Suggestions on TaurusDB Instance Selection

Overview

Before purchasing a TaurusDB instance, consider factors such as the price, performance, workload capacity, and service scenario to choose the right one for your needs. This section describes the differences between types, billing modes, and storage types of TaurusDB instances, helping you select the most suitable instance.

Storage Engine

By default, TaurusDB uses the InnoDB storage engine to provide high-performance and reliable transaction processing. The TaurusDB kernel provides functions such as parallel query, table recycle bin, hot row update, and multi-tenancy. It is widely used in high-performance and high-concurrency scenarios.

Storage Types

TaurusDB provides two storage types: DL6 and DL5. The table below explains their key differences to help you choose the best option for your needs.

**Table 1** Storage type description
Storage Type	Characteristic	Applicable Scenario
DL6	The shared storage is the default storage type for TaurusDB instances created before July 2024. DL6-based instances achieve zero RPO with a 3-AZ deployment and deliver better performance and higher peak throughput.	Core application systems that are sensitive to performance and have demanding requirements on storage I/O during peak hours, such as those in finance, e-commerce, government, and gaming
DL5	The new type of storage uses Huawei Cloud's hardware and network infrastructure technologies, ensuring that DL5-based instances maintain the same high availability (zero RPO in the 3-AZ deployment) as DL6-based instances. Although the peak performance may decrease, the cost per unit of capacity is significantly reduced.	CPU-intensive sub-core business systems, or application modules that need to minimize costs

Billing Modes

There are yearly/monthly, pay-per-use, and serverless billing modes to meet requirements in different scenarios.

**Table 2** Comparison of billing modes
Billing Mode	Yearly/Monthly	Pay-per-Use	Serverless
Payment	Prepaid Billed by the required duration specified in your order	Postpaid Billed for what you use	Postpaid Billed for what you use
Supported Instance Type	Single-node Cluster	Single-node Cluster Multi-primary	Serverless
Billing Cycle	Billed by the required duration specified in your order	Billed by second and settled by hour	Billed by second and settled by hour
Billing Item	Instance specifications (vCPUs and memory), storage space, backup space, EIP, and monitoring by seconds	Instance specifications (vCPUs and memory), storage space, backup space, EIP, and monitoring by seconds	Compute, storage space, backup space, EIP, and monitoring by seconds
Billing Mode Change	Can be changed to pay-per-use. The change is only applied after the yearly/monthly subscription expires. For details, see Yearly/Monthly to Pay-per-Use.	Can be changed to yearly/monthly. For details, see Pay-per-Use to Yearly/Monthly.	N/A
Applicable Scenario	Recommended for resources expected to be in use long-term. A cost-effective option for scenarios where the resource usage duration is predictable.	Good for short-term, bursty, or unpredictable workloads that cannot tolerate any interruptions, such as applications for e-commerce flash sales, temporary testing, and scientific computing.	The instance capacities automatically change based on application requirements.

The table below lists the billing items.

**Table 3** Billing items
Billing Item	Description	Billing Factor
(Mandatory) DB instance	You are billed for the selected instance specifications, including vCPUs, memory, and nodes.	vCPUs, memory, and number of nodes
(Mandatory) Storage	Pre-purchased storage is billed on a yearly/monthly basis. However, if the actual usage exceeds your purchased storage, you will be billed for additional storage on a pay-per-use basis.	Storage space, which is billed based on the unified standard
(Mandatory) Backup space	TaurusDB provides free backup space equal to the amount of your purchased storage space. After the free backup space is used up, charges are applied based on the backup space pricing details. Pricing is listed on a per-hour basis, but bills are calculated based on the actual usage duration.	Backup space, which is billed based on the unified standard
(Optional) EIP bandwidth	TaurusDB instances can be accessed through a public network, and traffic fees are generated accordingly. You are not billed for traffic generated through a private network.	Bandwidth, traffic, and EIP reservation (An EIP is required if a DB instance needs to access the Internet.) EIP for a yearly/monthly DB instance: billed by bandwidth
(Optional) Monitoring by seconds	TaurusDB provides monitoring every 60 seconds for free, but you are billed for monitoring by seconds. Its pricing is listed on a per-hour basis, but bills are calculated based on actual usage.	Monitoring frequency, which is billed based on the unified standard
(Optional) HTAP instance	If you create a standard HTAP instance for a TaurusDB instance, you will be billed for the HTAP instance.	Instance specifications and storage type
(Optional) Cross-region backup	If cross-region backup is enabled, you will be billed for backup space and network traffic required for dumping cross-region backups.	Backup space and network traffic required for dumping cross-region backups
(Optional) Proxy instance	Proxy instances are free.	N/A
(Optional) DRS migration	If you use Data Replication Service (DRS) for data migration, you will be billed based on the DRS pricing standard.	For details, see DRS Billing.

The table below lists the billing items.

