the global switch (refer to adding an alert rule).

2. Go to 【Service List】 > 【Alerts】, switch to the Alert Rules tab, click the Add button, and select the relevant alert rule to add.

Three-Role Management

There are three special roles: System Administrator, Security Confidentiality Officer, and Security Auditor. There is one and only one account per special role, generated when the system starts up, with passwords that cannot be changed (encrypted configuration in the configuration file).

• The System Administrator is responsible for setting up tenants and accounts, has the Tenants Management and User Management menus;
• The Security Confidentiality Officer is responsible for assigning tenants and roles (access permissions) to accounts, has the Authorization Management menu;
• The Security Auditor is responsible for reviewing the control panel operation logs, has the Operation Logs menu.

The actual users of the control panel functions fall into two categories:

1. Administrator Accounts: Administrators are responsible for creating and maintaining services for tenants;
2. Tenant Accounts: Corresponding to services assigned to tenants, they can use these services and have access to some service management functions. A tenant represents an independent environment, with data isolated between different tenants. Tenants can manage their own services on the control panel (by logging in with a tenant account), but they only have usage rights, not ownership (they cannot independently decide on the modification or deletion of cache services).

Tenant Management

Tenant management is used to create, edit, and delete the tenants associated with users. It is primarily used to distinguish between AMDC clusters that users can manage.

Create Tenant

Click the 【Add】 button in the top right corner to create a new tenant.

Edit Tenant

Select the tenant you need to edit and click on the "Tenant Name" column to edit the tenant.

Delete Tenant

Select the tenant you want to delete and click the Delete button to remove the tenant.

Authorization Management

Enter 【Users】 > 【Authorization Management】 to access the authorization management page.

Authorize Roles

In the tenant, unassigned users who have not been allocated to any tenant require tenant authorization first.

Change Password

New accounts do not have an initial password and cannot log in until the password has been changed once through the authorization management interface.

Activate or Freeze

Click 【Activate/Freeze】 to activate or freeze a user's account; after freezing, the account cannot be logged into.

Operation History

Operation history records the history of all user actions on the console.

Password and Security

Password Modification Guidelines: To ensure system security, the password length must be at least six characters long and include special characters. Passwords can be modified through the control platform or by editing the configuration file.

Initial Passwords in Three-Role Management

The three roles refer to: System Administrator (Account: SystemAdministrator), Security Confidentiality Officer (Account: KeysKeeper), and Safety Auditor (Account: SafetyAuditor). The initial password for each is 【admin!123】. Note that these three accounts cannot be deleted.

Changing the Current User's Password

Log in to the console, click on 【User Information】 in the top right corner of the homepage, then on the user information page click the 【Change Password】 button. Modify the login password for the current user in the pop-up window.

Password Modification by the Security Confidentiality Officer

The System Administrator can modify the information for "Administrators" and "Regular Users," while Administrators can modify the information for "Regular Users." After logging in, navigate to 【Users】 > 【Authorization Management】, click on the tenant where the user resides, select the user, then click the 【Change Password】 button to modify the user's password.

Cache Core

Distributed caching is the most essential capability of AMDC, serving as the core of the entire product, with all other features built upon data caching operations. AMDC stores data directly in memory and leverages multi-threaded read-write separation to achieve efficient storage, catering to various types of data storage, facilitating rapid development, and reducing type conversion. It supports multiple data eviction strategies to make rational use of memory space.

Cache Service Configuration File

The AMDC cache service configuration file is located at: /installation root directory/amdc/conf.yaml. Some configurations can be modified via the console. Below are the detailed configurations for the AMDC cache service:

