Lindorm provides Lindorm V1, Lindorm, and Lindorm Tunnel Service (LTS). You must use an instance to access the features of Lindorm. You can create and release instances by using the console, OpenAPI, or an SDK. This topic describes how to create and release instances of various Lindorm product types in the console, and provides links to reference documentation for creating instances by using OpenAPI and SDKs.
Free trial
If you are a verified user and have not used Lindorm before, you are eligible for a free trial. For the trial rules and tutorials, see Free Trial and Lindorm Query Practice with Massive Data.
Prerequisites
You have registered an Alibaba Cloud account. For more information, see Register an Alibaba Cloud Account.
If a RAM user creates a Lindorm instance, ensure the user has the following permissions.
Lindorm V1 and Lindorm
The required permissions depend on the billing method.
Subscription: AliyunLindormFullAccess, AliyunHBaseReadOnlyAccess, AliyunBSSOrderAccess.
Pay-as-you-go: AliyunLindormFullAccess, AliyunHBaseReadOnlyAccess.
Lindorm Tunnel Service (LTS)
The required permissions depend on the billing method.
Subscription: AliyunLindormFullAccess, AliyunVPCReadOnlyAccess, AliyunBSSOrderAccess.
Pay-as-you-go: AliyunLindormFullAccess, AliyunVPCReadOnlyAccess.
To learn how to view the permissions of a RAM user, see Manage RAM User Information.
If the RAM user lacks the required permissions, ask the Alibaba Cloud account owner to grant them. To learn how a primary account grants permissions, see Manage RAM User Permissions.
Select a product type
Lindorm offers Lindorm V1, Lindorm, Lindorm Tunnel Service (LTS) for different use cases.
Lindorm V1: A standardized deployment on a dedicated cluster. It provides wide-column, time series, search, file, compute, and streaming engines. You can use it in production environments for medium- to large-sized enterprises, for scenarios such as Internet of Vehicles (IoV), IoT, industrial IoT, risk control, billing, advertising, content feeds, and data dashboards. Lindorm V1 supports single-AZ and multi-AZ deployment to meet different disaster recovery requirements.
Lindorm: Supports multiple engines, including the wide-column engine, time-series engine, and Lindorm Tunnel Service (LTS). It is suitable for production use in medium- to large-sized enterprises, for learning purposes for individual developers, and for feature validation and development testing in small businesses.
Lindorm Tunnel Service (LTS): Provides features such as data migration and synchronization, real-time subscriptions, and backup and recovery.
For a detailed introduction to the product series, see Product Family.
Create an instance
Commodity Type: Lindorm
The configuration process and key parameters are highly similar for single-AZ, multi-AZ basic, and multi-AZ high-availability deployment modes. For clarity, this topic uses a single-AZ instance as an example and explains the settings specific to multi-AZ deployments in subsequent sections.
Go to the Lindorm purchase page.
In the Basic information section, configure the following parameters:
Parameter
Description
Commodity Type
Lindorm
Billing Method
Subscription: A prepaid billing method where you pay for the instance upon creation. This option is cost-effective for long-term use, and longer subscription periods typically offer larger discounts.
Pay-as-you-go: A postpaid billing method where you are billed hourly based on your instance configuration. This option is suitable for short-term needs, as you can release the instance at any time to save costs.
Deployment Scenario
Single Zone: The primary and secondary nodes are in the same availability zone. This mode is suitable for scenarios that require single-data-center disaster recovery.
Multi-zone Deployment (Basic): The primary and secondary nodes are in different availability zones, providing cross-AZ disaster recovery. All engines are supported.
Multi-zone Deployment (HA): The primary and secondary nodes are in different availability zones, providing cross-AZ disaster recovery.
ImportantThe Multi-zone Deployment (HA) mode currently supports only the wide-column engine.
Instance name
Set a name for the instance for easy identification.
Region Type
Select the region where the instance will be created.
Recommendations:
To minimize latency, choose a region close to your services.
If you use multiple Alibaba Cloud products that require internal network communication, deploy them in the same region.