**Table 4** Billing items
Billing Item	Description	Billing Factor
(Mandatory) DB instance	You are billed for the selected instance specifications, including vCPUs, memory, and nodes.	vCPUs, memory, and number of nodes
(Mandatory) Storage	Pre-purchased storage is billed on a yearly/monthly basis.	Storage space, which is billed based on the unified standard
(Mandatory) Backup space	After the free backup space is used up, charges are applied based on the backup space pricing details. Pricing is listed on a per-hour basis, but bills are calculated based on the actual usage duration.	Backup space, which is billed based on the unified standard
(Optional) EIP bandwidth	DB instances can be accessed through a public network, and traffic fees are generated accordingly. You are not billed for traffic generated through a private network.	Bandwidth, traffic, and EIP reservation (An EIP is required if a DB instance needs to access the Internet.) Billing by bandwidth: bandwidth fee + EIP reservation fee Billing by traffic: traffic fee + EIP reservation fee Shared bandwidth added: bandwidth fee + EIP reservation fee
(Optional) Monitoring by seconds	Enabling monitoring by seconds will incur additional fees. Its pricing is listed on a per-hour basis, but bills are calculated based on actual usage.	Monitoring frequency, which is billed based on the unified standard
(Optional) HTAP instance	If you create a standard HTAP instance for a TaurusDB instance, you will be billed for the HTAP instance.	Instance specifications and storage type
(Optional) Cross-region backup	If cross-region backup is enabled, you will be billed for backup space and network traffic required for dumping cross-region backups.	Backup space and network traffic required for dumping cross-region backups
(Optional) Proxy instance	Proxy instances are free.	N/A
(Optional) DRS migration	If you use Data Replication Service (DRS) for data migration, you will be billed based on the DRS pricing standard.	For details, see DRS Billing.

The table below lists the billing items.

**Table 5** Billing items
Billing Item	Description	Billing Factor
DB instance	You are billed for the selected instance specifications. The billing starts immediately after the instance is purchased.	vCPUs, memory, and number of nodes
Storage	You do not need to select storage when purchasing a DB instance. Storage will be scaled up dynamically based on how much data needs to be stored. It is billed hourly on a pay-per-use basis.	Storage space, which is billed based on the unified standard
Backup space	TaurusDB provides free backup space equal to the amount of your used storage space. If the backup space usage exceeds 100% of your provisioned database storage, the additional part will be billed based on the backup pricing.	Backup space, which is billed based on the unified standard
(Optional) Public network traffic	TaurusDB instances can be accessed through a public network, and traffic fees are generated accordingly. You are not billed for traffic generated through a private network.	Bandwidth, traffic, and EIP reservation (An EIP is required if a DB instance needs to access the Internet.)
(Optional) HTAP instance	If you create a standard HTAP instance for a TaurusDB instance, you will be billed for the HTAP instance.	Instance specifications and storage type

Instance Type

TaurusDB supports cluster, single-node, and multi-primary instances. You can select an instance type based on your service scenario and scale. For details, see Table 6.

**Table 6** Instance type description
Instance Type	Description	Highlight	Applicable Scenario
Cluster	DB engine version: 8.0 Node: 1 primary node + 1 to 15 read replicas	The primary node processes read and write requests, and the read replicas process only read requests. If the primary node becomes unavailable, TaurusDB automatically fails over to a read replica.	Medium- and large-sized enterprises in the Internet, taxation, banking, and insurance sectors
Single-node	DB engine version: 8.0 Node: 1 primary node + 0 read replicas	Single-node instances do not involve data synchronization between nodes and can easily ensure atomicity, consistency, isolation, and durability of transactions.	Development and testing of microsites, and small and medium enterprises, or learning about TaurusDB
Serverless	DB engine version: 8.0 Node: 1 primary node + 0 to 7 read replicas	The instance capacities automatically change based on application requirements.	Scenarios where capacity needs to be automatically expanded based on services
Multi-primary	DB engine version: 8.0 Node: 2 to 63 primary nodes + 0 read replicas	Such an instance can process multiple reads and writes, delivering excellent read/write performance at high concurrency.	Scenarios where high concurrent write performance is required

Instance Specifications

There are general-purpose and dedicated specifications for pay-per-use and yearly/monthly TaurusDB instances, as listed in Table 7. For a serverless instance, you do not need to select the instance specifications. You only need to specify a compute range. A serverless instance is billed by TCU. 1 TCU is approximately equal to 1 vCPU and 2 GB of memory.

**Table 7** Instance specifications
Instance Specification	Supported CPU Architecture	Description	Applicable Scenario
General-purpose	x86	The vCPUs and memory are shared with other general-purpose instances on the same physical machine. vCPU usage is maximized through resource overcommitment. General-purpose instances are cost-effective.	Scenarios where stable performance is not critical
Dedicated	x86 and Kunpeng	Your instance gets dedicated vCPUs and memory, so the performance is stable. It is not affected by other instances on the same physical machine.	Scenarios that require stable performance

Functions Supported by Different Instance Types

**Table 8** Reference for instance type selection
Reference		Single-node	Cluster	Serverless	Multi-primary (OBT)
Specifications change	Changing the vCPUs and Memory of a DB Instance or Node	√	√	x	x
	Changing the Compute Range	x	x	√	x
	Changing the Storage Space of a DB Instance	√ Supported only by yearly/monthly instances	√ Supported only by yearly/monthly instances	x	x
	Configuring Autoscaling for a DB Instance (OBT)	√ Supported only by yearly/monthly instances	√ Supported only by yearly/monthly instances	x	x
Version upgrade	Upgrading the Minor Version of a DB Instance	√	√	√	x
Backup and restoration	Data Backups	√	√	√	x
Backup and restoration	Data Restorations	√	√	√	x
Data migration	Migrating Data to TaurusDB Enterprise Edition (OBT)	x	√	x	x
Proxy instance	Changing Configurations of a Proxy Instance	√	√	√	√
Proxy instance	Upgrading the Kernel Version of a Proxy Instance	√	√	√	√
DBA Assistant	Performance Monitoring	√	√	√	x
	Problem Diagnosis	√	√	√	x
	SQL Analysis and Tunning	√	√	√	x
Security and encryption	Configuring SSL for a DB Instance	√	√	√	x
Security and encryption	Enabling TDE for a DB Instance	√	√	√	x
Cold and hot data separation (OBT)	Configuring a Cold Table	x	√	x	x
HTAP analysis	What Is HTAP Standard Edition?	x	√	√	x
Tag	Tag Management	√	√	√	√