Category	Parameter Name	Default Value	Notes
Network	Bind	0.0.0.0	IP address to listen on; 0.0.0.0 indicates accessibility via all local IPs. Multiple addresses can be bound. It is recommended to add local access IP and remote access IP, e.g., Bind: - "127.0.0.1" - "172.24.4.190"
	Port	6359	Port number
	MaxClients	10000	Maximum number of connections; the server refuses new connections beyond this limit. Setting it to 0 disables the maximum connection limit
	Timeout	0	The server actively closes the connection if idle for longer than the set time. When set to 0, the timeout mechanism is not enabled. Timeout unit is seconds
	TcpKeepAlive	300	Unit is seconds; when set to 0, tcp keepalive is not configured
	IOReadGoroutineNum	12	Number of read goroutines
	IOWriteGoroutineNum	15	Number of write goroutines
	IOGoroutineDoReads	“yes”	Whether parsing should occur in the goroutine? yes/on.
	ReadonlyProGoNum	6	Number of read-only goroutines
Security	RequirePass	""	Authentication password; in the presence of users.acl, the password in users.acl takes precedence
	ACLFile	"./users.acl"	Location of the ACL permission control file
	ACLPubsubDefault	"allchannels"	Default permissions for ACL channels; allchannels / resetchannels.
	ACLLogMaxLen	128	Maximum number of logs saved for ACL log
General	Databases	16	Number of databases
	LogLevel	"debug"	Log level for filtering output logs, including debug, info, warn, error, fatal
	LogPath	""	Directory for log file output; when an empty string is set, the log file is not written to disk. Example: LogFile: "/tmp/server.log"
	LicensePath	"./license.xml"	Path to the license file
	KbcLicensePath	"./license.lic"	Path to the kbclicense file
MemoryManagement	MaxMemory	"0"	Maximum memory limit; if maxmemory is 0, there is no restriction. If no unit follows the number, the default unit is bytes. Units are case-insensitive, e.g., "1gb", "1GB", "1000mb", "1000m", "1000000KB", "1000000kb", "1000000000B", "1000000000b"
	MaxMemoryPolicy	"noeviction"	Cache eviction policy; supported: noeviction: Prohibits eviction of data. Default is this setting. When memory usage reaches the threshold, all commands that cause memory requests will return errors. volatile-lru: Selects the least recently used data from the dataset with expiration times. volatile-ttl: Selects data nearing expiration from the dataset with set expiration times. volatile-random: Arbitrarily selects data from the dataset with set expiration times to evict. volatile-lfu: Selects infrequently used data from the dataset with set expiration times to evict. allkeys-lru: Selects the least recently used data from the dataset to evict. allkeys-random: Arbitrarily selects data from the dataset to evict. allkeys-lfu: Selects infrequently used data from the dataset to evict
	MaxMemorySamples	5	Sample count during each cache eviction
	LFULogFactor	10.000000	The lfu-log-factor adjusts the probability of counter growth; the larger the lfu-log-factor, the smaller the probability of counter growth. The calculation formula is: 1 / (old_value * lfu_log_factor + 1)
	LFUDecayTime	1.000000	LFU decay time is a value in minutes that adjusts the speed of counter reduction
	ReplSlaveIgnoreMaxmemory	"yes"	Whether slave nodes ignore maxmemory checks
Snapshotting	Save	- "3600 1" - "300 10" - "60 10000"	save , specifies how many update operations after a certain period the server will rewrite the RDB file to disk, e.g., Save: - "3600 1" - "300 10" - "60 10000" When Save: "" is empty, RDB auto-save is disabled
	StopWritesOnBgsaveError	"yes"	Whether the server stops accepting writes after bgsave save fails
	RdbCompression	"yes"	Whether to enable LZF compression for string objects, yes/no
	RdbCheckSum	"yes"	Whether to enable CRC64 checksum, yes/no
	DbFileName	"dump.rdb"	RDB filename, excluding path
	Dir	"./"	Working directory; RDB and AOF files will be stored under the Dir path
AppendOnlyMode	AppendOnly	"no"	Whether to enable AOF persistence, yes/no
	AppendFileName	"appendonly.aof"	AOF file name, excluding path
	AppendFSync	everysec	AOF file buffer flush strategy; options are always / everysec / no
	AutoAofRewritePercentage	100	Growth percentage of the current AOF file compared to the last rewrite
	AutoAofRewriteMinSize	"4M"	Minimum size to trigger AOF file rewriting
	AofNoFsyncOnRewrite	"yes"	Whether to reject fsync flushing when executing bgsave or bgrewiteaof, yes / no
	AofLoadTruncated	"yes"	# Behavior when errors occur during AOF file loading process: # yes: When errors occur during AOF file loading, the AOF file will be partially loaded and log messages will notify the user. # no: When errors occur during AOF file loading, the server will report an error and refuse to start
	AofUseRdbPreamble	"yes"	Use RDB format as the base file for AOF (smaller file)
SlowLog	SlowLogSlowerThan	10000	Specifies that commands that take more than a few microseconds to execute are logged to the logs
	SlowLogMaxLen	128	Specifies that commands that take more than a few microseconds to execute are logged to the logs
Script	LuaTimeLimit	5000	Maximum execution time for Lua scripts, measured in milliseconds. Set to 0 to impose no limit on maximum execution time
	LuaMaxLocalVarNum	600	Maximum number of parameters for Lua scripts
LazyFree	LazyEviction	"no"	Whether to adopt the lazy free mechanism when evicting keys, yes / no
	LazyExpire	"no"	For keys with TTL, whether to adopt the lazy free mechanism when cleaning up after expiration, yes / no
	LazyServerDel	"no"	For some commands that implicitly perform a DEL key operation when processing existing keys, such as the rename command, whether to adopt the lazy free mechanism, yes / no
	ReplicaLazyFlush	"no"	For full data synchronization of slaves, before loading the master's RDB file, the slave executes flushall to clear its own data; parameter settings determine whether asynchronous flush mechanisms are adopted, yes / no
Replication	Replicaof	""	Sets the server to be a slave node of a specified server, e.g., Replicaof: "127.0.0.1 6359"
	MasterAuth	""	Password for authentication between master and slave nodes
	ReplTimeout	60	Timeout for master-slave node connection, measured in seconds
	ReplServeStaleData	"yes"	Whether slave nodes continue to handle read requests during master-slave disconnection or sync phase
	MinReplicasToWrite	0	If the number of normal slave nodes is less than this configuration, the master node rejects command execution
	MinReplicasMaxLag	10	If slave nodes do not return ACK information beyond this configured time, they are judged abnormal; units are seconds
	ReplicaReadOnly	"yes"	Whether slave nodes only handle read requests and cannot modify data
	ReplicaAnnounceIp	""	If port forwarding or NAT is enabled, the value set by ReplicaAnnounceIp overrides the slave node's default IP value
	ReplicaAnnouncePort	0	If port forwarding or NAT is enabled, the value set by ReplicaAnnouncePort overrides the slave node's default Port value
	ReplDisableTcpNoDelay	"no"	Whether to disable TCP_NODELAY after SYNC
	ReplPingSlavePeriod	10	Time interval in seconds for the master node to send PING commands to slave nodes
	ReplBacklogSize	"1mb"	Size of the replication backlog buffer
	ReplBacklogTTL	3600	Time in seconds for the master node to remain in a slaveless state; beyond this configuration, the replication backlog area is released.
	ReplicaPriority	100	When the master node cannot work normally, Sentinel prioritizes promotion based on this value; the smaller the value, the higher the priority for promotion. A value of 0 indicates that the slave node can never be promoted to a master node
EventNotification	NotifyKeyspaceEvents	None	Supported types for key notifications
Cluster	ClusterEnabled	"no"	Whether to enable cluster mode, yes / no
	ClusterConfigFile	"./node.conf"	Name of the configuration file for each cluster node; names must be unique and automatically generated and updated by the node; manual editing is not allowed
	ClusterNodeTimeout	15000	Cluster node timeout, measured in ms
	ClusterReplicaValidityFactor	10	Slave nodes disconnected from the master (ReplPingSlavePeriod + ClusterReplicaValidityFactor * ClusterNodeTimeout) seconds do not participate in failover
	ClusterMigrationBarrier	1	Only when a master node has at least ClusterMigrationBarrier number of slave nodes in normal operation will slave nodes be allocated to isolated master nodes in the cluster
	ClusterRequireFullCoverage	"yes"	Whether all 16384 slots in the cluster need to be fully allocated; yes: If slots are not fully allocated, the entire cluster will be in an unavailable state until all slots are allocated, and the cluster will automatically become available; no: When slots are not fully allocated, parts of the cluster nodes are still available
	ClusterReplicaNoFailover	"no"	Whether this cluster node participates in automatic failover; yes: This cluster slave node will not participate in the automatic failover process but can be manually forced to execute failover; no: This cluster slave node participates in the automatic failover process.
	ClusterAnnounceIp	""	To make the cluster work in environments with port forwarding or NAT, statically configure ClusterAnnounceIp so that each node in the cluster knows its public IP address
	ClusterAnnouncePort	0	To make the cluster work in environments with port forwarding or NAT, statically configure ClusterAnnouncePort so that each node in the cluster knows its public port number
	ClusterAnnounceBusPort	0	To make the cluster work in environments with port forwarding or NAT, statically configure ClusterAnnounceBusPort so that each node in the cluster knows its public cluster message broadcast port
Advanced	ClientQueryBufferLimit	"1gb"	Maximum value of request input buffer; the server directly closes the connection when exceeding the maximum buffer value
	ProtoMaxBulkLen	"512mb"	Maximum length of each line of strings for RESP protocol Bulk multi-line requests
	GroutineMaxCpuNum	0	CPU usage limit; 0 imposes no usage limit
Proxy	Enabled	"no"	Whether to start the proxy
	Bind	"127.0.0.1"	Proxy's own IP address
	Port	8002	Port number used by the proxy
	Proxy2IP	"127.0.0.1"	IP address of the cache service for the proxy
	Proxy2Port	6359	Port number of the cache service for the proxy
	CriptEnabled	"yes"	Whether encryption is enabled
	GMflag	1	Whether national cryptography is enabled
	CertPath	"./certs/gm_cert"	Path to encryption and decryption authentication files
IpTable	Ip	""	Adds an IP address to the whitelist; e.g., Ip: - "127.0.0.1" "192.168.116.1"
	Segment	""	Adds a network segment to the whitelist; e.g., Segment: - "127.0.0.1/24"
Prometheus	Enabled	false	Whether or not to enable Prometheus to monitor metrics data
	ClusterReplicaNoFailover	"no"	Whether this cluster node participates in automatic failover; yes: This cluster slave node will not participate in the automatic failover process but can be manually forced to execute failover; no: This cluster slave node participates in the automatic failover process
	Bind	"127.0.0.1"	Address bound to Prometheus HTTP service; by default, 127.0.0.1 for local access; if exposed to the LAN or specific IP, fill in the LAN IP segment address accessible to all traffic, which can be set to: 0.0.0.0
	Port	8004	Access port for Prometheus HTTP metrics
	NameSapce	“amdc”	Prefix for each Prometheus metric, e.g., default is amdc: amdc_command_total
	MetricsPath	“/metrics”	HTTP access URL address for Prometheus metrics indicators, e.g., combined with the above bind + port: http://127.0.0.0:8004/metrics
	ConnectionTimeOut	15s	Timeout for client connections to amdc
	Export-client-list	true	Whether to display information about clients connected to amdc
	EnableHTTPS	false	Use HTTPS access; default is false for http; change to true for HTTPS use
	CertPath	"./certs/tls_cert"	HTTPS access requires providing SSL certificates; ensure that the certificate is named server.pem and server.key in the folder
	ClusterReplicaNoFailover	"no"	Whether this cluster node participates in automatic failover; yes: This cluster slave node will not participate in the automatic failover process but can be manually forced to execute failover; no: This cluster slave node participates in the automatic failover process
ApusicAcls	Enable	false	Whether to enable the apusic license authentication center; if enabled, use true; otherwise, use false
	AuthUrls	""	Authentication center address, must be in ip:port format; multiple addresses separated by commas
	Namespace	""	Namespace
	Tenant	""	Tenant name