For more information about how to select a region, see Regions and zones.
ImportantThe region cannot be changed after the instance is created.
Availability Zone
Select the availability zone for the instance.
NoteDifferent availability zones within the same region can communicate over the internal network.
VPC
Select the VPC for the instance.
If you use multiple Alibaba Cloud products, such as ECS, and they require internal network communication, deploy Lindorm in the same VPC. Otherwise, they cannot communicate directly over the internal network.
ImportantIf your VPC is not displayed on the purchase page, check the following:
Ensure that a VPC has been created in the selected region.
If you are logged in as a RAM user, ensure that the user has the AliyunLindormFullAccess and AliyunHBaseReadOnlyAccess permissions. After you grant these permissions, you can select a VPC.
vSwitch
Select the vSwitch for the instance.
ImportantIf your vSwitch is not displayed on the purchase page, check the following:
Ensure that a vSwitch has been created in the selected availability zone.
If you are logged in as a RAM user, ensure that the user has the AliyunLindormFullAccess and AliyunHBaseReadOnlyAccess permissions. After you grant these permissions, you can select a vSwitch.
In the Select Data Service section, configure the following parameters:
Parameter
Description
Storage Type
Options include Standard Cloud Storage, Performance Cloud Storage, Capacity Cloud Storage, local SSD, and local HDD. Select a storage type based on your business requirements. For more information, see Storage types.
Performance: Suitable for scenarios that require low-latency data access, such as ad bidding, user profiling, audience segmentation, real-time search, and risk control systems.
Standard: Suitable for real-time data access scenarios, such as content feeds, chats, real-time reporting, and online computing.
Big Data: Uses local disks from a single physical machine, offering low latency, high random IOPS, high throughput, and cost-effectiveness. This type is suitable for industries such as the internet and finance that require extensive computing and analysis for massive datasets and offline processing.
Local SSD: Uses local disks from a single physical machine, offering low latency, high random IOPS, high throughput, and cost-effectiveness. This type is suitable for online businesses such as online gaming, e-commerce, live streaming, and media that require low latency and high I/O performance for block storage.
NoteThe storage type cannot be changed after the instance is created.
Instance Configuration
General-purpose Mode: Select one or more data engines based on your business scenario. For an introduction to each engine and its use cases, see Engine types.
Select the required engines. Click Edit to configure engine settings, including settings for node specifications, node count, and attached cloud disks (such as type and capacity per node).
Scenario-based Mode: This mode simplifies the configuration process. You only need to provide estimates based on your business metrics, without worrying about underlying details.
In the Long-term Memory scenario, you only need to provide the Number of Users and Memories per User. The system automatically generates the corresponding parameters in the Instance Configuration section.
Whether to select capacity cloud storage
Capacity storage stores infrequently accessed historical data (cold data) to reduce storage costs. For more information, see Capacity-type cloud storage.
You can purchase capacity storage when you create the instance or enable it after the instance is created.
Whether to activate the computing engine
For more information, see Compute engine.
If you selected Billing Method for the Subscription, select a Subscription Length and configure Auto-renewal Recommended.
To prevent the instance from being released upon expiration, we recommend that you enable Auto-renewal Recommended. For more information about instance expiration and renewal, see Expiration and overdue payments and Enable auto-renewal.
ImportantIf a RAM user purchases a Lindorm instance, the RAM user must have the AliyunBSSOrderAccess permission to pay for subscription orders. This permission is not required for pay-as-you-go instances.
Set the Resource Group for the instance.
Resource groups allow you to manage, authorize, and bill resources in groups. For more information, see What is a resource group?.
Follow the instructions on the purchase page to complete the payment.