Operation Commands

Command	Description
ACL LOAD	Reload ACLs from the configured ACL file.
ACL SAVE	Save current ACL rules to the configured ACL file.
ACL LIST	List current ACL rules in ACL configuration file format.
ACL USERS	List usernames of all configured ACL rules.
ACL GETUSER username	Retrieve rules for a specific ACL user.
ACL SETUSER username [rule [rule ...]]	Modify or create rules for a specific ACL user.
ACL DELUSER username [username ...]	Delete specified ACL users and associated rules.
ACL CAT [categoryname]	List ACL categories or commands within a category.
ACL GENPASS [bits]	Generate pseudo-random secure passwords for ACL users.
ACL WHOAMI	Return the name of the user associated with the current connection.
ACL LOG [count or RESET]	List recent events denied due to ACL.
ACL HELP	Display useful text about different subcommands.
APPEND key value	Append a value to a key.
ASKING	Sent by cluster clients after -ask redirection.
AUTH [username] password	Authenticate with the server.
BGREWRITEAOF	Asynchronously rewrite the append-only file.
BGSAVE [SCHEDULE]	Asynchronously save the dataset to disk.
BITCOUNT key [start end [BYTE|BIT]]	Count set bits in a string.
BITFIELD key [GET encoding offset] [SET encoding offset value] [INCRBY encoding offset increment] [OVERFLOW WRAP|SAT|FAIL]	Perform arbitrary bit field integer operations on strings.
BITFIELD_RO key GET encoding offset	Read-only variant of BITFIELD for strings.
BITOP operation destkey key [key ...]	Perform bitwise operations between strings.
BITPOS key bit [start [end [BYTE|BIT]]]	Find first set or cleared bit in a string.
BLPOP key [key ...] timeout	Remove and get the first element of a list, or block until one is available.
BRPOP key [key ...] timeout	Remove and get the last element of a list, or block until one is available.
BRPOPLPUSH source destination timeout	Pop an element from a list, push it to another list and return it; or block until one is available.
BLMOVE source destination LEFT|RIGHT LEFT|RIGHT timeout	Pop an element from a list, push it to another list and return it; or block until one is available.
LMPOP numkeys key [key ...] LEFT|RIGHT [COUNT count]	Pop elements from lists.
BLMPOP timeout numkeys key [key ...] LEFT|RIGHT [COUNT count]	Pop elements from lists, or block until one is available.
BZPOPMIN key [key ...] timeout	Remove and return lowest scoring members from one or more sorted sets, or block until one is available.
BZPOPMAX key [key ...] timeout	Remove and return highest scoring members from one or more sorted sets, or block until one is available.
BZMPOP timeout numkeys key [key ...] MIN|MAX [COUNT count]	Remove and return members with scores from sorted sets or block until one is available.
CLIENT CACHING YES|NO	Indicate whether the server should track keys for the next request.
CLIENT ID	Return the client ID of the current connection.
CLIENT INFO	Return information about the current client connection.
CLIENT KILL [ip:port] [ID client-id] [TYPE normal|master|slave|pubsub] [USER username] [ADDR ip:port] [LADDR ip:port] [SKIPME yes/no]	Kill a client's connection.
CLIENT LIST [TYPE normal|master|replica|pubsub] [ID client-id [client-id ...]]	Get a list of client connections.
CLIENT GETNAME	Get the name of the current connection.
CLIENT GETREDIR	Get the client ID of the notification redirection (if any).
CLIENT UNPAUSE	Resume processing of paused clients.
CLIENT PAUSE timeout [WRITE|ALL]	Stop processing commands for clients for a given time.
CLIENT REPLY ON|OFF|SKIP	Indicate whether the server should reply to commands.
CLIENT SETNAME connection-name	Set the name of the current connection.
CLIENT TRACKING ON|OFF [REDIRECT client-id] [PREFIX prefix [PREFIX prefix ...]] [BCAST] [OPTIN] [OPTOUT] [NOLOOP]	Enable or disable server-assisted client-side caching support.
CLIENT TRACKINGINFO	Return information about server-assisted client-side caching for the current connection.
CLIENT UNBLOCK client-id [TIMEOUT|ERROR]	Unblock a client blocked by a different connection.
CLIENT NO-EVICT ON|OFF	Set the client eviction mode for the current connection.
COMMAND	Get an array of detailed information about Redis commands.
COMMAND COUNT	Get the total number of Redis commands.
COMMAND GETKEYS	Extract keys given a full Redis command.
COMMAND INFO command-name [command-name ...]	Get an array of detailed information about specific Redis commands.
CONFIG GET parameter [parameter ...]	Get values of configuration parameters.
CONFIG REWRITE	Rewrite the configuration file from memory configuration.
CONFIG SET parameter value [parameter value ...]	Set configuration parameters to given values.
CONFIG RESETSTAT	Reset statistics returned by INFO.
COPY source destination [DB destination-db] [REPLACE]	Copy a key.
DBSIZE	Return the number of keys in the selected database.
DEBUG OBJECT key	Get debugging information about a key.
DEBUG SEGFAULT	Crash the server.
DECR key	Decrement the integer value of a key by one.
DECRBY key decrement	Decrement the integer value of a key by the given number.
DEL key [key ...]	Delete one or more keys.
DISCARD	Discard all commands issued after MULTI.
DUMP key	Return a serialized version of the value stored at the specified key.
ECHO message	Echo the given string.
EVAL script numkeys [key [key ...]] [arg [arg ...]]	Execute a Lua script on the server side.
EVAL_RO script numkeys key [key ...] arg [arg ...]	Execute read-only Lua scripts on the server side.
EVALSHA sha1 numkeys [key [key ...]] [arg [arg ...]]	Execute a Lua script on the server side.
EVALSHA_RO sha1 numkeys key [key ...] arg [arg ...]	Execute read-only Lua scripts on the server side.
EXEC	Execute commands queued by MULTI.
EXISTS key [key ...]	Determine if a key exists.
EXPIRE key seconds [NX|XX|GT|LT]	Set a key's time to live in seconds.
EXPIREAT key timestamp [NX|XX|GT|LT]	Set the expiry of a key as a UNIX timestamp.
EXPIRETIME key	Get the expiry Unix timestamp of a key.
FAILOVER [TO host port [FORCE]] [ABORT] [TIMEOUT milliseconds]	Initiate coordinated failover between this server and one of its replicas.
FLUSHALL [ASYNC|SYNC]	Remove all keys from all databases.
FLUSHDB [ASYNC|SYNC]	Remove all keys from the current database.
GEOADD key [NX|XX] [CH] longitude latitude member [longitude latitude member ...]	Add one or more geospatial items in a sorted set representing a geospatial index.
GEOHASH key member [member ...]	Return members of a geospatial index as standard geohash strings.
GEOPOS key member [member ...]	Return the longitude and latitude of members of a geospatial index.
GEODIST key member1 member2 [m|km|ft|mi]	Return the distance between two members of a geospatial index.
GEORADIUS key longitude latitude radius m|km|ft|mi [WITHCOORD] [WITHDIST] [WITHHASH] [COUNT count [ANY]] [ASC|DESC] [STORE key] [STOREDIST key]	Query a sorted set representing a geospatial index for members matching a given maximum distance from a point.
GEORADIUSBYMEMBER key member radius m|km|ft|mi [WITHCOORD] [WITHDIST] [WITHHASH] [COUNT count [ANY]] [ASC|DESC] [STORE key] [STOREDIST key]	Query a sorted set representing a geospatial index for members matching a given maximum distance from a member.
GEOSEARCH key [FROMMEMBER member] [FROMLONLAT longitude latitude] [BYRADIUS radius m|km|ft|mi] [BYBOX width height m|km|ft|mi] [ASC|DESC] [COUNT count [ANY]] [WITHCOORD] [WITHDIST] [WITHHASH]	Query a sorted set representing a geospatial index for members within a box or circle area.
GEOSEARCHSTORE destination source [FROMMEMBER member] [FROMLONLAT longitude latitude] [BYRADIUS radius m|km|ft|mi] [BYBOX width height m|km|ft|mi] [ASC|DESC] [COUNT count [ANY]] [STOREDIST]	Query a sorted set representing a geospatial index for members within a box or circle area and store results in another key.
GET key	Get the value of a key.
GETBIT key offset	Return the bit value at offset in the string value stored at key.
GETDEL key	Get the value of a key and delete the key.
GETRANGE key start end	Get a substring of the string stored at a key.
GETSET key value	Set the string value of a key and return its old value.
HDEL key field [field ...]	Delete one or more hash fields.
HELLO [protover [AUTH username password] [SETNAME clientname]]	Handshake with Redis.
HEXISTS key field	Check if a hash field exists.
HGET key field	Get the value of a hash field.
HGETALL key	Get all fields and values in a hash.
HINCRBY key field increment	Increment the integer value of a hash field by the given amount.
HINCRBYFLOAT key field increment	Increment the float value of a hash field by the given amount.
HKEYS key	Get all the fields in a hash.
HLEN key	Get the number of fields in a hash.
HMGET key field [field ...]	Get the values of all the given hash fields.
HMSET key field value [field value ...]	Set multiple hash fields to multiple values.
HSET key field value [field value ...]	Set the string value of a hash field.
HSETNX key field value	Set the value of a hash field, only if the field does not exist.
HRANDFIELD key [count [WITHVALUES]]	Get one or more random fields from a hash.
HSTRLEN key field	Get the length of a hash field value.
HVALS key	Get all the values in a hash.
INCR key	Increment the integer value of a key by one.
INCRBY key increment	Increment the integer value of a key by the given amount.
INCRBYFLOAT key increment	Increment the float value of a key by the given amount.
INFO [section]	Get information about the server.
LOLWUT [VERSION version]	Display some computer art and Redis version.
KEYS pattern	Find all keys matching a given pattern.
LASTSAVE	Get the UNIX timestamp of the last successful save to disk.
LINDEX key index	Get an element from a list by index.
LINSERT key BEFORE|AFTER pivot element	Insert an element before or after another element in a list.
LLEN key	Get the length of a list.
LPOP key [count]	Remove and get the first element of a list.
LPOS key element [RANK rank] [COUNT num-matches] [MAXLEN len]	Return indices of matching elements in a list.
LPUSH key element [element ...]	Push one or more elements onto the head of a list.
LPUSHX key element [element ...]	Push elements onto the head of a list, only if the list exists.
LRANGE key start stop	Get a range of elements from a list.
LREM key count element	Remove elements from a list.
LSET key index element	Set the value of an element in a list by index.
LTRIM key start stop	Trim a list to the specified range.
MEMORY DOCTOR	Output a report on memory issues.
MEMORY HELP	Display useful text about different subcommands.
MEMORY MALLOC-STATS	Show internal allocator statistics.
MEMORY PURGE	Ask the allocator to release memory.
MEMORY STATS	Show details about memory usage.
MEMORY USAGE key [SAMPLES count]	Estimate the memory usage of a key.
MGET key [key ...]	Get the values of all the given keys.
MIGRATE host port key| destination-db timeout [COPY] [REPLACE] [AUTH password] [AUTH2 username password] [KEYS key [key ...]]	Atomically transfer keys from one Redis instance to another.
MONITOR	Monitor all requests received by the server in real time.
MOVE key db	Move a key to another database.
MSET key value [key value ...]	Set multiple keys to multiple values.
MSETNX key value [key value ...]	Set multiple keys to multiple values, only if none of the keys exist.
MULTI	Mark the beginning of a transaction block.
OBJECT ENCODING key	Check the internal encoding of a Redis object.
OBJECT FREQ key	Get the logarithmic access frequency counter of a Redis object.
OBJECT IDLETIME key	Get the time since last access of a Redis object.
OBJECT REFCOUNT key	Get the reference count of a key's value.
OBJECT HELP	Display useful text about different subcommands.
PERSIST key	Remove the expiry from a key.
PEXPIRE key milliseconds [NX|XX|GT|LT]	Set a key's time to live in milliseconds.
PEXPIREAT key milliseconds-timestamp [NX|XX|GT|LT]	Set the expiry of a key as a timestamp in milliseconds.
PEXPIRETIME key	Get the expiry Unix timestamp of a key in milliseconds.
PFADD key [element [element ...]]	Add specified elements to the specified HyperLogLog.
PFCOUNT key [key ...]	Return the approximate cardinality of the set observed in key(s).
PFMERGE destkey sourcekey [sourcekey ...]	Merge N distinct HyperLogLogs into one.
PING [message]	Ping the server.
PSETEX key milliseconds value	Set the value and expiry in milliseconds of a key.
PSUBSCRIBE pattern [pattern ...]	Listen for messages published to channels matching the given patterns.
PUBSUB CHANNELS [pattern]	List active channels.
PUBSUB NUMPAT	Get the count of unique pattern subscriptions.
PUBSUB NUMSUB [channel [channel ...]]	Get the number of subscribers for channels.
PUBSUB HELP	Display useful text about different subcommands.
PTTL key	Get the remaining time to live of a key in milliseconds.
PUBLISH channel message	Publish a message to a channel.
PUNSUBSCRIBE [pattern [pattern ...]]	Stop listening for messages published to channels matching the given patterns.
QUIT	Close the connection.
RANDOMKEY	Return a random key from the key space.
READONLY	Enable read queries for connections to cluster replica nodes.
READWRITE	Disable read queries for connections to cluster replica nodes.
RENAME key newkey	Rename a key.
RENAMENX key newkey	Rename a key, only if the new key does not exist.
RESET	Reset the connection.
RESTORE key ttl serialized-value [REPLACE] [ABSTTL] [IDLETIME seconds] [FREQ frequency]	Create a key using a provided serialized value, previously obtained with DUMP.
ROLE	Return the role of the instance in the replication context.
RPOP key [count]	Remove and get the last element of a list.
RPOPLPUSH source destination	Remove the last element of a list, add it to another list and return it.
LMOVE source destination LEFT|RIGHT LEFT|RIGHT	Pop an element from a list, push it to another list and return it.
RPUSH key element [element ...]	Append one or more elements to the tail of a list.
RPUSHX key element [element ...]	Append elements to the tail of a list, only if the list exists.
SADD key member [member ...]	Add one or more members to a set.
SAVE	Synchronously save the dataset to disk.
SCARD key	Get the number of members in a set.
SCRIPT DEBUG YES|SYNC|NO	Set the debug mode for executed scripts.
SCRIPT EXISTS sha1 [sha1 ...]	Check if scripts exist in the script cache.
SCRIPT FLUSH [ASYNC|SYNC]	Remove all scripts from the script cache.
SCRIPT KILL	Terminate the currently executing script.
SCRIPT LOAD script	Load a specified Lua script into the script cache.
SDIFF key [key ...]	Subtract multiple sets.
SDIFFSTORE destination key [key ...]	Subtract multiple sets and store the resulting set in a key.
SELECT index	Change the selected database for the current connection.
SET key value [EX seconds|PX milliseconds|EXAT timestamp|PXAT milliseconds-timestamp|KEEPTTL] [NX|XX] [GET]	Set the string value of a key.
SETBIT key offset value	Set or clear the bit at offset in the string value stored at key.
SETEX key seconds value	Set the value and expiry of a key.
SETNX key value	Set the value of a key, only if the key does not exist.
SETRANGE key offset value	Overwrite part of the string at key starting from the specified offset.
SHUTDOWN [NOSAVE|SAVE]	Synchronously save the dataset to disk then shut down the server.
SINTER key [key ...]	Intersect multiple sets.
SINTERCARD numkeys key [key ...] [LIMIT limit]	Intersect multiple sets and return the cardinality of the result.
SINTERSTORE destination key [key ...]	Intersect multiple sets and store the resulting set in a key.
SISMEMBER key member	Determine if a given value is a member of a set.
REPLICAOF host port	Make the server a replica of another instance, or promote it to master.
SLOWLOG GET [count]	Get entries from the slow log.
SLOWLOG LEN	Get the length of the slow log.
SLOWLOG RESET	Clear all entries in the slow log.
SLOWLOG HELP	Displays useful text about different subcommands
SMEMBERS key	Returns all the members of the set stored at key
SMOVE source destination member	Moves a member from one set to another
SORT key [BY pattern] [LIMIT offset count] [GET pattern [GET pattern ...]] [ASC	DESC] [ALPHA] [STORE destination]
SORT_RO key [BY pattern] [LIMIT offset count] [GET pattern [GET pattern ...]] [ASC	DESC] [ALPHA]
SPOP key [count]	Removes and returns one or more random members from a set
SRANDMEMBER key [count]	Returns one or more random members from a set
SREM key member [member ...]	Removes one or more members from a set
LCS key1 key2 [LEN] [IDX] [MINMATCHLEN len] [WITHMATCHLEN]	Finds the longest common substring
STRLEN key	Returns the length of the value stored at key
SUBSCRIBE channel [channel ...]	Subscribes to messages published to given channels
SUNION key [key ...]	Adds multiple sets
SUNIONSTORE destination key [key ...]	Adds multiple sets and stores the resulting set in a key
SWAPDB index1 index2	Swaps two Redis databases
SYNC	Internal command used for replication
PSYNC replicationid offset	Internal command used for replication
TIME	Returns the current server time
TOUCH key [key ...]	Modifies the last access time of keys. Returns the number of existing keys specified
TTL key	Gets the time to live of a key in seconds
TYPE key	Determines the type stored at key
UNSUBSCRIBE [channel [channel ...]]	Stops listening for messages published to given channels
UNLINK key [key ...]	Deletes a key asynchronously in another thread. Otherwise behaves like DEL but non-blocking
UNWATCH	Forgets all watched keys
WAIT numreplicas timeout	Waits until the given number of replicas has synchronized all write commands sent in the current connection context
WATCH key [key ...]	Watches the given keys to determine the execution of a MULTI/EXEC block
ZADD key [NX	XX] [GT
ZCARD key	Returns the number of members in a sorted set
ZCOUNT key min max	Counts the number of members in a sorted set with scores within the given values
ZDIFF numkeys key [key ...] [WITHSCORES]	Subtracts multiple sorted sets
ZDIFFSTORE destination numkeys key [key ...]	Subtracts multiple sorted sets and stores the resulting sorted set in a new key
ZINCRBY key increment member	Increments the score of a member in a sorted set
ZINTERCARD numkeys key [key ...] [LIMIT limit]	Intersects multiple sorted sets and returns the cardinality of the result
ZINTERSTORE destination numkeys key [key ...] [WEIGHTS weight [weight ...]] [AGGREGATE SUM	MIN
ZLEXCOUNT key min max	Counts the number of members in a sorted set within the given lexicographical range
ZPOPMAX key [count]	Removes and returns the highest-scoring members from a sorted set
ZPOPMIN key [count]	Removes and returns the lowest-scoring members from a sorted set
ZMPOP numkeys key [key ...] MIN	MAX [COUNT count]
ZRANDMEMBER key [count [WITHSCORES]]	Returns one or more random elements from a sorted set
ZRANGESTORE dst src min max [BYSCORE	BYLEX] [REV] [LIMIT offset count]
ZRANGE key min max [BYSCORE	BYLEX] [REV] [LIMIT offset count] [WITHSCORES]
ZRANGEBYLEX key min max [LIMIT offset count]	Returns a range of members in a sorted set, by lexicographical range
ZREVRANGEBYLEX key max min [LIMIT offset count]	Returns a range of members in a sorted set, by lexicographical range from high to low string sorting
ZRANGEBYSCORE key min max [WITHSCORES] [LIMIT offset count]	Returns a range of members in a sorted set by score
ZRANK key member	Determines the index of a member in a sorted set
ZREM key member [member ...]	Removes one or more members from a sorted set
ZREMRANGEBYLEX key min max	Removes all members in a sorted set within the given lexicographical range
ZREMRANGEBYRANK key start stop	Removes all members in a sorted set within the given indexes
ZREMRANGEBYSCORE key min max	Removes all members in a sorted set within the given scores
ZREVRANGE key start stop [WITHSCORES]	Returns a range of members from a sorted set by index, ordered from high to low score
ZREVRANGEBYSCORE key max min [WITHSCORES] [LIMIT offset count]	Returns a range of members from a sorted set by score, ordered from high to low score
ZREVRANK key member	Determines the index of a member in a sorted set, ordered from high to low score
ZSCORE key member	Gets the score associated with the given member in a sorted set
ZMSCORE key member [member ...]	Gets the scores associated with the given members in a sorted set
ZUNIONSTORE destination numkeys key [key ...] [WEIGHTS weight [weight ...]] [AGGREGATE SUM	MIN
SCAN cursor [MATCH pattern] [COUNT count] [TYPE type]	Iterates the keyspace incrementally
SSCAN key cursor [MATCH pattern] [COUNT count]	Iterates Set elements incrementally
HSCAN key cursor [MATCH pattern] [COUNT count]	Iterates hash fields and associated values incrementally
ZSCAN key cursor [MATCH pattern] [COUNT count]	Iterates sorted set elements and related scores incrementally
XINFO CONSUMERS key groupname	Lists consumers in the consumer group
XINFO GROUPS key	Lists consumer groups of a stream
XINFO STREAM key [FULL [COUNT count]]	Retrieves information about a stream
XINFO HELP	Displays useful text about different subcommands
XADD key [NOMKSTREAM] [MAXLEN	MINID [=
XTRIM key MAXLEN	MINID [=
XDEL key ID [ID ...]	Deletes specified entries from a stream. Returns the actual number of items deleted which might differ from the number of IDs passed if some IDs do not exist.
XRANGE key start end [COUNT count]	Returns a range of elements from a stream whose IDs match the specified ID interval
XREVRANGE key end start [COUNT count]	Returns a range of elements from a stream in reverse order (from larger to smaller IDs) where IDs match the specified ID interval compared to XRANGE
XLEN key	Returns the number of entries in a stream
XREAD [COUNT count] [BLOCK milliseconds] STREAMS key [key ...] ID [ID ...]	Returns elements from multiple streams unseen since the caller reported IDs for each stream. Can be blocked.
XGROUP CREATE key groupname id|$ [MKSTREAM]	Creates a consumer group
XGROUP CREATECONSUMER key groupname consumername	Creates a consumer in a consumer group
XGROUP DELCONSUMER key groupname consumername	Deletes a consumer from a consumer group
XGROUP DESTROY key groupname	Destroys a consumer group
XGROUP SETID key groupname id|$	Sets the consumer group to an arbitrary last delivered ID value
XGROUP HELP	Displays useful text about different subcommands
XREADGROUP GROUP group consumer [COUNT count] [BLOCK milliseconds] [NOACK] STREAMS key [key ...] ID [ID ...]	Returns new entries from streams using a consumer group, or accesses the history of pending entries for a given consumer. Can be blocked.
XACK key group ID [ID ...]	Marks pending messages as properly handled, effectively removing them from the pending entries list of the consumer group. The return value of this command is the number of messages successfully acknowledged, i.e., the IDs we were actually able to parse in the PEL.
XCLAIM key group consumer min-idle-time ID [ID ...] [IDLE ms] [TIME ms-unix-time] [RETRYCOUNT count] [FORCE] [JUSTID]	Changes (or gets) ownership of messages in a consumer group as if the messages were delivered to the specified consumer
XAUTOCLAIM key group consumer min-idle-time start [COUNT count] [JUSTID]	Changes (or gets) ownership of messages in a consumer group as if the messages were passed to the specified consumer
XPENDING key group [[IDLE min-idle-time] start end count [consumer]]	Returns information and entries from the pending entries list of a stream's consumer group, i.e., messages that have been fetched but never acknowledged.