Unlike single-AZ instances, multi-AZ deployment modes (including multi-AZ basic and multi-AZ high-availability) require you to specify deployment locations for different cluster roles to build a cross-data-center disaster recovery architecture. You must also configure the following parameters:
Parameter | Role and function | Recommendation |
Primary Zone | Handles the primary data read and write traffic and acts as the cluster's main workspace. Requires an associated vSwitch. | For optimal performance, choose the availability zone where your application (such as ECS) is located or the one with the lowest network latency. |
vSwitch for Primary Zone | Select the vSwitch for the instance's primary AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Secondary Zone | Deploys redundant data replicas and synchronizes with the primary AZ in real time. It takes over service if the primary AZ fails, ensuring business continuity. | Select an availability zone that is physically isolated from the primary AZ. |
vSwitch for Secondary Zone | Select the vSwitch for the instance's secondary AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Client Node Zone | Deploys the coordinator node and log node. It does not handle data reads or writes but arbitrates between the primary and secondary AZs to prevent split-brain issues, which is critical for cluster decision consistency. | Select a third, independent availability zone to form a complete three-data-center disaster recovery architecture. |
vSwitch for Client Node Zone | Select the vSwitch for the instance's coordinator AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Commodity Type: Lindorm V1
The configuration process and key parameters are highly similar for single-AZ, multi-AZ basic, and multi-AZ high-availability deployment modes. For clarity, this topic uses a single-AZ instance as an example and explains the settings specific to multi-AZ deployments in subsequent sections.
Go to the Lindorm purchase page.
In the Basic information section, configure the following parameters:
Parameter
Description
Commodity Type
Lindorm V1
Billing Method
Subscription: A prepaid billing method where you pay for the instance upon creation. This option is cost-effective for long-term use, and longer subscription periods typically offer larger discounts.
Pay-as-you-go: A postpaid billing method where you are billed hourly based on your instance configuration. This option is suitable for short-term needs, as you can release the instance at any time to save costs.
Deployment Mode
Single Zone: The primary and secondary nodes are in the same availability zone. This mode is suitable for scenarios that require single-data-center disaster recovery.
Multi-zone Deployment (Basic): The primary and secondary nodes are in different availability zones, providing cross-AZ disaster recovery. All engines are supported.
Multi-zone Deployment (HA): The primary and secondary nodes are in different availability zones, providing cross-AZ disaster recovery.
ImportantThe Multi-zone Deployment (HA) mode currently supports only the wide-column engine.
Region
Select the region where the instance will be created.
Recommendations:
To minimize latency, choose a region close to your services.
If you use multiple Alibaba Cloud products that require internal network communication, deploy them in the same region.
For more information about how to select a region, see Regions and zones.
ImportantThe region cannot be changed after the instance is created.
Availability Zone
Select the availability zone for the instance.
NoteDifferent availability zones within the same region can communicate over the internal network.
Virtual Private Cloud (VPC)
Select the VPC for the instance.
If you use multiple Alibaba Cloud products, such as ECS, and they require internal network communication, deploy Lindorm in the same VPC. Otherwise, they cannot communicate directly over the internal network.
ImportantIf your VPC is not displayed on the purchase page, check the following:
Ensure that a VPC has been created in the selected region.
If you are logged in as a RAM user, ensure that the user has the AliyunLindormFullAccess and AliyunHBaseReadOnlyAccess permissions. After you grant these permissions, you can select a VPC.
vSwitch
Select the vSwitch for the instance.
ImportantIf your vSwitch is not displayed on the purchase page, check the following:
Ensure that a vSwitch has been created in the selected availability zone.
If you are logged in as a RAM user, ensure that the user has the AliyunLindormFullAccess and AliyunHBaseReadOnlyAccess permissions. After you grant these permissions, you can select a vSwitch.
Cluster Name
Set a name for the cluster for easy identification.
In the Select Data Service section, configure the following parameters:
Parameter
Description
Storage Type
Options include Standard Cloud Storage, Performance Cloud Storage, Capacity Cloud Storage, local SSD, and local HDD. Select a storage type based on your business requirements. For more information, see Storage types.
Performance: Suitable for scenarios that require low-latency data access, such as ad bidding, user profiling, audience segmentation, real-time search, and risk control systems.
Standard: Suitable for real-time data access scenarios, such as content feeds, chats, real-time reporting, and online computing.