AMDC Cluster

The AMDC cluster is a highly available and scalable cluster, and the clustering modes are Master-Slave mode, Sentinel mode, and Cluster mode.

AMDC Master-Slave mode

AMDC Master-Slave mode, i.e. (Master-Slave Replication) Master-Slave Replication, uses one AMDC instance as the master and the rest as backup machines. The data of the master and the backup machine are identical, and the master supports various operations such as writing and reading of data, while the slave supports synchronisation and reading of data with the master. When one of the AMDCs fails, it is sufficient to access the AMDC cache service of the other node.

Master-Slave Command

• Transform into a slave of a node: Replicaof (Replicaof NO ONE will transform the slave into a master.)

AMDC Sentinel Mode

Sentinel is an AMDC cache server-side application that automatically monitors and handles the work of transferring failed nodes between AMDC cache services. AMDC provides commands to Sentinel, and Sentinel monitors multiple AMDC instances running by sending commands and waiting for a response from the AMDC cache service.

Sentinel Commands

1. Command to check the status of Sentinel：
   info
2. Command to obtain all monitored master nodes in Sentinel：
   sentinel masters
3. Command to get the status information of the main node named：
   sentinel master
4. Command to get the status information of all slaves under the master-name node：
   sentinel slaves
5. Command to get the IP address via the node name in Sentinel：s
   entinel get-master-addr-by-name
6. Command to add a node：
   sentinel monitor
7. Command to reset the status matched by the amdc name：
   sentinel reset
8. Command to delete a node：
   sentinel remove
9. Command to force a node to go subjectively down：
   sentinel failover

AMDC Cluster Mode

The cluster mode is for the elastic scalability of the main node implemented by AMDC. Its main function is to increase or decrease the memory capacity that AMDC can use, and to meet the caching requirements of the business by expanding nodes instead of increasing the server memory. Moreover, the cluster mode has an automatic failover function similar to Sentinel, with high availability, and is a better choice. All nodes in the cluster mode can communicate with each other and perceive each other's status information. The cluster can automatically allocate master and slave nodes or manually specify these nodes.

Cluster Commands

1. Allocate new hash slots to the receiving node:
   CLUSTER ADDSLOTS slot [slot ...]
2. Allocate new hash slots to the receiving node:
   CLUSTER ADDSLOTSRANGE start-slot end-slot [start-slot end-slot ...]
3. The command triggers an increment in the cluster configuration age from the connected node. If the configuration age of the node is zero or less than the maximum age of the cluster, the age will be incremented:
   CLUSTER BUMPEPOCH
4. Return the number of active failure reports for the given node:
   CLUSTER COUNT-FAILURE-REPORTS node-id
5. Return the number of local keys in the specified hash slot:
   CLUSTER COUNTKEYSINSLOT slot
6. Set the hash slot as unbound in the receiving node:
   CLUSTER DELSLOTS slot [slot ...]
7. Set the hash slot as unbound in the receiving node:
   CLUSTER DELSLOTSRANGE start-slot end-slot [start-slot end-slot ...]
8. Force the replica to perform a manual failover of its master:
   CLUSTER FAILOVER [FORCE|TAKEOVER]
9. Delete the slot information of the node itself:
   CLUSTER FLUSHSLOTS
10. Delete the node from the node table:
    CLUSTER FORGET node-id
11. Return the local key names in the specified hash slot:
    CLUSTER GETKEYSINSLOT slot count
12. Provide information on the status of AMDC Cluster nodes:
    CLUSTER INFO
13. Return the hash slot of the specified key:
    CLUSTER KEYSLOT key
14. Force one node cluster to handshake with another node:
    CLUSTER MEET ip port
15. Return the node id:
    CLUSTER MYID
16. Obtain the cluster configuration of the node:
    CLUSTER NODES

Prometheus API

AMDC has a built-in Prometh API. After enabling the Prometheus API, multiple standard monitoring metric contents can be output, allowing third-party monitoring devices to access and monitor.

Prometheus Configuration Items

Parameter Name	Resolution	Usage
Enabled	false	Whether to enable Prometheus monitoring metric data
Bind	"127.0.0.1"	The address bound by the Prometheus HTTP service. The default is 127.0.0.1 for local access. If it needs to be exposed to the local area network or a specified IP, please fill in the local area network IP segment of the current environment. All traffic can be accessed and can be set to: 0.0.0.0
Port	8004	The access port of the Prometheus HTTP metrics
NameSapce	"amdc"	The prefix of each metric of the Prometheus metrics. For example, the default is amdc: amdc_command_total. If it is set to redis: redis_command_total
MetricsPath	"/metrics"	The HTTP access URL address of the Prometheus metrics. For example, combined with the above bind + port: http://127.0.0.0:8004/metrics
ConnectionTimeOut	15s	The timeout period for the client to connect to amdc.
Export-client-list	true	Whether to display the client information connected to amdc.
EnableHTTPS	false	Use HTTPS access. The default is false for using http. If HTTPS needs to be used, it needs to be changed to true
CertPath	"./certs/tls_cert"	SSL certificates need to be provided for HTTPS access capabilities. Ensure that the names of the certificates in the folder are: server.pem and server.key

Prometheus Usage

The usage of Prometheus is very simple. The usage steps are as follows:

1. Open the conf.yaml configuration file and change Enabled to true;
2. Change the required configuration items. Important configuration items include Bing/Port;
3. Start/restart the AMDC cache core;
4. Use the monitoring system to access the Prometheus API of AMDC.

Prometheus Metric Items

Metric	English Description	Type (Type)	Chinese Description
amdc_aof_current_rewrite_duration_sec	aof current rewrite duration sec	gauge	Current aof rewrite duration
amdc_aof_last_bgrewrite_status	aof last bgrewrite status	gauge	The status of the last background rewrite of aof
amdc_aof_last_rewrite_duration_sec	aof last rewrite duration sec	gauge	The duration of the last write of aof
amdc_aof_last_write_status	aof last write status	gauge	The last write status of aof
amdc_aof_rewrite_in_progress	aof rewrite in progress	gauge	Aof is being rewritten
amdc_aof_rewrite_scheduled	aof rewrite scheduled	gauge	Aof rewrite is scheduled
amdc_blocked_clients	blocked clients	gauge	Blocked clients
amdc_cluster_enabled	cluster enabled	gauge	Whether it is in cluster mode
amdc_cluster_current_epoch	cluster current epoch	gauge	The value of the local Current Epoch variable of the cluster. This value is a unique auto-incremented version number created during the node failover period
amdc_cluster_healthy_status	cluster healthy status	gauge	Whether the cluster is healthy
amdc_cluster_known_nodes	cluster know nodes	gauge	The total number of known nodes in the cluster, including nodes that are in the handshake (HANDSHAKE) state and have not yet become formal members of the cluster
amdc_cluster_size	cluster size	gauge	The number of master nodes that contain at least one hash slot and can provide services
amdc_cluster_slots_assigned	cluster slots assigned	gauge	The number of slots that have been assigned
amdc_cluster_slots_fail	cluster slots fail	gauge	The number of hash slots with the status of FAIL. If this number is not zero, the node cannot provide queries unless cluster-require-full-coverage is set to no in the configuration
amdc_cluster_slots_ok	cluster slots ok	gauge	The number of hash slots whose status is not Fail and PFail
amdc_cluster_slots_pfail	cluster slots pfail	gauge	The number of hash slots with the status of PFAIL. PFAIL only means that we cannot currently communicate with the node, but it may just be a temporary error
amdc_cluster_my_epoch	cluster my epoch	gauge	The current configuration version assigned to this node
amdc_cluster_messages_received_total	cluster messages received total	gauge	The total number of messages received through the node-to-node binary bus
amdc_cluster_messages_sent_total	cluster messages sent total	gauge	The total number of messages sent through the node-to-node binary bus
amdc_cluster_stats_messages_meet_received	cluster stats messages meet received	gauge	The number of meet messages received through the node-to-node binary bus
amdc_cluster_stats_messages_ping_received	cluster stats messages ping received	gauge	The number of ping messages received through the node-to-node binary bus
amdc_cluster_stats_messages_ping_sent	cluster stats messages ping sent	gauge	The number of ping messages sent through the node-to-node binary bus
amdc_cluster_stats_messages_pong_received	cluster stats messages pong received	gauge	The number of pong messages received through the node-to-node binary bus
amdc_cluster_stats_messages_pong_sent	cluster stats messages pong sent	gauge	The number of pong messages sent through the node-to-node binary bus
amdc_commands_processed_total	commands processed total	counter	The total number of processed commands
amdc_config_maxclients	config maxclients	gauge	The set maximum number of clients
amdc_config_maxmemory	config maxmemory	gauge	The set maximum memory capacity
amdc_connected_clients	connected clients	gauge	The number of connected clients
amdc_connected_slaves	connected slaves	gauge	The number of connected slave nodes
amdc_connections_received_total	connections received total	counter	The total amount of data received by the connection
amdc_cpu_sys_children_seconds_total	cpu sys children seconds total	counter	The execution time of the child process in the kernel state
amdc_cpu_sys_seconds_total	cpu sys seconds total	counter	The execution time in the kernel state
amdc_cpu_user_children_seconds_total	cpu user children seconds total	counter	The execution time of the child process in the user state
amdc_cpu_user_seconds_total	cpu user seconds total	counter	The execution time in the user mode
amdc_db_keys	Total number of keys by DB	gauge	The number of keys in each db
amdc_db_keys_expiring	Total number of expiring keys by DB	gauge	The number of expiring keys in the DB
amdc_evicted_keys_total	evicted keys total	counter	The total number of evicted keys
amdc_expired_keys_total	expired keys total	counter	The total number of expired keys
amdc_expired_stale_percentage	expired stale percentage	gauge	The proportion of expired and stale
amdc_expired_time_cap_reached_total	expired time cap reached total	gauge	The total number of reaching the timeout
amdc_exporter_last_scrape_connect_time_seconds	exporter last scrape connect time seconds	gauge	The duration of the last connection establishment of the Prometheus service interface
amdc_exporter_last_scrape_duration_seconds	exporter last scrape duration seconds	gauge	The duration of the last processing request of the Prometheus service interface
amdc_exporter_last_scrape_error	The last scrape error status.	gauge	The last error of the Prometheus service interface
amdc_instance_info	Information about the Redis instance	gauge	Version and other information of AMDC
amdc_keyspace_hits_total	keyspace hits total	counter	The total number of hit keys
amdc_keyspace_misses_total	keyspace misses total	counter	The total number of missed keys
amdc_last_key_groups_scrape_duration_milliseconds	Duration of the last key group metrics scrape in milliseconds	gauge	The duration of obtaining key data information and processing
amdc_last_slow_execution_duration_seconds	The amount of time needed for last slow execution, in seconds	gauge	The required time for the last slow query log execution
amdc_lazyfree_pending_objects	lazyfree pending objects	gauge	The number of lazy deletion objects
amdc_master_repl_offset	master repl offset	gauge	The offset of the master node interaction processing
amdc_memory_max_bytes	memory max bytes	gauge	The maximum memory (bytes)
amdc_memory_used_bytes	memory used bytes	gauge	The used memory (bytes)
amdc_memory_used_peak_bytes	memory used peak bytes	gauge	The maximum used memory
amdc_memory_used_rss_bytes	memory used rss bytes	gauge	The total memory size occupied by the amdc process's exclusive memory and shared libraries
amdc_process_id	process id	gauge	The AMDC process ID
amdc_pubsub_channels	pubsub channels	gauge	The total number of publish-subscribe channels
amdc_pubsub_patterns	pubsub patterns	gauge	The total number of publish-subscribe patterns
amdc_rdb_bgsave_in_progress	rdb bgsave in progress	gauge	Whether RDB is in progress
amdc_rdb_changes_since_last_save	rdb changes since last save	gauge	How much dirty data there is since the last RDB
amdc_rdb_current_bgsave_duration_sec	rdb current bgsave duration sec	gauge	The current RDB duration
amdc_rdb_last_bgsave_duration_sec	rdb last bgsave duration sec	gauge	The duration of the last RDB
amdc_rdb_last_bgsave_status	rdb last bgsave status	gauge	The status of the last RDB
amdc_rdb_last_save_timestamp_seconds	rdb last save timestamp seconds	gauge	The timestamp of the last RDB
amdc_rejected_connections_total	rejected connections total	counter	The total number of rejected connections
amdc_repl_backlog_first_byte_offset	repl backlog first byte offset	gauge	The offset size of the accumulated interactive commands (bytes)
amdc_repl_backlog_history_bytes	repl backlog history bytes	gauge	The size of the interactive command history (bytes)
amdc_repl_backlog_is_active	repl backlog is active	gauge	Whether there is accumulation of interactive commands
amdc_replica_partial_resync_accepted	replica partial resync accepted	gauge	The number of accepted PSYNC commands
amdc_replica_partial_resync_denied	replica partial resync denied	gauge	The number of rejected PSYNC commands
amdc_replica_resyncs_full	replica resyncs full	gauge	The number of full synchronizations to slave nodes
amdc_replication_backlog_bytes	replication backlog bytes	gauge	The size of the commands accumulated for RDB synchronization
amdc_second_repl_offset	second repl offset	gauge	The offset time during the RDB synchronization process
amdc_slave_expires_tracked_keys	slave expires tracked keys	gauge	The total number of expirations on the slave node
amdc_slowlog_last_id	Last id of slowlog	gauge	The ID of the last slow query log
amdc_slowlog_length	Total slowlog	gauge	The total number of slow query logs
amdc_start_time_seconds	Start time of the Redis instance since unix epoch in seconds.	gauge	The start time (seconds)
amdc_up	Information about the Redis instance	gauge	Whether AMDC is running
amdc_uptime_in_seconds	uptime in seconds	gauge	The time since startup (seconds)
go_gc_duration_seconds	A summary of the pause duration of garbage collection cycles.	summary	The duration of go garbage collection
go_goroutines	Number of goroutines that currently exist.	gauge	The number of go coroutines
go_info	Information about the Go environment.	gauge	Go version information
go_memstats_alloc_bytes	Number of bytes allocated and still in use.	gauge	The size of the memory allocated and in use in the go program
go_memstats_alloc_bytes_total	Total number of bytes allocated, even if freed.	counter	The total memory allocated in the go program, including the size of the memory that has been freed
go_memstats_buck_hash_sys_bytes	Number of bytes used by the profiling bucket hash table.	gauge	The size of the memory occupied by the profiling bucket hash table in go
go_memstats_frees_total	Total number of frees.	counter	The size of the free memory
go_memstats_gc_sys_bytes	Number of bytes used for garbage collection system metadata.	gauge	The size of the memory used for garbage collection metadata
go_memstats_heap_alloc_bytes	Number of heap bytes allocated and still in use.	gauge	The size of the heap memory allocated and in use
go_memstats_heap_idle_bytes	Number of heap bytes waiting to be used.	gauge	The size of the heap memory waiting to be used
go_memstats_heap_inuse_bytes	Number of heap bytes that are in use.	gauge	The size of the heap memory in use
go_memstats_heap_objects	Number of allocated objects.	gauge	The number of allocated objects in the heap
go_memstats_heap_released_bytes	Number of heap bytes released to OS.	gauge	The size of the heap memory released to the operating system
go_memstats_heap_sys_bytes	Number of heap bytes obtained from system.	gauge	The size of the heap memory obtained from the system
go_memstats_last_gc_time_seconds	Number of seconds since 1970 of last garbage collection.	gauge	The time of the last garbage collection
go_memstats_lookups_total	Total number of pointer lookups.	counter	The total number of pointer lookups
go_memstats_mallocs_total	Total number of mallocs.	counter	The number of mallocs
go_memstats_mcache_inuse_bytes	Number of bytes in use by mcache structures.	gauge	The size of the memory in use by mcache structures
go_memstats_mcache_sys_bytes	Number of bytes used for mcache structures obtained from system.	gauge	The size of the memory used for mcache structures obtained from the system
go_memstats_mspan_inuse_bytes	Number of bytes in use by mspan structures.	gauge	The size of the memory in use by mspan structures
go_memstats_mspan_sys_bytes	Number of bytes used for mspan structures obtained from system.	gauge	The size of the memory used for mspan structures obtained from the system
go_memstats_next_gc_bytes	Number of heap bytes when next garbage collection will take place.	gauge	The size of the heap memory when the next garbage collection occurs
go_memstats_other_sys_bytes	Number of bytes used for other system allocations.	gauge	The number of bytes used for other system allocations
go_memstats_stack_inuse_bytes	Number of bytes in use by the stack allocator.	gauge	The size of the memory in use by the stack allocator
go_memstats_stack_sys_bytes	Number of bytes obtained from system for stack allocator.	gauge	The size of the memory obtained from the system for the stack allocator
go_memstats_sys_bytes	Number of bytes obtained from system.	gauge	The size of the memory obtained from the system
go_threads	Number of OS threads created.	gauge	The number of OS threads created by the process
process_cpu_seconds_total	Total user and system CPU time spent in seconds.	counter	The total CPU time spent by the process in user and system mode in seconds
process_max_fds	Maximum number of open file descriptors.	gauge	The maximum number of open file descriptors of the process
process_open_fds	Number of open file descriptors.	gauge	The number of open file descriptors of the process
process_resident_memory_bytes	Resident memory size in bytes.	gauge	The resident memory size of the process in bytes
process_start_time_seconds	Start time of the process since unix epoch in seconds.	gauge	The start time of the process since the Unix epoch in seconds
process_virtual_memory_bytes	Virtual memory size in bytes.	gauge	The virtual memory size of the process in bytes
process_virtual_memory_max_bytes	Maximum amount of virtual memory available in bytes.	gauge	The maximum virtual memory size available for the process in bytes