Big Data: Uses local disks from a single physical machine, offering low latency, high random IOPS, high throughput, and cost-effectiveness. This type is suitable for industries such as the internet and finance that require extensive computing and analysis for massive datasets and offline processing.
Local SSD: Uses local disks from a single physical machine, offering low latency, high random IOPS, high throughput, and cost-effectiveness. This type is suitable for online businesses such as online gaming, e-commerce, live streaming, and media that require low latency and high I/O performance for block storage.
NoteThe storage type cannot be changed after the instance is created.
Select data engine
Select one or more data engines based on your business scenario. For an introduction to each engine and its use cases, see Engine types.
Lindorm has a distributed cluster architecture. Each data engine has its own dedicated nodes and must run on at least two nodes to enable distributed task processing.
For data engines you do not need, keep the number of nodes at 0.
ImportantAfter the instance is created, you can change the node specifications and node count for data engines, and enable engines you have not purchased. However, you cannot disable an engine that is already enabled.
Select storage capacity
Storage capacity determines how much data your instance can store. The configuration depends on the storage type.
For Standard Cloud Storage, Performance Cloud Storage, or Capacity Cloud Storage:
Select the Storage Space.
The purchased storage capacity is the actual usable storage capacity.
For example, if you purchase 200 GB of storage and your raw data size is 100 GB, the data occupies 50 GB of storage after being compressed by Lindorm, leaving 150 GB of available space.
If you later need more storage or have excess capacity, you can scale the storage up or down without manual data migration.
Configure Whether to select capacity cloud storage.
Capacity storage stores infrequently accessed historical data (cold data) to reduce storage costs. For more information, see Capacity-type cloud storage.
You can purchase capacity storage when you create the instance or enable it after the instance is created.
NoteWhen the storage type is Capacity Cloud Storage, the storage capacity you select is for capacity storage, so you do not need to configure this setting.
When the deployment mode is multi-AZ high-availability, configure the following parameters:
Select the Single-Node Capacity for Core Nodes.
The actual usable storage space is half of the selected capacity.
For example, if you select 400 GB for the single-node capacity and purchase four wide-column nodes, the actual storage space is 800 GB.
If you later need more storage or have excess capacity, you can scale the disks to increase or decrease storage space.
Select the Disk Type and Single-Node Capacity for Log Nodes.
Choose the disk type for logs based on your business scenario.
ESSD: Use for scenarios with frequently accessed data or high I/O loads.
Ultra Disk: Use for I/O-intensive scenarios.
The single-node capacity for log nodes is the actual usable storage capacity for logs.
Purchasing capacity storage is not currently supported.
NoteIf you later need more storage or have excess capacity, you can scale disks to increase or decrease the Single-Node Capacity for Core Nodes and Single-Node Capacity for Log Nodes.
For Local SSD or Big Data storage:
View the Single-Node Capacity of Local Disk.
Local disk capacity per node is determined by the specifications of local disk nodes and the number of nodes in the data engine, and cannot be customized. If the Local disk capacity per node is too small to meet your data storage requirements, you can mount cloud disks and use them in a pool with local SSDs.
Lindorm instances with local disks use a three-replica architecture. Therefore, the actual usable storage capacity is one-third of the total raw disk capacity (single-node capacity × number of nodes).
For example, if you select Local SSD, the single-node disk capacity is 894 GB, and the number of data engine nodes is 3, the console shows 2,682 GiB of storage space, but the actual usable capacity is 894 GB (894 * 3 * 1/3).
To improve storage utilization, you can enable erasure coding to reduce replica redundancy. For more information, see Erasure coding (EC).
Configure Purchase Capacity Cloud Storage.
Capacity storage stores infrequently accessed historical data (cold data) to reduce storage costs. For more information, see Capacity-type cloud storage.
You can purchase capacity storage when you create the instance or enable it after the instance is created.
Attach Cloud Disks.
When the local disk capacity is insufficient and you do not want to purchase more nodes, you can attach cloud disks, which are pooled with the local SSDs.