SSL Usage

SSL is the built-in encryption communication protocol of AMDC. Activating SSL enables bidirectional authentication encrypted communication with clients, which can be achieved using certificates and key files generated by OPENSSL.

SSL Configuration Items

Parameter Name	Description	Usage
Enable	Whether to enable SSL. Use true if enabled, otherwise false.	Enable: true
Port	SSL listening port. If SSL listening is enabled exclusively, set the Port under NetWork to 0.	Port: 6369
TlsCertFile	Server SSL certificate.	TlsCertFile: "./certs/ssl_tls_cert/server.crt"
TlsKeyFile	Key for the server SSL certificate.	TlsKeyFile: "./certs/ssl_tls_cert/server.key"
TlsCaCertFile	Certificate authority's certificate file, i.e., the trusted root certificate.	TlsCaCertFile: "./certs/ssl_tls_cert/ca.crt"
TlsCaCertDir	Path to the certificate authority's certificate files. Use this parameter if there are multiple trusted root certificates.	TlsCaCertDir: ""
TlsClientCertFile	Client SSL certificate, used for cluster/master-slave mode.	TlsAuthClients: "./certs/ssl_tls_cert/client.crt"
TlsClientKeyFile	Key for the client SSL certificate, used for cluster/master-slave mode.	TlsAuthClients: "./certs/ssl_tls_cert/client.key"
TlsAuthClients	Whether the server verifies client certificates. By default, requires the client to provide a certificate and the server performs validation. Can be configured as "no" if client certificate is not required. Setting to "optional" allows the client to optionally provide a certificate for verification.	TlsAuthClients: ""
TlsReplication	Whether TLS communication is used in master-slave mode. When the master node has this parameter set to true, the slave nodes must also be set to true.	TlsReplication: false

Generating Certificates with OPENSSL

You can obtain the required certificates using other methods, as long as they are legally valid and usable.

For SSL bidirectional authentication:

• The server requires a CA certificate, a server certificate, and a server private key.
• The client requires a CA certificate, a client certificate, and a client private key.

Steps:

1. Download and install OpenSSL from the official website: source/index.html (opens new window)

2. Create an openssl.cnf file:

   [ server_cert ]
   keyUsage = digitalSignature, keyEncipherment
   nsCertType = server
   
   [ client_cert ]
   keyUsage = digitalSignature, keyEncipherment
   nsCertType = client
   

3. Create the CA certificate:

   • a. openssl genrsa -out ca.key 4096
   • b. openssl req -x509 -new -nodes -sha256 -key ca.key -days 3650 -subj "/O=APUSIC/CN=AMDC.com" -out ca.crt

4. Create the server private key and certificate:

   • a. openssl genrsa -out server.key 2048
   • b. openssl req -new -sha256 -subj "/O=APUISC/CN=AMD.com" -key server.key | openssl x509 -req -CA ca.crt -CAkey ca.key -CAcreateserial -days 365 -extfile openssl.cnf -extensions server_cert -out server.crt

5. Create the client private key and certificate:

   • a. openssl genrsa -out client.key 2048
   • b. openssl req -new -sha256 -subj "/O=APUISC/CN=AMDC.com" -key client.key | openssl x509 -req -CA ca.crt -CAkey ca.key -CAcreateserial -days 365 -extfile openssl.cnf -extensions server_cert -out client.crt

SSL Client Connection

• Connecting using amdc-cli: ./amdc-cli -p 6369 --tls --cert ./client.crt --key ./client.key --cacert ./ca.crt
• Connecting using Go language:

Data Persistence

AMDC offers two persistence methods, RDB and AOF, for users. These methods are also operable through commands listed in the Command List. This section will detail the usage of both persistence methods along with associated tools.

RDB

The RDB persistence method stores data from AMDC in RDB format on the hard drive.

There are three ways to generate an RDB file:

1. Configure the Save directive in the configuration file to set conditions for RDB generation. When these conditions are met, the system automatically generates an RDB file using the bgsave command.
2. Use the save command. This command blocks all requests until the RDB file is generated; then, request processing resumes.
3. Use the bgsave command. This command does not block requests but generates the RDB file at a slower speed compared to the save command.

There are two ways to restore data from an RDB file:

1. Utilize the RDB Data Migration Tool.
2. Place one or more RDB files in the directory specified in the cache's configuration file under the Dir directive. Upon starting AMDC, it will automatically load the RDB data.

AOF

The AOF persistence method writes data from AMDC in AOF format into a text file, stored in the appendonlydir folder, and AMDC appends new requests to the text file in an append-only manner.

To generate an AOF file:

Set the AppendOnly directive in the configuration file to "yes".

Restoring AOF data

Place one or more AOF files or the entire appendonlydir folder in the directory specified in the cache's configuration file under the Dir directive. Upon starting AMDC, it will automatically load the AOF data.

Recovering a corrupted AOF file

AMDC provides an AOF file repair tool. However, this tool is only applicable if the last appended file is damaged. This design choice is made because if the base file is already corrupt, it implies that most of the data is lost and there is no point in attempting a recovery.

Usage: ./amdc-aof-check --fix [filename.aof|filename.manifest]

Command Audit

AMDC offers the monitor command, which can surveil all request information. By combining this with a shell client, you can direct the monitoring content to a specific file, thereby implementing command auditing.

The format of recorded request information: timestamp [database number client address: client port number] command

Usage: amdc-cli -h [ip] -p [port] [-a password] monitor > amdc_commands.log

Shell Client

For ease of use with AMDC, the shell client provides users with a convenient and swift command-line interface, offering practical features such as cluster establishment, cluster slot management, LRU testing, and more.

AMDC Client Arguments

Usage: amdc-cli [arguments] [command [command arguments ...]]

Argument	Description
-h	IP address of the cache service (default: 127.0.0.1)
-p	Port of the cache service (default: 6359)
-a	Password for requirepass authentication of the cache service; you can set and input the password using the AMDCCLI_AUTH environment variable, which is safer
--user	Username when logging in with an ACL user
--pass	Password corresponding to the ACL user during login
--askpass	Disregard the AMDCCLI_AUTH environment variable and force the use of the directly entered password
-u	URI address of the cache service (compatible with Redis protocol)
-r	Execute a specific command times
-i	When using -r, set the interval in seconds between each repeat execution
-n	Database number (there are 16 databases by default, numbered 0-15)
-3	Switch to RESP3 protocol
-x	Treat input read from stdin as the last argument for amdc-cli Example: amdc-cli -x set key < /opt/file
-d	Delimiter between response blocks in raw format (such as: \n)
-D	Delimiter between multiple responses in raw format (such as: \n)
-c	Connect in cluster mode
-e	Return error codes when a command execution fails
--raw	Use raw format output (when tty is not the default output device)
--no-raw	Force formatted output, even if standard output is not a tty
--quoted-input	Force the treatment of input as a quoted string
--csv	Output in CSV format
--show-pushes	Whether to print resp3 push messages, enabled by default
--stat	Dynamically print status information about the cache service: memory/client connections, etc.
--lru-test	Simulate a cache workload with an 80-20 distribution (key usage frequency)
--replica	Simulate a replica node displaying commands received from the master node
--slave host:ip	Specify a node as the master node for the current connection node
--rdb	Obtain an RDB file from the cache service as a local file
--pipe	Transmit raw amdc protocol (compatible with Redis) from stdin to the cache service
--pipe-timeout	Timeout time in pipe mode
--bigkeys	Find keys with large value memory consumption
--memkeys	Find keys with large key-value memory consumption
--memkeys-samples	Find keys with large key-value memory consumption, with more concise output
--hotkeys	Find keys with high usage counts, requires cache policy to be LFU (Least Frequently Used)
--scan	List all keys using the SCAN command
--pattern	Use regex pattern matching for keys when using --scan/--bigkeys/--hotkeys (default: *)
--quoted-pattern	Similar to --pattern, but quoted strings must be enclosed in quotes to avoid being interpreted as non-binary-safe strings
--intrinsic-latency	Test system intrinsic latency for a duration of seconds
--eval	Send and execute a LUA script file
--ldb	Use with --eval to enable the AMDC Lua Debugger
--ldb-sync-mode	Similar to --ldb, but synchronizes with the debugger, blocking the cache service
--cluster [args…] [opts…]	Cluster management commands and arguments; detailed commands and parameters can be viewed with
--cluster help	View help information
--verbose	Verbose mode
--no-auth-warning	Suppress warning information when directly entering the password with -a
--help	Help information
--version	AMDC CLI version information

Usage: amdc-cli --cluster [command [command parameters...]]