Like local disks, attached cloud disks use a three-replica architecture. Therefore, the usable storage capacity from attached disks is one-third of the total provisioned cloud disk capacity.
NoteLocal HDD does not support purchasing capacity storage or attaching cloud disks.
When the deployment mode is multi-AZ high-availability, configure the following parameters:
Select the disk type and single-node capacity for log nodes.
Choose the disk type for logs based on your business scenario.
ESSD: Use for scenarios with frequently accessed data or high I/O loads.
Ultra Disk: Use for I/O-intensive scenarios.
The single-node capacity for log nodes is the actual usable storage capacity for logs.
View the Single-Node Capacity of Local Disk.
Local disk capacity per node is determined by the specifications of local disk nodes and the number of nodes for the data engine, and cannot be customized. If the Local disk capacity per node is too small to meet your data storage needs, you can mount a cloud disk and use it in a pool with local SSDs.
Lindorm instances with local disks use a three-replica architecture. Therefore, the actual usable storage capacity is one-third of the total raw disk capacity (single-node capacity × number of nodes).
For example, if you select Local SSD, the single-node disk capacity is 894 GB, and the number of data engine nodes is 3, the console shows 2,682 GiB of storage space, but the actual usable capacity is 894 GB (894 * 3 * 1/3).
To improve storage utilization, you can enable erasure coding to reduce replica redundancy. For more information, see Erasure coding (EC).
Purchasing capacity storage is not currently supported.
Purchase capacity cloud storage
Capacity storage stores infrequently accessed historical data (cold data) to reduce storage costs. For more information, see Capacity-type cloud storage.
You can purchase capacity storage when you create the instance or enable it after the instance is created.
Whether to activate the computing engine
For more information, see Compute engine.
If you selected Billing Method for the Subscription, select a Subscription Length and configure Auto-renewal Recommended.
To prevent the instance from being released upon expiration, we recommend that you enable Auto-renewal Recommended. For more information about instance expiration and renewal, see Expiration and overdue payments and Enable auto-renewal.
ImportantIf a RAM user purchases a Lindorm instance, the RAM user must have the AliyunBSSOrderAccess permission to pay for subscription orders. This permission is not required for pay-as-you-go instances.
Set the Resource Group for the instance.
Resource groups allow you to manage, authorize, and bill resources in groups. For more information, see What is a resource group?.
Follow the instructions on the purchase page to complete the payment.
Unlike single-AZ instances, multi-AZ deployment modes (including multi-AZ basic and multi-AZ high-availability) require you to specify deployment locations for different cluster roles to build a cross-data-center disaster recovery architecture. You must also configure the following parameters:
Parameter | Role and function | Recommendation |
Multi-AZ Combination | A preset, optimized set of availability zones that bundles the primary, secondary, and coordinator AZs. | Alibaba Cloud recommends combinations with low and stable network latency between availability zones. Selecting a combination helps prevent manual selection errors. |
Primary Zone | Handles the primary data read and write traffic and acts as the cluster's main workspace. Requires an associated vSwitch. | For optimal performance, choose the availability zone where your application (such as ECS) is located or the one with the lowest network latency. |
vSwitch for Primary Zone | Select the vSwitch for the instance's primary AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Secondary Zone | Deploys redundant data replicas and synchronizes with the primary AZ in real time. It takes over service if the primary AZ fails, ensuring business continuity. | Select an availability zone that is physically isolated from the primary AZ. |
vSwitch for Secondary Zone | Select the vSwitch for the instance's secondary AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Client Node Zone | Deploys the coordinator node and log node. It does not handle data reads or writes but arbitrates between the primary and secondary AZs to prevent split-brain issues, which is critical for cluster decision consistency. | Select a third, independent availability zone to form a complete three-data-center disaster recovery architecture. |
vSwitch for Client Node Zone | Select the vSwitch for the instance's coordinator AZ. | Choose the same vSwitch as your application servers for efficient internal communication with the Lindorm instance. In complex network topologies, a dedicated vSwitch for databases can simplify network policy and security management. |