Primary Parameter	Secondary Parameter	Description
create host1:port1... hostN:portN		Create cluster
	--cluster-replicas	The number of slave nodes
check host:port		Check cluster
	--cluster-search-multiple-owners	Check if there are slots assigned to multiple nodes simultaneously
info host:port		View the cluster status
fix host:port		Repair cluster
	--cluster-search-multiple-owners	Repair the problem of duplicate slot allocation
reshard host:port		Specify any node of the cluster to migrate slot and redistribute slots
	--cluster-from	Slots need to be migrated from which source nodes. Migration can be completed from multiple source nodes, separated by commas. The passed is the node id of the node. You can also directly pass --from all. In this case, the source nodes are all nodes of the cluster. If this parameter is not passed, the user will be prompted to enter during the migration process.
	--cluster-to	The node id of the destination node to which the slot needs to be migrated. Only one destination node can be specified. If this parameter is not passed, the user will be prompted to enter during the migration process.
	--cluster-slots	The number of slots to be migrated. If this parameter is not passed, the user will be prompted to enter during the migration process.
	--cluster-yes	Specify the confirmation input during migration
	--cluster-timeout	Set the timeout time of the migrate command
	--cluster-pipeline	Define the number of keys retrieved by the cluster getkeysinslot command at a time. The default value is 10 if not passed.
	--cluster-replace	Whether to directly replace to the destination node
rebalance host:port		Specify any node of the cluster to balance the number of slots of cluster nodes
	--cluster-weight <node1=w1...nodeN=wN>	Specify the weight of cluster nodes
	--cluster-use-empty-masters	Set that master nodes without assigned slots can participate. It is not allowed by default.
	--cluster-timeout	Set the timeout time of the migrate command
	--cluster-simulate	Simulate the rebalance operation and will not actually perform the migration operation
	--cluster-pipeline	Define the number of keys retrieved by the cluster getkeysinslot command at a time. The default value is 10 if not passed.
	--cluster-threshold	If the slot threshold of migration exceeds threshold, perform the rebalance operation
	--cluster-replace	Whether to directly replace to the destination node
add-node new_host:new_port existing_host:existing_port		Add a node and add the new node to the specified cluster. The default is to add the master node.
	--cluster-slave	The new node is a slave node. By default, a random master node is selected.
	--cluster-master-id	Specify the master node for the new node
del-node host:port node_id		Delete a given node and shut down the node service after success.
call host:port command arg arg.. arg		Execute related commands on all nodes of the cluster
set-timeout host:port milliseconds		Set cluster-node-timeout
import host:port		Import external amdc data into the cluster (nodes in non-cluster mode)
	--cluster-from	Import the data of the specified instance into the cluster
	--cluster-copy	Specify copy when migrating
	--cluster-replace	Specify replace when migrating

AMDC Client Usage

The AMDC client languages cover mainstream programming languages such as Java, PHP, Python, C, C++, Node.js, etc. It can be connected through amdc-cli, and also supports the connection of the redis client to amdc for operations. The effects of both are the same.

General Operations

Connect to amdc-server via url

./amdc-cli -u amdc://user:password@127.0.0.1:6359/db

Display the command help information of amdc-cli --help

Omitted.

Connect to amdc using the amdc-cli client

./amdc-cli -h host -p port -a password

-h is used to specify the ip
-p is used to specify the port
-a is used to specify the authentication password

Output the returned data to the current command line --raw (hide the data type) and --no-raw

Run the same command continuously for n times -r set the running interval time -i (seconds)

Connect to the specified db: -n

Execute the command output in comma-separated format (csv format) --csv

Read data from standard input (stdin) as the last parameter of amdc-cli -x

RDB backup --rdb

Slave mode, treat the client as a slave node of the server --slave

Pipeline mode, execute multiple commands at once --pipe

Statistical Operations

Continuous statistical mode, view the number of amdc keys, occupied memory and other data in real time --stat

Find large keys --bigkeys

Find hot keys --hotkeys

Query Operations

Get the list of all keys --scan

Scan the specified key --pattern (regular expression)

View the minimum, maximum, and average access latency of amdc within a certain period of time --latency or --latency-history

Test Operations

Measure the latency time of amdc within n seconds --intrinsic-latency

LRU simulation, simulate the cache elimination strategy --lru-test

Simulate and display the copy of the commands received from the master server --replica

LUA Operations

Execute lua script

Start lua debugging mode: --ldb or --ldb-sync-mode

Cluster Operations

Interactive command line cluster mode -c

./amdc-cli -h host -p port -a password -c

Create cluster

./amdc-cli --cluster create <ip:port ip:port …> --cluster-replicas <num> (The number of slave nodes for each master node)

Add a master node to the cluster

./amdc-cli --cluster add-node ip:port Any node of the original cluster (Note: The new master node is not assigned a slot and needs to move or balance the slot)

Move slot

./amdc-cli --cluster reshard Any node of the original cluster --cluster-from The id of the original slot node --cluster-to The id of the receiving slot node --cluster-slots The number of moving slots

Balance slot

Balance the cluster of non-new nodes:

./amdc-cli --cluster rebalance --cluster-weight node_id=ratio node_id=ratio .... Any node of the original cluster

Balance the cluster with new nodes (empty nodes):

./amdc-cli --cluster rebalance [Any ip:port] –cluster-use-empty-masters –cluster-threshold 1

Check the cluster

./amdc-cli --cluster check ip:port --cluster-search-multiple-owners

Repair the cluster

./amdc-cli --cluster fix ip:port --cluster-search-multiple-owners

Remove the cluster node

./amdc-cli --cluster del-node Any node of the cluster The id of the removed node (If the master node is removed, all slots need to be moved to other master nodes)

Add a slave node to the cluster

./amdc-cli --cluster add-node [The ip:port to be added] [The ip:port of the master] --cluster-slave --cluster-master-id [master_id (The first parameter of each item in cluster nodes)]

Import the data of the external node into the cluster

./amdc-cli --cluster import Any node of the cluster --cluster-from The external node --cluster-replace

View the cluster information

./amdc-cli --cluster info Any node of the cluster

Execute the command in the cluster

./amdc-cli --cluster call Any node of the cluster cmd

View the help information of the cluster command

./amdc-cli --cluster help

RDB Cluster Data Migration Tool

AMDC provides an RDB cluster data migration tool for replacing Redis clusters, allowing users to quickly migrate Redis cluster data to the AMDC cluster.

Data Migration Tool Usage

The binary file is named amdc-data-mirgration: Run ./amdc-data-mirgration -help to see what parameters are available.

-f：Followed by the file directory, the program will automatically obtain the rdb files in this directory
-h：Destination IP address
-p：Destination port
-a：Password
-c：Whether it is in cluster mode

Among them, -f, -h, and -p are mandatory parameters. After -f, followed by the file directory, the program will automatically obtain the rdb files in this directory. If the cluster has a password, the parameter -a password needs to be added; if amdc is in cluster mode, the parameter -c needs to be added.

If amdc is in cluster mode:

For example:./amdc-data-mirgration -c -h 127.0.0.1 -p 6359 -f /root/server

If amdc is in stand-alone or master-slave mode:

For example:./amdc-data-mirgration -h 127.0.0.1 -p 6359 -f /root/server

Performance Testing Tool

amdc-benchmark is a dedicated performance testing tool provided to users to test AMDC, so as to quickly understand the performance of AMDC and provide a reliable basis for optimization.

AMDC benchmark parameters

All parameters of benchmark are to output the corresponding performance test data. Example: amdc-benchmark [-h <host>] [-p <port>] [-c <clients>] [-n <requests>]

Command parameters

-h <hostname> Instance host name (default 127.0.0.1)
-p <port>      Instance port (default 6379)
-s <socket>    Instance socket (will overwrite host and port parameters)
-c <clients>   Concurrent connection count (default 50)
-n <requests>  Total number of requests (default 100000)
-d <size>      The size of SET/GET data, in bytes (default 2)
-dbnum <db>    Specify the db library (default AMDC has libraries 0-15, the current parameter defaults to 0)
-k <boolean>   1 = Keep the connection 0 = Reconnect (default 1)
-r <keyspacelen> Use random key for SET/GET/INCR, use random value for SADD, keyspacelen refers to the maximum length of random data (The maximum length range is 0-12)
-P <numreq>    Merge the specified number of requests using pipelines, default 1 (Do not use pipelines)
-q             Only display the query/sec value
--csv          Output in CSV format
-l             Loop and keep testing
-t <tests>     Specify the test command
-I             Idle mode, open N idle connections and wait