Commodity Type: LindormTunnel channel service
Parameter | Description |
Billing Method |
|
Region and Zone | Select the region where the instance will be created. Note
|
Zone | Select the availability zone for the instance. Note If your application is deployed on an ECS instance, we recommend that you select the same availability zone to reduce network latency. |
Network Type | This is fixed to VPC. |
Virtual Private Cloud (VPC) | Select the VPC for the instance. If you do not have an available VPC, you can create one in the VPC console. For instructions, see Create and manage a VPC. |
vSwitch | Select the vSwitch for the instance. If you do not have an available vSwitch, you can create one in the VPC console. For instructions, see Create a vSwitch. |
Cluster Name | Set a name for the instance for easy identification. |
Node Specifications | Select a node specification based on your requirements. |
Nodes | Enter the required number of nodes. |
Duration | If you are creating a subscription instance, select the subscription duration. You can choose from 1 to 9 months, or 1 to 3 years. Note If you want the instance to renew automatically upon expiration, select Auto-renewal Recommended and ensure your account has a sufficient balance before the expiration date. |
Resource Group | Select a resource group to manage the instance. You can choose an existing resource group or create a new one. For instructions, see Create a resource group. |
Release an instance
When you no longer need an instance and its Status is Running, you can use the Lindorm console to manually release or unsubscribe from the idle instance to prevent unnecessary charges.
When an instance is released, all its data is permanently deleted and cannot be recovered. Proceed with caution.
Log on to the Lindorm console.
In the upper-left corner of the page, select the region where the instance is deployed.
On the Instances page, find the target instance and click the
icon in the Actions column. Then, choose the appropriate action based on the billing method: for a pay-as-you-go instance, select Release; for a subscription instance, select Unsubscribe.In the confirmation dialog box that appears, click Determine.
For the rules and refund information related to unsubscribing from subscription instances, see Unsubscription Rules and Refunds.
FAQ
Q: The purchase page indicates insufficient inventory. What should I do?
A: This means the selected availability zone does not have enough capacity. Select a different availability zone to create the instance.
Q: The purchase page shows an authentication failure. What should I do?
A: This is usually caused by insufficient permissions for the current account. See Prerequisites and verify that your account and permissions meet the operational requirements.
Q: Why can't I select a VPC or vSwitch?
A: There are two common reasons:
No VPC or vSwitch exists in the selected region and availability zone. Ensure that you have created a VPC and a vSwitch in the selected region and availability zone.
The RAM user has insufficient permissions. Ensure that the RAM user has the AliyunLindormFullAccess and AliyunHBaseReadOnlyAccess permissions. After granting these permissions, you can select a VPC and vSwitch.
Q: Why can't I see the instance I just purchased in the console?
A: Provisioning a Lindorm instance takes about 20 minutes. If the instance is still not visible after 20 minutes, refresh the console. If it still does not appear, check for a refund in My Orders. A refund indicates that the instance creation failed. Try creating the instance again. If it fails again, please submit a ticket.
Q: Are there other charges besides the instance creation fee?
A: You may incur additional charges when using Lindorm in the following situations:
Enabling paid features, such as the backup feature for the wide-column engine or the compute engine. When you enable a paid feature, the console informs you of the cost, allowing you to decide whether to proceed. For a list of Lindorm's billable items, see Billing overview.
Upgrades and scaling operations may increase costs. Examples include adding nodes, changing node specifications, and expanding storage. The console displays the cost when you perform these operations. For billing rules on configuration changes, see Billing for configuration changes.
Q: Can I stop an instance?
A: Lindorm instances of all types cannot be stopped. If you no longer need an instance, you can release it. Billing stops after the instance is released.
Related documents
To create a Lindorm instance by using OpenAPI, see CreateLindormInstance.
To release a Lindorm instance by using OpenAPI, see ReleaseLindormInstance. This API operation currently supports only releasing pay-as-you-go instances.
To create a Lindorm instance by using an SDK, see the SDK documentation.